Jun 1, 2011 ... Renewable Energy Engineering and Management. Department of ..... Fiscal
Measures. FMEA. Failure Mode and Effects Analysis. FMECA.
MODELING THE RELIABILITY OF WIND TURBINE GENERATORS (WTGS)
Major Project Report
Submitted by
OKORONKWO JONATHAN ONWUKWE
For the partial fulfillment of the requirements of the degree of
MASTER OF TECHNOLOGY in Renewable Energy Engineering and Management
Department of Energy and Environment
TERI University
June, 2011
ii DECLARATION This is to certify that the work that forms the basis of this report, entitled “Modeling the Reliability of Wind Turbine Generators (WTGs)” is an original work carried out by me and has not been submitted anywhere else.
I certify that all sources of information and data are fully acknowledged in the project report.
Signature (OKORONKWO JONATHAN ONWUKWE) Date: 1st June, 2011
iii CERTIFICATE
iv ACKNOWLEDGEMENT All the Glory is to GOD Almighty who has done this marvelous work in our sight. My deep sincere appreciation goes to Sindacatum Climate Change Foundation, the Director – Mr. Mike Wheelhouse, the Secretary-Renita du Toit and the entire SCCF team without whom we won’t have had this opportunity to India. To my academic Supervisors Prof. VVN Kishore (HOD) and Prof Prateek Sharma Dean, the Vice Chancellor-Prof Bhavik, the Registrar-Prof Seth, our Faculties – Prof Sawhney, Dr. Najmur, Dr. Ashu, Dr. Ramikishore, Dr. Arun, Dr. Nandini, Dr. Rao, Dr. Ritu, Prof. Surender, Dr. Mande, Dr. Jami, Dr. Sharma, Dr. Ishan, Dr. Kaushik, Dr. Suresh, Dr. Bharati, Dr. Kamal, Dr. Poornima, Mr. Sushil, Mr. Rohan, Mr. Arora, Mr. Sharma, Mr. Abraham, Mr. Giri, Mr. Ratan, Mr. Prem, Mr. Sunil, Mr. Danesh including the library, the Security, the Cleaners , the Caterers and a host of others too numerous to mention, we are very grateful to TERI University.
To Federal Scholarship Board, Abuja, Nigeria and Nigerian High Commission, New Delhi, India, we are thankful for your fairness and transparency.
We are also very grateful to the entire staff of the Suzlon India Business Group and Global Learning Development Excellence team for such a wonderful opportunity given to carry out this interesting research study under the scrutiny of my amiable technical Supervisor Shri Prakash Vora and the entire OMS team under the Leadership of Shri Nilesh Vaishnav and Shri Anil Gagvani respectively.
On a personal note, we are very grateful to Mrs. Hemangini, Mr. Sharma, Mr. Nitin, and Mr. Amul amongst all our numerous friends who gave us support during our stay for this study.
To my Sweet dear wife, life is just beginning my hearty cheers to the corrections and contributions you made to the manuscript and to the entire families of Living Faith Church, Int’l and the Okoronkwo’s thank you for your fervent prayers.
v TABLE OF CONTENTS DECLARATION ................................................................................................... ii CERTIFICATE .................................................................................................... iii ACKNOWLEDGEMENT ................................................................................... iv TABLE OF CONTENTS ...................................................................................... v LIST OF ABBREVIATION .............................................................................. viii LIST OF FIGURES ............................................................................................. xi LIST OF TABLE ................................................................................................ xii ABSTRACT........................................................................................................ xiii INTRODUCTION ............................................................................................. - 1 1.0
Wind Power Generation .................................................................... - 1 -
1.1
Research Objective ........................................................................... - 3 -
2.0 LITERATURE REVIEW ........................................................................... - 5 2.1
Reliability Methods ........................................................................... - 5 -
2.1.1 Reliability Estimation Model ............................................................ - 6 2.1.1.1
Similar Item Prediction Method .............................................. - 6 -
2.1.1.2
Field Data Measurement Method ............................................ - 6 -
2.1.2 Reliability Prediction Methods .......................................................... - 7 2.1.2.1
MIL-HDBK 217 ..................................................................... - 7 -
2.1.2.1.1 Parts Count Prediction........................................................... - 8 2.1.2.1.2 Parts Stress Analysis Prediction ............................................ - 8 2.1.2.2
Telcordia ................................................................................ - 9 -
2.1.2.3
HRD5 ..................................................................................... - 9 -
2.1.2.4
RBD ....................................................................................... - 9 -
2.1.2.5
Markov Model ........................................................................ - 9 -
2.1.2.6
FMEA / FMECA .................................................................. - 10 -
2.1.2.7
Fault Tree ............................................................................. - 10 -
2.1.2.8
HALT ................................................................................... - 10 -
2.1.3 Reliability Growth Models .............................................................. - 11 2.1.3.1
Duane Model ........................................................................ - 12 -
2.1.3.2
AMSAA/Crow Growth Model .............................................. - 12 -
2.2
Model Iteration ............................................................................... - 13 -
vi
3.0
2.3
Development and Reliability Growth Testing .................................. - 13 -
2.4
Uses of Reliability Growth Performance Results ............................. - 14 -
2.5
Growth without Performance .......................................................... - 14 -
MATERIAL AND METHODS ........................................................... - 15 3.1
The Importance of WTG in the Energy Industry.............................. - 15 -
3.2
Technical Aspect Relating to WTGs and the Physics of Operation .. - 15 -
3.3
WTG Total Power Output ............................................................... - 21 -
3.4.1 WTG Components and System Failure ........................................... - 25 3.5
WTG Components .......................................................................... - 26 -
3.5.1 Pitch System ................................................................................... - 26 3.5.1.1
Battery Bank ......................................................................... - 27 -
3.5.2 Gearbox .......................................................................................... - 28 3.6
Reliability and System Failure ........................................................ - 29 -
3.7
Challenges Relating to Failures and Defect of Components of the WTGs ....................................................................................................... - 31 -
3.8
Reliability Analysis Techniques Adequate for WTG Failures .......... - 31 -
3.8.1 Condition Monitoring Techniques ................................................... - 33 3.8.2 Vibration Analysis .......................................................................... - 33 3.8.2.1
Types of Vibrations .............................................................. - 34 -
3.8.2.2
Static and Dynamic Deflection .............................................. - 35 -
3.8.2.3
Variations of Multiple Time Waveforms ............................... - 38 -
3.9
Data and Analysis of Failure ........................................................... - 39 -
3.10
What are considered to be a Failure in WTGs? ................................ - 40 -
3.11
Reliability Parameters of MTTR & MTBF ...................................... - 40 -
3.12
MTBF Analysis Using the Field Data Measurement Method........... - 42 -
3.13
Collection of Data ........................................................................... - 43 -
3.13.1 Definition and Estimation of Size of Population .............................. - 44 3.13.2 Determination of Sample Time Range for Collecting Data .............. - 44 3.13.3 Definition of a Failure ..................................................................... - 45 3.13.4 Downtime ‘Start’ and ‘End’ Time ................................................... - 46 3.13.5 Computing the Annual Failure Rate ................................................ - 47 3.13.6 Converting AFR to MTBF .............................................................. - 47 -
vii 3.13.6.1
Variables Affecting AFR ...................................................... - 48 -
3.13.6.2
Constant Failure Rate Assumption ........................................ - 48 -
3.13.6.3
Population Size ..................................................................... - 48 -
3.13.6.3
Decision Criterion beyond MTBF ......................................... - 49 -
3.14
Analysis of Results ......................................................................... - 49 -
3.14.1 Operations and Maintenance ........................................................... - 49 3.14.2 The WTGs Reliability Growth Pattern ............................................ - 49 4.0
DISCUSSION AND CONCLUSION .................................................. - 51 4.1
Direct Drives and Geared Systems .................................................. - 51 -
4.2
Summary ........................................................................................ - 52 -
5.0
REFERENCES .................................................................................... - 54 -
6.0
ANNEXURE(S) ................................................................................... - 71 -
viii LIST OF ABBREVIATION A
Availability
AC
Alternating Current
AFR
Annual Failure Rate
AH
Average-Hour
AHP
Analytic Hierarchy Process
AMSAA
Army Materiel Systems Analysis Activity
BAO
Bad as Old
BAP
Basic Approaches to Pricing
CD
Drag Coefficient
CG
Centre of Gravity
CL
Coefficient of Lift
CMO
Complimentary Metal Oxide Semi Conductor
CMS
Condition Monitoring System
DCI
Design Change Instruction
DCM
Design Change Management
DFIG
Double Fed Induction Generator
DG
Distributed Generation
EMF
Electromotive Force
FET
Field Effect Transistor
FiT
Feed in Tariffs
FM
Fiscal Measures
FMEA
Failure Mode and Effects Analysis
FMECA
Failure Mode, Effects and Criticality Analysis
G1
Low Wind Generator
G2
High Wind Generator
GC
Green Certificates
HALT
Highly Accelerated Life Testing
HRD 5
Handbook
for
Reliability
Data
for
Electronic
Components IC
Integrated Circuit
IEEE
Institute of Electrical & Electronics Engineers
ix IGBT
Insulated Gate Bipolar Transistor
IO
Input - Output
IV
Current - Voltage
KE
Kinetic Energy
KW
Kilo Watt
KW
Kilo Watt
KWH
Kilo Watt per Hour
M&R
Maintenance and Repair
MCB
Miniature Circuit Breaker
MILSTD
Military Standard
MITA
Wind Power Operating Software
MPC
Main Power Controller
MTBF
Mean Time Between Failures
MTTF
Mean Time To Failure
MTTR
Mean Time To Repair
MV
Medium Voltage
MVAR
Reactive Energy
MW
Mega Watt
MWH
Active Energy
NCR
Non Conformity Report
O&M
Operation and Maintenance
OMS
Operation and Maintenance Services
PE
Potential Energy
PI
A Linear Proportional Integral Controller
PIV
Peak Inverse Voltage
PLC
Programmable Logic Circuit
PLC
Programmable Logic Unit
RBD
Reliability Block Diagram
RGM
Reliability Growth Management
RGP
Reliability Growth Performance
RGT
Reliability Growth Test
RLC
Resistance Inductor Capacitor Circuit
x RMS
Root Mean Square
RO
Renewable Obligations
RPM
Revolution per Minute
RPS
Renewable Portfolio Standards
RTU
Remote Terminal Unit
S&R
Subsidies and Rebates
SCADA
Supervisory Control and Acquisition System
SCR
Silicon Controlled Rectifier
SCS
Suzlon Controlling Software
SEL
Suzlon Energy Limited
SR
Series Model
SSR
Solid State Relay
TCI
Technical Change Instruction
UGF
Universal Generating Functions
UPS
Uninterrupted Power Supply
WPC
Wind Power Controller
WT
Wind Turbine
WTG
Wind Turbine Generator
xi LIST OF FIGURES Figure1: Reliability Growth Begins with Design Iteration through Life cycle analysis……………………………………………………………..……………..-11Figure2: Broken part of the WTG Blade near the Centre of Gravity…….…........-19Figure3: Force analysis at the WTG Rotor…………..……………….…………..-19Figure4: Resolution of Lift and Drag forces on the Aerofoil………...………......-22Figure5: Battery Bank…………………………………………………….……....-27Figure6: The Gearbox………………………………….………………….….......-28Figure7: The Thermograph Instrument…………………………..….....................-36Figure8: Use of the thermograph instrument on the Control Panel…....................-37Figure9: Output of thermograph instrument depicting high rise in Temperature in the cables…...................................................................................-37Figure10: Vibration analyzer..................................................................................-38Figure11: Bathtub curve ........................................................................................-41-
xii LIST OF TABLE Table 1:
Duane Model for Reliability Growth…………………...………….26
Table 2:
AMSAA/Crow Growth Model…………………………..…………26
xiii ABSTRACT We have investigated the reliability of more than 440 Suzlon WTGs and their parts in five states of India and particularly changes in reliability of the mechanical and electrical components coupled with their material consumptions. We first started by considering the failure rate of WT populations, then the annualized frequency rate and then their MTBF. This analysis yields some surprising results about reliability and efficiency of Suzlon WTGs. Then we proceed to consider the major breakdown duration function Tmed variation with time for WT using the downtime and frequency failure rate of these populations. We adopted the field performance data measurement method for the reliability analysis of the errors occurring in the WTGs. This analysis shows that WT gearboxes seem to be achieving reliabilities based on their frequency rate rather than downtime while the electrical system is affected by both. However, WTGs and converters are both achieving reliabilities considerably depending on the stability of the grid. The study also considers different aspects of onsite performance of the maintenance team and ways of improving their approach. Then we conclude by proposing that WTGs should be subject to more rigorous reliability improvement measures, through the internal reliability growth process already in place.
INTRODUCTION 1.0
Wind Power Generation Energy access is essential to tackle global poverty. It needs to happen at the
lowest cost and in the cleanest and most sustainable way possible to help both developed and developing countries establish a low-carbon route to development, it is imperative and noteworthy to observe in most recent times the oil spillage in Gulf of Mexico, the nuclear impasse in Japan and the proactive response of all concerned were as a result of effective environmental management leading to a more developmental deal to handle these creeping disaster. Wind power has a number of economical impacts on power system operation, (Ummels, et al., 2009, p282) which are all related to the low marginal cost of wind power thus contributing towards the fulfillment of the continuously increasing electric (Papaefthimiou,et al., 2009p.298) energy demands and reduction of the pollution caused by the thermal electric energy generating units but it is necessary to reduce cost (Piegari & Rizzo, 2010p318) and increase efficiency in order to make wind energy attractive. The environmental aspects of wind power are the most important driving force behind the development of wind power (Ummels, et al., 2009, p279)
The use of distributed renewable energy resources is proposed as a possible solution for today’s energy challenges .However, the limited controllability, restricted to curtailment and reactive power control if applied, and the overall low capacity factor of renewable energy sources would demand significant transmission and distribution grid reinforcements in an arbitrary integration of these units up to the 20% renewable generation (Alarcon-Rodriguez, et al., 2009p233) objective that has been put forward for the year 2020 globally.
The population of the developing world is expected to increase by 3 billion over the next forty years, and energy demand per capita will grow rapidly as countries economic development proceeds, their per capita consumption of commercial energy increases (Brendan, et al., 2007pp166). The effect of high wind penetration on the operation and control of power systems has been the subject of
-2numerous publications, for example (Hajizadeh & Golkar, 2009; Karki & Billinton, 2004) where technical, operational, and economic (Papaefthimiou,et al., 2009p.299) implications have been identified, mainly resulting from the intermittency and limited predictability of wind power generation. In wind energy’s recent renaissance there was initially little thought given to the issue of variability (Ummels, et al., 2009, p280) except in so far as it affected the total energy production, and thus the revenues from the WTGs. Nevertheless, the relevance of wind’s variability to the design of the WTG was extreme structural loads and fatigue. These issues were in turn closely related to the unsteady aerodynamics of the blades and structural dynamics of the entire WTG (Editorial, 2008pp.1) in unusual or extreme environments with turbines of larger scale (Ummels, et al., 2008pp.35-37) higher reliability and lower cost.
Although grid electrification (Morales, et al., 2008pp.47-48) is the traditional means of providing reliable electricity supplies, connection to distant grids will be too expensive to be cost effective for many developing countries. Fortunately, there are a number of promising alternatives for increasing energy supplies even in these developing nations, based on renewable energy conversion methods using WTGs and these WTs are expected to behave like conventional (Conroy & Watson 2007pp.183) synchronous generators during voltages dips, remaining connected and supplying reactive power during and after the voltage dip has ended.
Hence to ameliorate the numerous problems often encountered during operations of WTGs with its varying components ratings, a reliability model need to be developed because wind speeds larger than nominal wind speed is worst case (Poller, 2008, pp1-5) regarding the stability and fault ride through of the WTGs. During low winds speed periods, the power that can be extracted is obviously reduced (Teninge, et al., 2009, pp359) with emphasis given on the parameterization of input data that describes operational and reliability characteristics of the system, so a global investigation of the examined system can be done. Evaluation (Katsigiannis, et al., 2008pp.75-76) is based on the calculation of system’s basic indices related to reliability estimation and energy production.
-3Understanding the changing O & M methodologies and how the process can become more industrialized to benefit the bottom line and what steps are needed in order to succeed in the short, medium and long term is the challenge facing the industry, however, measuring performance, under performance and output considering best methodological practices can help the industry in understanding the electro-thermal behavior (Jupe & Taylor, 2009, 373) of the network, both in terms of power flows and the identification of components that could be thermally at risk of damage.
Deciphering where failures typically occur from subsystems upwards and how we can drive improvement using effective troubleshooting to minimize downtime, save of manpower time spent and overall profit loss, considering the thin line between fixing a problem and running to failure, it will be very effective to consider achieving the right operational balance for maintenance optimization. Implementation of maintenance procedures and condition monitoring (Tavner, 2008pp.218) for optimized operations will give a clear understanding on the variations in energy capture, and combining it with control to impact on reliability, improve efficiency and long sustainability of wind energy.
However to turn maintenance into proactive rather reactive measure, we have to assess performance through advance pattern recognition by implementing condition monitoring and improving the WTG health management with effective reliability process. The overall function of the electrical system is to collect power from individual WTGs, to transmit and convert it to the appropriate grid voltage and frequency (Quinonez-Varela, et al., 2007pp.108). Electrical systems can be designed depending on the level of collector reliability. Future increased levels of WTGs will also influence the operational (Pearmine, et al., 2007pp142) characteristics of the system and warrant investigation. 1.1
Research Objective The prime objective of this study is to extract information from existing
historic data under prevalent diverse environment, these machines are exposed, so that the reliability of the WTs can be estimated and suggest ways to improve
-4performance of the WTG and its components because the system with lowest cost of energy is considered to be the economically optimal configuration in the work of Lee & Chen (2009, pp152-153)
The data collected from the supervisory control and data acquisition system (SCADA) and the manual maintenance records from the various wind farms including the materials utilized at these sites have being used. The results achieved will however act as a fillip in complementing the efforts of the OMS team to achieve the following: •
Technical Improvement
•
Occupational health and safety of personnel
•
Less unscheduled maintenance occurrences
•
Machine improvement in terms of machine generation, grid(Morales, et al., 2008pp.49) availability, reduction of errors and breakdown
Comparisons have been made between the various models or capacities currently deplored by SEL in their various wind farms. The main purpose of this comparison was to discuss the practical methods of estimating deployment of more reliable models to meet customers’ satisfaction and encourage frequent patronage from prospective ones. It will equally help in predicting (Pecht &Nash, 1994) deployment of larger turbines in the near future. In order to achieve this; WT model and configuration, time, weather and possibly maintenance affected the extracted results and had been used to estimate the reliability of the system further. Even if the wind speed is considered constant for the whole wind farm, its layout can provoke different winds speeds to be considered for different WTGs, (Caramia, et al., 2007pp.116) especially when the direction of wind is taken into account and this direction is made to coincide with the alignment of the WTGs.
-52.0 LITERATURE REVIEW 2.1
Reliability Methods Reliability primarily depends on WTG construction and is intrinsically
estimated whereas availability, yearly production and capacity factor depend not only on reliability but also more strongly on wind conditions and the consequences of faults, which in turn depend on WT location, access logistics and maintenance regime, not primarily to the WTG construction.
Special attention have been given to WTG breakdowns and their associated errors with emphasis on some of the most vital components directly related to these errors, components like rotor blades , air brake , mechanical brake , main shaft ,gearbox ,generator ,yaw system , electrical controls ,hydraulic system, grid system, electrical system , Mechanical or pitch control system and others but the XMW and YMW have been considered.
Various reliability models have been used previously to describe this scenario, such as induction model reliability model (Hamid & Gerald, 2004pp645). Machine life and reliability model, Power law process model, the homogenous Poisson process, Universal generating functions (UGF) , simulative (Monte-Carlo) or analytical techniques based on Markov-based models (Tavner, 2008 and Di Fazio & Russo,2008). However, all the techniques to a large extent provides a comprehensive understanding assessment of WTGs but frequently used are based on Weibull functions. Considerable research has been conducted to develop mathematical models and techniques for the reliability evaluation of power systems containing wind energy (IEEE 90). Monte Carlo simulation technique was used to calculate the frequency indices and can incorporate different state duration distributions. The impact of wind farms on the systems reliability indices using sequential Monte Carlo simulation approach can only yield less results compared to the field data measurement analyses.
-6There are various methods of reliability models and approaches previously used by organisations and companies. •
Reliability Estimation Model
•
Reliability Prediction Model
•
Reliability Growth Model
2.1.1 Reliability Estimation Model 2.1.1.1 Similar Item Prediction Method This method provides a quick means of estimating reliability based on historical reliability data of a similar item. The effectiveness of this method is mostly dependent on how similar the new equipment is to the existing equipment for which field data is available. Similarity should exist between manufacturing processes, operating environments, product functions and designs. For products that follow an evolutionary path, this prediction method is especially useful since it takes advantage of the past field experience. However, differences in new designs should be carefully investigated and accounted for in the final prediction. 2.1.1.2 Field Data Measurement Method The field data measurement method is based on the actual field experience of WTGs. This method is perhaps the most used method by manufacturers since it is an integral part of their quality control program. These programs are often referred to as Reliability Growth Management. By tracking the failure rate of WTG in the field, a manufacturer can quickly identify and address problems thereby driving out WTGs defects. Because it is based on real field failures, this method accounts for failure modes that prediction methods sometimes miss. The method consists of tracking a sample population of new WTG and gathering the failure data. Once the data is gathered, the failure rate and MTBF are calculated. The failure rate is the percentage of a population of units that are expected to "fail" in a calendar year. In addition to using this data for quality control, it also is used to provide customers and partners with information about the WTG reliability and quality processes.
MTBF is commonly used in the industry and it is an indication of reliability, it does not represent the expected service life of the WTG, ultimately an MTBF
-7value is meaningless if failure is undefined and assumptions are unrealistic or altogether missing. Oftentimes the terms “prediction” and “estimation” are used interchangeably as shown above, however this is not correct. Methods that predict MTBF, calculate a value based only on a system design, usually performed early in the WTG lifecycle. Prediction methods are useful when field data is scarce or nonexistent as is the case of the new product designs. When sufficient field data exists, prediction methods should not be used rather estimation should be used (Miller, 2007pp552-563).
2.1.2 Reliability Prediction Methods The earliest methods of reliability prediction (Miller, 2007pp558) came about in the 1940’s with a German scientist named Von Braun and a German mathematician named Eric Pieruschka. While trying to improve numerous reliability problems with the V-1 rocket, Pieruschka assisted Von Braun in modeling the reliability of his rocket thereby creating the first documented modern predictive reliability model (Muhando, et al., 2010pp24). Subsequently, NASA along with the growth of the nuclear industry prompted additional maturation in the field of reliability analysis. Today, there are numerous methods for predicting MTBF. 2.1.2.1 MIL-HDBK 217 Published by the U.S. military in 1965, the Military Handbook 217 was created to provide a standard for estimating the reliability of electronic military equipment and systems so as to increase the reliability of the equipment being designed. It sets common ground for comparing the reliability of two or more similar designs. The Military Handbook 217 is also referred to as Mil Standard 217, or simply 217.
There are two ways that reliability is predicted under MIL STD 217; (MIL-HDBK189) •
Parts Count Prediction
•
Parts Stress Analysis Prediction.
-82.1.2.1.1 Parts Count Prediction Parts Count Prediction is generally used to predict the reliability of a system early in the system development cycle to obtain a rough reliability estimate relative to the reliability goal or specification. A failure rate is calculated by literally counting similar components of a system (i.e. capacitors) and grouping them into the various component types (i.e. film capacitor). The number of components in each group is then multiplied by a generic failure rate and quality factor found in MIL STD 217. Lastly, the failure rates of all the different part groups are added together for the final failure rate. By definition, Parts Count assumes all components are in series and requires that failure rates for non-series components be calculated separately. 2.1.2.1.2 Parts Stress Analysis Prediction Parts Stress Analysis Prediction is usually used much later in the system’s development cycle, when the design of the actual components and sub components are nearing production. It is similar to Parts Count in the way the failure rates are summed together. However, with Parts Stress, the failure rate for each and every component is individually calculated based on the specific stress levels the component is subjected to (i.e. humidity, temperature, vibration, voltage,). In order to assign the proper stress levels to each component, a system design and its expected environment must be well documented and understood. The Parts Stress Method usually yields a lower failure rate then the Parts Count Method. Due to the level of analysis required, this method is time consuming compared to other methods. Today, MIL STD 217 is rarely used (Leonard, C., 1991). In 1996 the U.S. Army announced that the use of MIL-HDBK-217 should be discontinued because it "has been shown to be unreliable, and its use can lead to erroneous and misleading reliability predictions" MIL STD 217 has been cast off for many reasons, most of which have to do with the fact that component reliability has improved greatly over the years to the point where it is no longer the main driver in product failures. The failure rates given in MIL STD 217 are more conservative (higher) then the electronic components available today. A thorough investigation of the failures in today’s systems would reveal that failures were most likely caused by misapplication (human error), process control or product design.
-92.1.2.2 Telcordia The Telcordia reliability prediction model evolved from the telecom industry and has made its way through a series of changes over the years. It was first developed by Bellcore Communications Research under the name Bellcore as a means to estimate telecom equipment reliability. Although Bellcore was based on MIL STD 217, its reliability models (equations) were changed in 1985 to reflect field experiences of their telecom equipment. The latest revision of Bellcore was TR-332 Issue 6, dated December 1997. SAIC subsequently bought Bellcore in 1997 and renamed it Telcordia. The latest revision of the Telcordia Prediction Model offers various calculation methods in addition to those of MIL STD 217. Today, Telcordia continues to be applied as a product design tool within this industry. 2.1.2.3 HRD5 HRD5 is the Handbook for Reliability Data for Electronic Components used in telecommunication systems. HRD5 was developed by British Telecom and is used mainly in the United Kingdom. It is similar to 217 but doesn’t cover as many environmental variables and provides a reliability prediction model that covers a wider array of electronic components including telecom. 2.1.2.4 RBD The Reliability Block Diagram or RBD is a representative drawing and a calculation tool that is used to model system availability and reliability. The structure of a reliability block diagram defines the logical interaction of failures within a system and not necessary their logical or physical connection together. Each block can represent an individual component, sub-system or other representative failure. The diagram can represent an entire system or any subset or combination of that system which requires failure, reliability or availability analysis. It also serves as an analysis tool to show how each element of a system functions, and how each element can affect the system operation as a whole. 2.1.2.5 Markov Model Markov modeling provides the ability to analyze complex systems such as electrical architectures. Markov models are also known as state space diagrams or state graphs. State space is defined as a collection of all of the states a system can be
- 10 in. Unlike block diagrams, state graphs provide a more accurate representation of a system. The use of state graphs accounts for component failure dependencies as well as various states that block diagrams cannot represent, such as the state of a UPS being on battery. In addition to MTBF, Markov models provide various other measures of a system, including availability, MTTR, the probability of being in a given state at a given time and many others (Markov, assessed:20-03-11). 2.1.2.6 FMEA / FMECA FMEA (Failure Mode and Effects Analysis) is a process used for analyzing the failure modes of a product. This information is then used to determine the impact each failure would have on the product, thereby leading to an improved product design. The analysis can go a step further by assigning a severity level to each of the failure modes in which case it would be called a FMECA (Failure Mode, Effects and Criticality Analysis). FMEA uses a bottom to top approach. For instance, in the case of a UPS, the analysis starts with the circuit board level component and works its way up to the entire system. Apart from being used as a product design tool, it can be used to calculate the reliability of the overall system. Probability data needed for the calculations can be difficult to obtain for various pieces of equipment, especially if they have multiple states or modes of operation. 2.1.2.7 Fault Tree Fault tree analysis is a technique that was developed by Bell Telephone Laboratories to perform safety assessments of the Minuteman Launch Control System. It was later applied to reliability analysis. Fault trees can help detail the path of events, both normal and fault related, that lead down to the component-level fault or undesired event that is being investigated (top to bottom approach). Reliability is calculated by converting a completed fault tree into an equivalent set of equations. This is done using the algebra of events, also referred to as Boolean algebra. Like FMEA, the probability data needed for the calculations can be difficult to obtain. 2.1.2.8 HALT Highly Accelerated Life Testing (HALT) is a method used to increase the overall reliability of a product design. HALT is used to establish how long it takes to reach the literal breaking point of a product by subjecting it to carefully measured
- 11 and controlled stresses such as temperature and vibration. A mathematical model is used to estimate the actual amount of time it would have taken the product to fail in the field. Although HALT can estimate MTBF, its main function is to improve product design reliability. 2.1.3 Reliability Growth Models Various models have been used to analyze the reliability growth model (Lamarre,2007 and Duane,1964) amongst them are Duane Model, frequently used because of its simplicity and AMSAA/Crow Growth Model developed for U.S. Army Materiel Systems Analysis Activity (AMSAA) by Dr. Larry H. Crow.
WTG Design
“Pure” Design Portion of R (t) Growth
Design – Test Portion of R (t) Growth
Identified Deficiencies
Analysis Analysis Analysis CA Analysis B WTG ‘A'
Identified Deficiencies
WTG Design
Test Failure Analysis
Figure 1: Reliability Growth Begins with Design Iteration through Life cycle analysis. Reliability Growth Models: •
Duane Model
•
U.S. Army Materiel Systems Analysis Activity (AMSAA) developed by Dr. Larry H. Crow.
- 12 2.1.3.1 Duane Model In this method, failure and the accumulated performance time is calculated and the cumulative failure rate (total failures/ total test time) plotted against it on log-log paper. The equation parameters (K and α) are determined, often by fitting the data points to a straight line by least-square analysis. With this information, the current failure rate and the time required to achieve a desired failure rate can be computed. Duane Model for Reliability Growth is shown in the table below in table 1 (Duane, 1964) Table 1: Duane Model for Reliability Growth Objective
Find failures during test and learn from those failures by remodeling to eliminate them
Key
Relationship between mean time between failure (MTBF) and test
Assumptions time will be a straight line when plotted on log-log paper. Requires that model changes (fixes) be incorporated immediately after a failure and before testing resumes. Parameters
Growth rate, α = change in MTBF/time interval over which change occurred K, a constant which is a function of the initial MTBF T, the test time
Equations
Cumulative MTBF: MTBFc = (1 / K) Tα Instantaneous MTBF: MTBFi = MTBFc / 1 - a Test Time: T = [(MTBFi)(K)(1 - α)]1 / α
2.1.3.2 AMSAA/Crow Growth Model Another popular growth model was developed at the U.S. Army Materiel Systems Analysis Activity (AMSAA) by Dr. Larry H. Crow (Lamarre,2007 and Crow,1983). This model is based on the assumption that reliability growth is a nonhomogeneous Poisson process. That is, the number of failures in an interval of time (or cycles, miles, etc., as appropriate) is a random variable distributed in accordance with the Poisson distribution, but the parameters of the Poisson distribution change with time. It is an analytical model which permits confidence interval estimates to be
- 13 computed from the test data for current and future values of reliability (MTBF) or failure rate. In addition, the model can be applied to either continuous or discrete reliability systems, single or multiple systems, and tests which are time or failure truncated. Some details are provided in Table 2 (Crow,1983). Table 2: AMSAA/Crow Growth Model Objective, Key Assumptions Parameters
Equations
Same as for the Duane Model •
λ, the initial failure rate (1/MTBF)
•
β, the growth rate
•
T, the test time
Cumulating Failure Rate: •
λc = λTβ-1
Instantaneous Failure Rate: •
λi = λβTβ-1
Test Time: •
2.2
T = [λi / λβ]1/(β-1)
Model Iteration Many WTG models are extrapolations of previous models while some are
truly "new" with no predecessor evolving through the system process. In either case, the initial model is subjected to close scrutiny before any prototypes or model are built for performance and long before the final model review. As weaknesses in the initial model of the WTGs are uncovered through analyses of duty performance, the model is changed and the analyses repeated. This iterative pure model process normally continues until it is believe that further model iteration on the basis of analyses of reliability alone, without some performance testing. 2.3
Development and Reliability Growth Testing Ideally the WTG model process would be perfect, with no on duty
performance required to improve reliability to meet the requirement. However, analytical tools, models, and engineering judgment are not perfect, so some development performance is always needed to fill in the gaps in our knowledge and understanding. As performance deficiencies of the WTGs are observed and failures
- 14 are uncovered, the management should take distinct actions to examine the models and tools used to revise, refine, or otherwise improve them.
The analysis itself must provide information on the underlying failure mechanism of the WTG, the probability of recurrence in actual use, and the corrective actions that can be taken to prevent recurrence or minimize the effects of failure. If the WTG model changes are identified as the needed corrective action, reliability growth performance will occur when and if effective changes are incorporated. Often, improvements in reliability are claimed on the basis of planned changes that have yet to be validated. Making decisions based on planned changes is risky. Changes must be incorporated and the effectiveness of the changes in correcting the problem verified. 2.4
Uses of Reliability Growth Performance Results The primary purpose of growth performance of WTG and its components is
to validate the tools and models. Using RGP for determining compliance can affect the way in which such performance is approached. From an engineering perspective, failures in the WTGs or their components are not "bad" because they provide valuable information to the manufacturing plants regarding the adequacy of the models. When RGP is used to determine technical compliance, it becomes a passfail test, and failures are unwelcome occurrence. The original purpose of the trials can become obscured, the motivation to uncover problems compromised, and the real value of the trials lost. 2.5 Growth without Performance Balancing requirements (Kennedy, et al., and 2007pp.166) is demanding task, iteration is needed because not all analyses can be done simultaneously. Consequently, the model may be changed as the result of a particular analysis, only to be changed again when the results of a subsequent analysis are available. As iterations take place, the model is refined, and each revised model is an improvement over its predecessor. Some of the analyses observed in the WTG developmental evolvement shows that the model process directly addresses the reliability of the model. So, the reliability of the model improves as successive model changes are made based on analysis (Pudjianto, et al., 2007pp.10-16).
- 15 3.0 MATERIAL AND METHODS 3.1
The Importance of WTG in the Energy Industry Wind power production varies with wind speed and cannot be controlled
except by curtailing the wind power production (Meibom et al., 2009pp.76). Furthermore , it is only partly predictable and improved maintenance procedure of WTG to increase the machine availability, use of high-capacity machine, low-wind regime turbine, higher tower height, wider swept area of the rotor blade, better aerodynamic and structural design, faster computer-based machining technique, increasing power factor and better policies from government (Bevrani , et al., 2010; Chiang, et al., 2010).The impacts of wind speed variability (Editorial, 2008pp.2) could be addressed by developing accurate wind speed models and estimation techniques (Zeineldin, et al., 2009pp.83) this is often based on the reliability of WTG to be installed.
Extracting information from existing prevalent diverse environment these machines (Billinton & Allan, 1996) are exposed can help to determine reliability of the WTG in order to enhance its productive response with effective safety inspection standards observing the relevant procedural policies in place. There is a wide range of reliability techniques utilized in generating capacity planning and operation (McMillan & Ault 2008) basically; generating capacity adequacy evolutions evolves the development of a generation model, the development of a load model and the combination of the two models to produce a risk model.
3.2
Technical Aspect Relating to WTGs and the Physics of Operation The principal characteristic of a WTG is its rigidity. Under normal
circumstances, its size and shape vary only slightly under stress and changes in temperature , so we have in relation between angular momentum and angular velocity for a WTG rotor rotating about an axis – shaft.
We can define the centre of mass of the WTG (Di Fazio & Russo, 2008pp241); in large multi WTGs, the WT blades turn slowly in the order of 10rpm (McDonald, et al., 2008), the power of a generator is the product of the machine’s
- 16 torque, T and its angular velocity, Ω and so the directly driven electrical torque must produce a very large peak torque. P = T Ω----------------------------------------------------------------------------------------3.0 There are practical limits to electrical and magnetic loading in the electrical machines and thus there is an upper limit to the shear stress that can be developed. The torque of a rotating electrical machine is given by T=2πσR2 l--------------------------------------------------------------------------------------3.1 Where R is the machine airgap radius and l the axial length of the generator
If a generator designed to run at 1500rpm is connected to rotor blades rotating at 15rpm (with a 1:100 gearbox), then the generator only has to develop 1/100th of the torque of the equivalent direct drive electrical machine. If the two generators are of the same type (i.e. they generate the same shear stress, σ in the airgap) but have different dimensions R and l (where the aspect ratio 1/2R = constant), then the direct drive electrical machine will have a radius and length 4.6 (3√100) times greater than those of the original geared electrical machine. Direct drive leads to large, heavy electrical machines which can be expensive to build, transport and install.
One way to help reduce further the performance of the geared WTG is to improve the blade centre of gravity (CG).Using modeling equations for a rotating body on a common axis. •
••
P = M R --------------------------------------------------------------------------------------3.2 Where; M = Total mass of the WTG ••
R = The position R of the centre of mass •
P = The rate of change of momentum Hence, we can say that the rate of change of momentum is equal to the sum of the external forces alone; ••
i.e.
M R = ∑ F --------------------------------------------------------------------------3.3
- 17 Under this assumption of central internal forces; we can know according to the law of conservation of energy that; Kinetic energy (K.E) + Potential Energy (P.E) = Constant In order words; The moment of inertia is central which means; •
J = ∑ r Λ F -----------------------------------------------------------------------------------3.4
Since our WTG rotates about a fixed axis and assuming our angular momentum is constant in other axis, •
J z = ∑ mρvϕ = Iω -------------------------------------------------------------------------3.5
Where; I = ∑ mρ 2 ------------------------------------------------------------------------------------3.6
And; I = Moment of inertia about the z-axis. However, it should be noted that WTG is defined to rotate in a defined axis but yaws as the wind direction changes, hence in cylindrical polar which clearly defines the movement of WTG rotating parts of nacelle and rotor, we can say that the z and ρ
co-ordinates varies according to
•
ϕ = ω ------------------------------------------------------------------------------------------3.7 Where;
ω = angular velocity of the WTG It is obvious from the discussion above that z component is constant and yields •
•
J z = Iω =
∑ ρFϕ --------------------------------------------------------------------------3.8
Equation 3.2 represents the equation of motion of the WTG which determines the rate of change of velocity of the angular velocity. If we consider the Kinetic Energy (K.E) to defined by; 2
1 K.E = ∑ m( ρϕ ) -------------------------------------------------------------------------3.9 2 Then; K.E =
1 2 Iω --------------------------------------------------------------------------------3.10 2
- 18 This means we can express kinetic energy (K.E) in terms of moment of inertia I; hence the equation for the rate of change of kinetic energy can be expressed in this form; •
•
T = Iω ω =
∑ ( ρϕ )Fϕ
= ω ∑ ρFϕ -------------------------------------------------------3.11
However, the momentum equation of equation 3.2 determines the reaction at the axis, which of course has no moment about the axis. It means if we denote the force on the body at the axis by Q, then our equation 3.11 becomes; •
••
P = M R = Q + ∑ F -----------------------------------------------------------------------3.12 Therefore, the centre of mass is fixed in the body, so that; ••
•
•
•
R = ω Λ R + ω Λ R = ω Λ R + ωΛ (ωΛR) ----------------------------------------------3.13
The first term is the tangential acceleration i.e. •
•
R ω in the ϕ direction, and second the radial acceleration
•
ω 2 R in the ρ direction.
Both equations of 3.12 and 3.13 determine the Force at the Q at the axis. From this if therefore will consider as the case of horizontal axis WTG with three blades, with each of the blades signifying respective positions of the WTG blade as it rotates on its axis under gravity. Assuming the forces is acting on the centre of gravity (CG) of the blades and has the component F1, F2 & F3;
⇒ F = F1 + F2 + F3 --------------------------------------------------------------------3.14 F = (Mg,0,0) + (0, Mg,0) + (0,0, Mg) ----------------------------------------------------3.15 For a single blade, F = F1 -----------------------------------------------------------------------------------------3.16 ⇒ F = (Mg,0,0) -----------------------------------------------------------------------------3.17 Then, the equation of motion (3.12) becomes ••
∑ ρFϕ = I ϕ From equation 3.7 ----------------------------------------------------------3.18
- 19 From the Figure 2 below depicting equal spaced blades at an angle of 1200 from each other, with Blade 3 making an angle of ϕ with x-axis while our z-axis as the axis of rotation.
Figure 2: Broken part of the WTG Blade near the Centre of Gravity
Blade 1
Y
ϕ
Blade 2
Blade 3
X
F
C.G
Figure 3: Force analysis at the WTG Rotor
- 20 ••
∴ I ϕ = - MgR sin ϕ ------------------------------------------------------------------------3.19 The equation 3.19 is identical with the equation of motion of a simple pendulum of length l, Where Lenght (L) =
I ------------------------------------------------------------------3.20 MR
For any rotation of the blade, it can be assumed that, 1 2
1 2
L L -----------------------------------------------------------------3.21 2Π = 2Π g MgR Applying the energy conversation to the movement of the blades; E= E1 + E2 + E3 ----------------------------------------------------------------------------3.22 Hence; E = Kinetic Energy (K.E) + Potential Energy (P.E), so equation 3.19, multiplying by •
ϕ and integrating as shown below; ••
•
•
∫ I ϕ x ϕ = - MgR sin ϕ x ϕ --------------------------------------------------------------3.23 ⇒E =
1 •2 I ϕ - MgR cos ϕ ----------------------------------------------------------------3.24 2 •
This means for any given inclination ϕ of the blade, the angular velocity ϕ will be fixed accordingly and the components of the reaction by the Force Q in polar coordinates will be given as follows: ⇒ Q z = 0 ----------------------------------------------------------------------------------3.25 •
⇒ Q ρ = - MgR cos ϕ - MR ϕ 2 -----------------------------------------------------------3.26 ••
⇒ Q = Mgsin ϕ + MR ϕ ----------------------------------------------------------------3.27
From equation 3.19; ••
ϕ=
- MgR sin ϕ ---------------------------------------------------------------------------3.28 I
And from equation 3.27;
- 21 •
E + MgR cos ϕ --------------------------------------------------------------------3.29 1 I 2
ϕ =
Therefore using equations 3.26 & 3.27, we can determine the Force Q by writing equations 3.24, 3.25 & 3.26 as follows; ⇒Q
z
= 0 -----------------------------------------------------------------------------------3.30
E + MgR cos ϕ -----------------------------------------3.31 ⇒ Q ρ = - MgR cos ϕ - MR 1 I 2
- MgR sin ϕ ⇒ Qϕ = Mgsin ϕ + MR -------------------------------------------------3.32 I It is very important to determine the Force Q at the axis as the blades rotates on the rotor, however, as the inclination ϕ changes, this force continues to change and can be determined. From the foregoing, it is very clear that the wind energy is available in the form of kinetic energy, depending on the turbulence, will be imparting on WTG at varying angles and inclination with unequal torques (Muyeen, et al., 2007pp134) and if the centre of gravity (CG) of the blade is improved by locating it towards the tip of the blade not increasing the weight then more force will be developed at the rotor enhancing the power developed by the WTG. So the blade tip can be redesigned to minimize loads and reduce noisy levels, also minimizing root leakage will provide greater lift. Hence taller towers are an essential complement to longer blades. Longer blades capture more energy and in turn improve return on investment for wind farm developers. 3.3 WTG Total Power Output The objective of power control is to maintain the output power of a WTG and
subsequently the wind farm at a certain level even with stochastic wind (Geng & Yang, 2009; Mohammed & Nwankpa, 2010). The controller can not only level the wind energy conversion but also be applicable in a wide wind speed region even subject to large parametric or non parametric disturbances. The Bernoulli equation which is the conservation of mechanical energy, in terms of kinetic energy, can be used to calculate the dynamic pressure, which is the difference between the total pressure and static pressure, this means if the velocity is
- 22 zero at stagnation pressure, then the pressure will be maximum at that point giving the total and static pressure, hence if velocity is zero at stagnation points, then the pressure will be maximum at same point.
As a condition, the velocity of the WTG must always be less than the wind velocity V∞ and because every force stream velocity gets converted into power, the above equation becomes the maximum lifting system as shown in the equations below:
v∞
Fr
V∞
L
α
v
W Fq
D
Figure 4: Resolution of Lift and Drag forces on the Aerofoil
From the figures 17 above; tan α =
v ---------------------------------------------------------------------------------3.33 V∞
v V∞ v P = C L (1 2)lV∞2 A 1 − --------------------------------------------------------3.34 E V∞ From equation 3.34 above;
(1 +
v2 L ) ; E = ------------------------------------------------------------------------3.35 2 V∞ D
C pmax =
Pmax 1 lAV∞ 2
= CL
2E 4E (1 + 9 9
-----------------------------------------3.36
- 23 The upward force was as a result of the upward movement of this complex relationship not merely because of the rotational velocity ωR of WTG blade and the wind speed as seen by the blade is V∞ , so the forces can be resolved into the rotational force Fr which turns the blade, so the thrust force (Brendan, et al., 2007pp166) on the rotor has to be resisted by the structure, Fq, hence the velocity ratio for maximum power coefficient is given. The force which will be available for conversion of this wind power is given as; T = m(V∞ - V) = lAv(V∞ - V) -------------------------------------------------------------3.37 Since the swept area A of the rotor is known and the area of the wind velocity V is not in our control. Thrust can also be written as
T = A∆p -----------------------------------------------------------------------------------3.38 Where
∆p is the difference between the two pressures developed upstream and
downstream of the TG and it is obvious that the only term changing is the T=l
v=
∆p
A 2 (V ∞ - V 2 ) , by Bernoulli equation---------------------------------------------3.39 2
(V∞ - V) ; By Momentum expression-----------------------------------------------2
3.40 That implies the blade velocity v is an average of the initial and final velocities; hence, (V∞ - v ) = aV∞ hence (V∞ - V) = 2aV∞ ----------------------------3.41 where ‘a’ is a factor which will depend on the interference which is created by the turbine this factor ‘a’ is what determines the reliability of the WTG (Burton, et al., 2001), any input taken to optimize the WTG directly affects the value of ‘a’. From Ist law of thermodynamics when pressures are same on both sides and there are no observed changes in the temperature condition of the system, pressure can be expressed as; 2
P = lAv{
V2∞ V } -----------------------------------------------------------------------3.42 2 2
- 24 Where; V = downstream velocity V∞ = upstream velocity A = Swept area of the rotor
v = blade velocity l = Air density
Equation 3.42 translates to = l
A v(V∞ + V)(V∞ - V) --------------------------------3.43 2
From 3.41 (V∞ - V) = 2aV∞ P
Therefore l
A 3 V ∞ 2
= 4a(1 - a 2 ) ---------------------------------------------------------3.44
This 4a (1 – a2) becomes a maximum when a =1/3, which means the maximum we can derive is 59.3%, however it is to be noted that it is never the total velocity which gets converted into power, rather as the blades rotates we can only extract 59.3% of the wind power available. The WTG extracts energy from the air, so this interference which is known as axial interference factor is a measure of reliability, it actually defines the influence of the turbine on the air converting only 59.3% of the available energy and depending on the efficiency of the WTG components, and the maximum we could extract can never be more than 59.3%.
The WTG components are mechanically structured, hence higher solidity ratios [26] are needed for these parts to work efficiently while it is reverse for the electrical components were lower solidity ratios are needed, however, higher solidity ratio signifies higher torque production and lower torque for the electrical components. Solidity ratio for the WTG is defined as; total surface are of the blades ------------------------------------------------------------3.45 sweep area of the blades It is obvious for a mechanical system, higher torques will make the WTG to be very big and robust but in order to sustain an efficient WTG for power
- 25 production, it means we will need low torque system making the reliability of the WTG to depend largely on the electrical systems, because we will need the surface area of the mechanical parts to be small considering that the amount of mass flow that will flow through the system is AV and not AV∞.
3.4.1 WTG Components and System Failure WTG components and parts are connected electrically in the control panel
and each of their status can be viewed from it. Depending on whether it is geared or gearless (Bleijs,2007; Polinder,et al.,2006) the primary purpose of all WTGs remains effective generation of power under varying conditions of the wind energy (Billinton & Chowdhury 1992; Ameri et al.,2006). Each of these components is electromechanically (Bleijs, 2007) connected and the failure of any leads to total ineffectiveness of some others. Most models used in estimating the failure rates of the TGs considers them either in series or parallel which implies that the failure of each component directly or indirectly affects the functioning of other parts, which may in some case lead to fatal accident or some undesirable conditions. However, compliance with request to back down if needed on considerations of grid security or safety to equipment/ personnel, during start-up WTGs should ensure that MVAR drawl shall not affect (Brendan, et al., 2007; Billinton, & Hua Chen 1998) grid performance. Relevant to these replacement policies should be that WTGs are efficient, reliable and efforts made to evacuate available wind power and treat as must-run stations, also observing procedures for connection and site responsibilities, depending on capacity payment linked to plant availability to recover fixed costs, energy charges and variable charges like fuel cost. Reliability is the ability to produce useful data over a period of time. Motors with intermittent duty operation had a failure rate that was about half that of those with continuous duty and it is not known how different this would have been if running time had been used rather than calendar time (Tavner, et al., 2006pp.1-18).
- 26 Wind power can also be considered as distributed generation (DG) (Billinton & Allan, 1996); hence its reliability evaluation and enhancement are an important factor in modern power system planning and operation.
Another important feature of wind turbines is the productivity of wind turbine, referred to as capacity credit (Billinton & Allan, 1996; Bianchi, et al., 2008). Wind capacity credit is significant technology issue related to the substitution of conventional power installed capacity. There have been a number of valuable references comparing the productivity of different WT configurations (Billinton & Chowdhury 1992).
However, no accurate analysis has yet been made including reliability in these comparisons. 3.5
WTG Components Most WTG components are known to often experience failures and few of
these we have considered in this report. 3.5.1 Pitch System The main purposes of pitch control are to limit the maximum thrust and
torque and to give controllability (i.e., to be able to stop the rotor safely in a fast flow without relying on a brake). It must also be able to accomplish diverse tasks such as (Bianchi, et al., 2008) power production maximization, power balance, delta and gradient control. The pitch system was designed for large wind turbines when the rotor diameter became larger, for the purpose of reducing thrust, loads and to make the WTGs operate more stable considering the high failure rate. As compared to a stall regulated turbine which need to withstand blade root bending moments that can be two or three times (or more) greater than with a pitch controlled rotor. Pitch control is also used as part of a patented feature to allow efficient operation in a bi-directional flow of the wind by pitching the rotor blades through 180 degrees when the wind direction changes. Of course, there are also efficiency gains from pitch control, so it will also help by generating more electricity than a stall regulated fixed pitch system.
- 27 In a hostile wind environment this will happen sooner rather than later. So, the simple philosophy would be to limit the number of moving parts to lengthen the life span of a WTG and minimize maintenance requirements. Active power and frequency control are requirements which the pitch control action performs (Tsili & Papathanassiou, 2009). Wind power plants can actively take part in grid operation and control by regulation (Bevrani et al., 2010) of their output power, the large increase in the installed wind capacity in transmission systems necessitates that wind generation remains in operation in the event of network disturbances (Ma & Chowdhury, 2010). Reactive power regulation capabilities are requested by many grid codes. This can be done by externally providing a specific reactive power value or by a specific power factor. The pitch drive can be said to a vector controlled mechanism because it works according to direction as well as magnitude sensing and helping the blade to pitch in the desired angle and direction. 3.5.1.1 Battery Bank The pitching of the blades when grid is not available is aided by power
provided by the battery banks; it is very useful and acts more as a safety device. (Yang, 2008). This is for battery top up if for a long period of time (Jenkins, et al., 2008) the WTG is predominately running sub-synchronously, as well as to provide a route to discharge battery if the opposite happens. If preferred the battery charging /discharging can be controlled to take place when the wind speed is less than the cut in speed, or greater than the cut-off speed.
Figure 5: Battery Bank
- 28 Two of the battery banks are connected in series to give 264 V DC for single motor with a diode connected between them to ensure flow of current in one direction from battery charger to battery bank . 3.5.2 Gearbox
A gearbox provides speed and torque conversions from a rotating power source to another device using gear ratios. A total rotor torque is transferred to the generator shaft by a transmission system (Khan, et al., 2010) consisting of a rotor shaft, a gearbox with gear ratio G and several bearings.
Figure 6: The Gearbox The gearbox is usually lubricated by forced lubrication system to maintain their rotating condition to avoid friction and save internal parts. It is pressure sensitive control device which is used to maintain the pressure within predetermined values irrespective of the other conditions. As the pressure exceeds the predefined value it changes its contacts.
The diameter of the rotor shaft is less than the diameter of the gearbox shaft (Li & Chen, 2008pp124) in this case it is used to maximize the tensile strength and high compression force.
- 29 The active and reactive powers (Foster, et al., 2010) of the stator depend on the phase and magnitude of the induced EMF, unlike traditional control strategies such as stator flux orientation vector control and flux magnitude and angle control, magnitude and frequency control manipulates the magnitude and frequency of the rotor voltage. This simplifies the design of the control system and improves the reliability (Wang & Hsu, 2010; Wang, et al., 2010).The co-ordinate transformations, rotor position detection, and measurement of rotor currents and rotor speed are not required.
3.6 Reliability and System Failure
The modeling of windfarms for reliability or grid integration studies is now an important issue, owing to the significant increase in connected wind power capacity (Conroy & Watson, 2009pp.40). In a power system model, a conventional generating plant will have each of its individual synchronous generators modeled separately. On the hand, a windfarm contains a large number of individual WTGs, possibly in our case more than 440 WTGs. Modeling each of these separately increases the complexity and compromises the correctness and effectiveness of the desired result. In a power system that may contain a large number of windfarms (Jupe, et al., 2010pp.285) the need for aggregate windfarm models becomes obvious. This study looks at the aggregate modelling option considering the reliability of the WTGs and their components. Data from supervisory control and data acquisition system (SCADA) and the maintenance records (Petersson, 2005) respectively from the wind farms may be used to achieve the target of defining the health condition of these WTGs. The rapid growth of the wind energy industry and the colossal size of the equipment involved make high reliability standards very important. As the wind energy market matures, (Ault et al., 2007) developers are being forced to site their projects in rugged terrain as the number of undeveloped, less complex sites dwindles. A good wind site considers two aspects, that is wind resource and WTG, which means to carry out wind site survey both conditions that would affect the wind energy output from the site and also non-wind conditions such as soil type,
- 30 building and structures and so on that will affect the choice of the WTG (Prasad, et al., 2009) are considered.
Rugged, forested terrain introduces a host of issues relating to increased wind speed, turbulence and wind behavior that can make WTG reliability an imperative hence, determining where the turbulent spots are, which can overload (Sayas & Allan, 1996) the turbine and lead to early failure hence reducing its lifetime. A slope changes the angle of the wind as it hits the WTG rotor, which in turn, has an effect on the turbine loading and production. Higher turbulence and increased loads leads to early failure contributing to lower reliability hiking operation and maintenance costs, while production from the turbines is lower if they are sited incorrectly. It is more difficult to identify the optimal sites for turbines in complex terrain, but determining reliability of these turbines will enable us to select sites where the load is manageable and can be curtained optimally if we want to maximize production and minimize loads to achieve ultimate balance. These induction generators may not be able to contribute to power system regulation and control in the same way as a conventional field excited synchronous generator (Chen, et al., 2007). The induction generators need reactive power support, ideally, to be connected to stiff grids.
The cost of developing wind farms on complex sites is higher though but there are fewer options on access to the WTGs because the physical terrain features forces to route in other directions i.e. not necessarily the shortest path, this means it is also difficult to (Arabian, et al., 2010pp.192) move large equipment when we have such limited access. But if the WTGs are positioned in relation to reliability standards as well as where the cost of electricity production justifies the installation cost, then it makes sense to do it.
- 31 Complex terrain as found in the hilly regions of Karnataka state in India is often at higher elevations, which generally implies a better wind resource and higher WTG reliability need to be deplored. But along with higher wind speed, we have higher turbulence. Again the reliability of the WTG comes in. The rapid development of onshore wind in India has raised some real concerns for WTG reliability; these have made onshore wind developments despite a record level of growth in India, to reconsider reliability issues along with the unprecedented growth.
3.7
Challenges Relating to Failures and Defect of Components of the WTGs
Developments in the industry have sought to address these issues, with companies like Suzlon Energy Limited discussing where to move forward in the space, there is no doubt that increased communication and collaboration within its key segment and structures throughout the different continents where located is imperative to identifying and successfully addressing the essential strategies that will need to be employed. However, it has been observed (Hamid & Gerald, 2004pp656) that a wider, less exclusive and more open internal communications are needed for all aspects of the WTG assemblies to function effectively and share key remote alignment in this crucial area to achieve maximum reliability.
3.8 Reliability Analysis Techniques Adequate for WTG Failures
Reliability engineering is a field of engineering that drives product evolvement through the various stages of design and manufacturing. It is a measure that every company uses to drive its quality and maintain a high level of competition in the market especially in a growing industry as obtained in ours. The analysis of system implications is a complex matter (Klobasa, 2010pp57) to analyze the impacts of wind generation on WTGs and eventually on the power system operation should be based on time series and power system analyses.
Determining the impacts of such connection arrangements is more complex than with the traditional assessment at (Boehme, et al., 2010pp67) critical loads in response to grid faults, a sharp voltage dip at the generator terminals is detected and may sustain longer than the fault duration. The stator currents will dramatically
- 32 increase and exceed their rated values. Consequently, larger rotor currents will be observed pushing the dc link voltage to higher values and might lead to converter(Melicio, et al., 2010,pp141) damage, further electromagnetic torque will fluctuate which may cause a mechanical stress (Kasem, et al., 2008pp202) on the drive train system of the WTG. As a result the WTG will disconnect from grid. Using long term performance records with detailed network analysis is necessary to account for the differences in output of the wind farms at any moment. The correlation of the output of the wind farms is a complex matter, but application of appropriate time series analysis can be effective in estimating the reliability of the WTGs and provide a link between what is the cost that wind power imbalances actually cause to power system and what is the cost allocation to wind power producers (Helander et al., 2010pp76) through balance settlement. Previously, research has focused on the problem of windfarm modelling for reliability studies mainly at the generation level (Di Fazio & Russo, 2008, pp243) to assess the generation capacity adequacy in the presence of WTGs. However, some interesting models have proposed starting from a different perspective i.e. looking from the WTGs into the network. The aim of this approach is account for the effects of all the main factors affecting windfarm generation and provide models used in reliability analysis by WTG manufacturers, windfarm owners as well as by distribution and transmission system operators. In view of the above, technical impacts of high levels of WTGs on the low voltage networks (Trichakis, et al., 2008pp254) are often focused on voltage regulation, cable and transformer thermal limits, network losses etc. The reliability analysis must provide information on the underlying failure mechanism of the WTG, the probability of recurrence in actual use, and the corrective actions that can be taken to prevent recurrence or minimize the effects of failure. If the WTG model changes are identified as the needed corrective action, reliability growth performance will occur when and if effective changes are incorporated. Often, improvements in reliability are claimed on the basis of planned changes that have yet to be validated. Making decisions based on planned changes is
- 33 risky perhaps inspection is critical (Bakirtzis, 1992); findings and changes must be incorporated and the effectiveness of the changes in correcting the problem verified.
3.8.1 Condition Monitoring Techniques To keep the WTGs in operation, implementation of condition monitoring
system (CMS) and vibration analysis is paramount and for this purpose ample knowledge of these two types of systems is mandatory (Hameed et al., 2009; Hameeda et al., 2009). One means of achieving this level is to devise and implement efficient, adaptable and responsive systems of condition monitoring and vibration analysis especially on rotational structures (Tavner, 2008pp.220) of the WTG. Condition monitoring system plays a pivotal role in establishing a conditionbased maintenance and repair (M&R), which can be more beneficial than corrective and preventive maintenance. With CMS, a prediction of impending failure is given for each component, and therefore M&R’s can be better scheduled. The CMS for the gearbox primarily measures vibrations, but a supervision of the oil is also necessary. So we will always have to consider if the mass of the system should be increased or its stiffness. There is always a correlation between the mass and stiffness of the system. 3.8.2 Vibration Analysis Vibration is defined as a to and fro motion about a reference point. Any
object which moves to and fro creates a mechanical vibration with some magnitude. This vibration is useful in most cases but should have a limit .The adverse effect of vibration is always very high once it crosses its limits.
For every vibrating system, there must be a •
Spring
•
Mass
•
Damper
The equations below describes the these quantities in their proportion S F = R s x X --------------------------------------------------------------------------------3.46
- 34 F = M x a -----------------------------------------------------------------------------------3.47
D r = D c x V --------------------------------------------------------------------------------3.48 F = kX + mX " + cX ' -----------------------------------------------------------------------3.49 Where: SF
= Spring Force
Rs
= Spring Rate
X
= Deflection
a
= Acceleration
Dr -
= Damper
Dc
= Damping Coefficient
v
= Velocity
M
= Mass
F
= Force applied While k, m & c are Constants
3.8.2.1 Types of Vibrations There are two types of vibrations:
•
Forced Vibrations
•
Free Vibration
The Free Vibration is due to a single disturbances force with: •
Applied frequency of oscillation remaining same irrespective of force applied
•
Higher the force level the more amplitude of vibration1 and higher the setting time
In considering the effect of mass when force is applied, the natural frequency reduces while stiffness (Shek, et al., 2007pp.23) increases natural frequency of the system, this decides if the mass of the system should be increased or its stiffness. There is always a correlation between the mass and stiffness of the system in this regard.
- 35 However increasing damping force (Liu, et al., 2010, pp200) decreases the amplitude and shortens the setting time reducing the amplitude during resonance under forced vibrations. 3.8.2.2 Static and Dynamic Deflection During forced vibration applied forces which induce continuous vibration
with mass of the WTG remains constant with stiffness changing due to the vibration increase. The changes may be due to the horizontal or vertical deflection Ss =
F F or -------------------------------------------------------------------------------3.50 X Y
Where; X – Deflection in the x axis-direction Y - Deflection in the y axis-direction In WTGs, vibration changes because of the changes in stiffness and this may occur even at slightness loosening of a bolt. Natural Frequency is the frequency at which a system would like to vibrate .Natural frequency is dependent on the mass and stiffness of the system (Yang, et. al. 2009pp9), a system can have number of natural frequencies according to the number of component making it up, this is not necessarily a problem, but becomes a problem if the excitation frequency of the WTG coincides with the system natural frequency. Condition monitoring technology has been developed and is applied to WTs, presently as signal processing because of its convenience and low cost. These systems (McMillan & Ault, 2008, pp66) provide information to wind manufacturer with the goal of improving operational efficiency via more informed decision making capturing the deterioration and failure characteristics of WT. Resonance is a condition which occurs when natural frequency of a system coincides with excitation frequency. It is characterized by substantial amplitude increase and related phase shift and its impact is reduced with good damping characteristics. Excitation sources could be as a result of unbalance, misalignment, blade or vane passing pulsation etc.
- 36 During vibration, forces are transmitted from the rotor to the foundation and in each of these cases, spring mass system analysis is used to consider the effect of these transmission of forces, there are about six types of spring mass systems in WTGs(Yang,et.al.2009,pp9) nevertheless it is pertinent to identify components/parts associated with each and their sources. So the bearing is not the only source of vibration whenever it occurs, we should be able in any circumstance to identify the source of any vibration and where applicable then the bearing can be our last resort of analysis. Force applied to a generator or part is not equal to the vibration dissipated (Shek, et al., 2007pp.23), as the system response or mobility will contribute to it, making our vibration different from any isolated part, this is because of the stiffness associated with the equipment which depends directly on the weight of the equipment (Faried, et al., 2010pp305) and should be more stable to hold any coming vibration.
Vibration = Input forces + System response (mobility) --------------------------------3.51 The vibration condition to be monitored depends on what we are measuring, like gearbox, mostly all mechanical component have to adapt to this analysis because vibration analysis is an indication of the health condition of the WTG while in operation, but in a gearbox, we can monitor the condition through oil analysis; same for the bearing teeth etc, with the understanding that the oil is circulating everywhere within the internal parts of the bearing.
In the case of control panel, we cannot do vibration analysis rather specialized tools like thermograph and power analyzers are used. The thermograph instrument is shown below.
- 37 -
Figure 7: The Thermograph instrument Where loose ends exist, they are hot spots and with the use of thermograph they can be located and corrective actions taken (See figure 24 below). Standards in condition monitoring are generalized as a guide and can be used to know the operational limits only, there is no thumb rule to this but the use of vibration analysis can be used to know which component is experiencing problem.
Figure 8: Use of the thermograph instrument on the Control Panel
- 38 -
Figure 9: Output of thermograph instrument depicting high rise in temperature in the cables. 3.8.2.3 Variations of Multiple Time Waveforms Transient stability is the ability of the power system to synchronize when
subjected to a severe transient disturbance (Faried, et al., 2010pp301) stochastic models as well as the spring constant of the two mass shaft model of the WTG can be investigated to estimate reliability (Mohammed & Nwankpa, 2010pp88).
Figure 10: Vibration analyzer:
- 39 Everything in the WTG vibrates, so we have to know the size and phase angle. The phase angle indicates the angular displacement or deflection resulting in the stiffness of the body or equipment. Common causes of vibration are unbalance, bearing damage, misalignment, faulty gear-mesh, mechanical Looseness, electrically Induced unbalance. Electrically induced imbalance in the air gap causes vibration as the flux will not be proportional within the air gap, so before starting any vibration analysis, the speed, gears, fans, drawing of the gear box, gearbox mesh frequency should be well understood, and a thorough study of the components done and evaluated. The higher component frequency must be given priority because frequency shows the sources of the failure or impending problem. 3.9
Data and Analysis of Failure There are two basic definitions of a failure:
•
The termination of the ability of the WTG as a whole to perform its required function (IEEE 90).
•
The termination of the ability of any individual component like gearbox, rotor etc; to perform its required function but not the termination of the ability of the WTG as a whole to perform (IEEE 90; Hamid & Gerald, 2004pp657).
These assumptions are required to simplify the process of estimating MTBF. It would be nearly impossible to collect the data required to calculate an exact number. However, all assumptions are realistic. Throughout this study common assumptions used in estimating MTBF are described. MTBF impacts both reliability and availability. The difference between reliability and availability is often unknown or misunderstood. High availability and high reliability often go hand in hand, but they are not interchangeable terms. Reliability is the ability of a system or component to perform its required functions under stated conditions for a specified period of time (IEEE 90).
- 40 In other words, it is the likelihood that the system or component will succeed within its identified mission time, with no failures. Availability, on the other hand, is the degree to which a system or component is operational and accessible when required for use (IEEE 90). Clearly, the importance of defining a failure should be evident and must be understood before attempting to interpret any MTBF value. Assumptions are required to simplify the process of estimating MTBF. It would be nearly impossible to collect the data required to calculate an exact number. However, all assumptions must be realistic. Throughout this study common assumptions used in estimating MTBF are described. 3.10
What are considered to be a Failure in WTGs? However it is quite impossible to collect the actual or required data needed
for calculation as human errors and lack of sufficient information may be a hindrance. 1.
It can be viewed as the likelihood that the system or component is not in a
state to perform its required function under given conditions at a given instant in time. Availability is determined by a system’s reliability, as well as its recovery time when a failure does occur. When systems have long continuous operating times (for example, a 20-year WTG), failures are inevitable. Availability is often looked at because, when a failure does occur, the critical variable now becomes how quickly the system can be recovered. Having a reliable WTG design is the most critical variable, but when a failure occurs, the most important consideration must be getting the WTG running as fast as possible to keep downtime to a minimum. Downtime is a serious condition (Sae-Kok, et al., 2010pp460) and should be avoided. 3.11
Reliability Parameters of MTTR & MTBF Mean Time between Failure (MTBF) is a basic measure of a system’s
reliability. It is typically represented in units of hours. The higher the MTBF number is, the higher the reliability of the WTG. Equation 3.52 illustrates this relationship (Miller, 2007pp556).
λ(t) = ρβ l-(β t ) --------------------------------------------------------------------------------3.52
- 41 Where λ is the failure rate and can be modified appropriately to define the Reliability.
MTBF is not equivalent to the expected number of operating hours before a system fails, or the “service life”, this is why these numbers are often so high because they are based on the rate of failure of the WTG while still in their “useful life” or “normal life”, and it is assumed that they will continue to fail at this rate indefinitely, in this phase of the WTGs life, it is experiencing its lowest (and constant) rate of failure. In reality, wear-out modes of the WTG would limit its life much earlier than its MTBF figure. Therefore, there should be no direct correlation made between the service life of the WTG and its failure rate or MTBF. It is quite feasible to have a WTG with extremely high reliability (MTBF) but a low expected service life. MTTR, or Mean Time to Repair (or Recover), is the expected time to recover a WTG or its component from failure. This may include the time it takes to diagnose the problem, the time it takes to get the repair maintenance onsite (MIL-HDBK781A) , and the time it takes to physically repair the system. Similar to MTBF, MTTR is represented in units of hours. As equation 6.2 shows, MTTR impacts availability and not reliability. The longer the MTTR, the worse off a system is. If it takes longer to recover a system from a failure, the system is going to have a lower availability. The equation 3.53 below illustrates how both MTBF and MTTR impact the overall availability of a WTG. As the MTBF goes up, availability goes up. As the MTTR goes up, availability goes down. Availability =
MTBF ---------------------------------------------------------3.53 MTBF+ MTTR
For Equation 3.52 and 3.53 (Miller, 2007pp557; Lamarre, 1998pp376).above to be valid, a basic assumption must be made when analyzing the MTBF of a WTG. Figure 11, referred to as the bathtub curve, illustrates the origin of this constant failure rate assumption mentioned previously.
- 42 Early Failure
Normal Operating Wear out
Period
Period
Period
Rate of Failure
Constant Failure Rate Region O
Time
Figure 11: Bathtub curve (Piegari & Rizzo, 2010; Miller, 2007pp556)
The "normal operating period" or “useful life period" of this curve is the stage at which the WTG is in use in the field. This is where WTG has leveled off to a constant failure rate with respect to time. The sources of failures at this stage could include undetectable defects, low design safety factors, higher random stress than expected, human factors, and natural failures. Ample burn-in periods for components by the manufacturers, proper maintenance, and proactive replacement of worn components should prevent the type of rapid decay curve shown in the "wear out period". The discussion above provides some background on the concepts and differences of reliability and availability, allowing for the proper interpretation of MTBF (Piegari & Rizzo, 2010).
Methods that estimate MTBF have been used because they represent actual measurements of failures. MTBF estimation is by far the most widely used method of calculating MTBF, mainly because it is based on the particular WTGs that are experiencing actual usage in the field. This method is statistical in nature providing only an approximation of the actual MTBF. No one method is standardized across an industry. 3.12
MTBF Analysis Using the Field Data Measurement Method Avoiding failures in any WTG is always a top priority. When minutes of
downtime can negatively impact the market value of a business, it is crucial that the
- 43 maintenance and operations environment be reliable. MTBF is the most common means of comparing reliabilities. Ultimately MTBF is meaningless if the definition of failure is not clear or assumptions are unrealistic or misinterpreted.
Several methods are shown above for predicting MTBF (Pecht & Nash, 1994). The Field Data Measurement Method uses actual field failure data and therefore is a more accurate measure of failure rate than simulations. The basic assumptions made are: •
This data may not be available for WTGs manufactured in high volume like the Suzlon WTGS
•
3.13
It is based on the constant failure rate assumption
Collection of Data We were faced with so much data to analysis, over 460 WTGs, but after
sufficient review it was noted that only 440WTGs have available information needed. Out of these number 193WTGs were YMW models and 247 were of XMW along with their material consumption pattern. A technical approach unique to this huge data has been used to define a practical and direct solution of calculating uniformly the various stages of reliability parameters of MTTR, MTBF and may be consider them in series or parallel for each of the models. But due to the nature of the data which has been made simple by the SCADA we had to use field data measurement method.
However, it was observed that downtime and frequency are having different values, we decided to decipher why the variations as it was given different results, hence a unique statistical method had to be employed to evaluate these discrepancies as noted. Besides failure rates, the downtimes of the machines (Hahn et. al., 2007pp329) after a failure are an important value to describe the reliability of a machine.
- 44 As a result of the variations in the downtime and number of incidents, we determined the TMED which simply logically represents the major breakdown time to be used for models considered in order to have comparative bases to evaluate the performance of the maintenance team along the WTGs and its components resilience pattern. It will interest us to note specifically the determination of the operating major breakdown duration was done in order to classify the failures. TMED was used to limit breakdown incidents and considered failures occurring either on frequency type or downtime type or both and selected the range above 28.95mins, evaluating the performance of the components and errors affecting them and identifying the origin of such errors is a good beginning of root cause analysis.
3.13.1 Definition and Estimation of Size of Population The first step in the process of determining the annual failure rate (AFR), and
ultimately the MTBF of the WTG, is to identify a reasonable population size to be analyzed. The calculation was based on two models of the WTGs for ease of comparison. Though the dates were varying for each of them but a period of 59weeks was maintained in order to have the WTG exposed to same conditions and seasons.
3.13.2 Determination of Sample Time Range for Collecting Data The data failure rate varies for each of the models, a total of five states where
considered namely: •
Tamil Nadu with a total of 247 nos. of XMW & 38 nos. of YMW
•
Maharashtra with a total of 20nos. of YMW
•
Rajasthan with a total of 103 nos. of YMW WTGs
•
Gujarat with a total of 12nos. YMW WTGs
•
Karnataka with a total of 24nos. of YMW WTGs
A total of four (4) WTGs of YMW model does not have complete data as required so was neglected from the study.
- 45 The data from the SCADA is often recorded automatically when the WTG suddenly experiences any error or mal functions. If at any time these errors are not recorded they are to be manually entered at the time of restoration of operation by the SCADA. The appropriate amount of time between the commissioning date and the date of extraction of data had been used. There are two important variables here: (1) Sufficient time between each WTG’s commissioning date and the date of extraction of data (2) A total of 440WTG’s - big enough window of data collection to ensure confidence in the results, though four (4) WTGs lack sufficient data and have been excluded from calculations. Under this condition two effects happen. First, since WTGs that are not deployed cannot fail, there is a tendency to underestimate the failure rate. The second effect is that the sample period tends to include a large number of varying numbers of WTGs. With new WTGS less than one year, their lifespan may exhibit a failure rate of the classical "bathtub" shape. Although we know that both of these counteracting effects are very strong, we can't say that they balance one another.
The other important consideration with regard to sample time is the duration of the window which we choose irrespective of the size of the WTG. How many days are adequate to collect data on failures? The sample time window was chosen to be wide enough to remove statistical "noise" from the sample. The duration needed to obtain reasonable accuracy is dependent on the size of the population which in this case was large enough. 3.13.3 Definition of a Failure In the Wind industry, the most popular “definition” of a WTG failure is a “load
drop” failure. This means that the power supplied to the grid fell outside the acceptable limits and caused the load to turn off. However, is it useful to count load drops caused by other factors.
- 46 •
Is it possible that the WTG design itself increases the failure probability of an already risky procedure?
•
Is it useful to count failures due to customer request?
•
Is the expected wear out of a bearing outer race such as in a generator considered a failure if it failed prematurely (McMahon, et al., 2006)?
•
Were recurring failures counted?
•
Are failures caused during installation counted as failures?
•
Are failures of certain components of the system excluded?
•
If a cascading failure occurs, which brings subsequent errors, each system is counted as a failure.
The de facto definition of failure used in the wind industry especially in the SCS SCADA was followed to compute MTBF.
Three straightforward definitions are: Type I:
The WTG/component has a defect or failure that prevents it from being put into operation.
Type II:
The termination of the ability of the WTG as a whole to perform its required function
Type III:
The termination of the ability of any individual component to perform its required function but not the termination of the ability of the WTG as a whole to perform.
In addition to these definition(s) chosen, it is imperative to state that human causes of failure are not included. Human errors have been excluded in the MTBF calculation; this is because there are many ways in which human error can result in failure (IEC, 1995) .These failures as included represent all things the O&M has control over. 3.13.4 Downtime ‘Start’ and ‘End’ Time Sufficient time has be allowed between the end of the sample period and the
Annual failure rate (AFR) calculation to allow time for WTG with reported failures
- 47 to have a comprehensive information of the WTG for the year. The SCADA determines the type of failure, while the repair validates the end time.
3.13.5 Computing the Annual Failure Rate
The annual failure rate was computed to illustrate the expected number of failures in one calendar year of a particular WTG. The first step in calculating this number was to “annualize” the failure data. This is done by multiplying the number of failures in the sample period by the number of sample periods per year. The second step was to determine the ratio of failures to the entire population (Miller, 2007pp557; Lamarre, 1998pp376).
52 weeks per year ) Number of weeks in sample period ------6.3 Number of units in population
Failures in the sample period x ( AFR =
This statement makes the following 2 assumptions: 1. The WTGs operated 24 hours a day, 365 days a year 2. All WTGs in the population begin operation at the same time though their commissioning dates are different but still within the sample there are still relevant number of WTGs which commenced on the same year. So, even though this formula could be used for any WTG, it is more relevant for WTG that are continuously operating and this is the main criteria for subjecting them to annualized calculation to bring them to the same level of performance in terms of duration equivalent.
3.13.6 Converting AFR to MTBF Converting AFR to MTBF (in hours) is the easiest of all the steps but is
perhaps the most frequently misinterpreted. Converting AFR to MTBF is valid only under the constant failure rate assumption which we have earlier assumed due to the nature of the WTGs.
- 48 The formula is as shown below in equation 6.4: MTBF =
Hours in a year 8760 = -------------------------------------------------------6.4 AFR AFR
3.13.6.1 Variables Affecting AFR Often times, MTBF values are obtained without any underlying data to back
them up. As mentioned previously, when looking at MTBF figures (or AFR figures) it is important to understand the underlying assumptions and variables used in this analysis, particularly the way failures are defined. When comparison is done without this understanding, the risk of a biased comparison becomes high and variations of 500% or more should be expected. This can ultimately lead to unnecessary business expense and even unexpected downtime. In general, the MTBF values between two or more WTGs should never be compared without an explicit definition of the variables, assumptions, and definitions of failure. Even if two MTBF values appear similar, there is still the risk of a biased comparison. Therefore it is imperative to look beyond the MTBF results and dissect and understand what goes into those values. 3.13.6.2 Constant Failure Rate Assumption For the field data measurement method of computing AFR and MTBF to be
valid, the WTGs being analyzed must assume a constant failure rate. This is generally an accepted assumption for electronic systems or components (Lamarre, 1998pp378). 3.13.6.3 Population Size It is very clear that the WTGs and their applications are similar, it is also
very important to note the method used to collect the field data collection process was same. Defining the population size (number of units commissioned) was the first critical variable in this study. If the volume of WTGs defined in the population is too small, the resultant MTBF estimate is likely to be useless. Although commissioning dates of the WTGs being compared may differ, the important consideration is the number of units in the population as per site, state and model.
- 49 3.13.6.3 Decision Criterion beyond MTBF While MTBF can be a useful decision tool for WTG specifications and
selection (when methods, variables, and assumptions are the same for all systems compared), it should never be the sole criterion. There are many other criteria used in this comparison in evaluating WTGs from multiple models. For instance: •
What is the Capacity and model of the WTG?
•
What type of generators are these WTGs using and in what environment?
•
What is the materials consumption pattern?
3.14
Analysis of Results
3.14.1 Operations and Maintenance Operations and maintenance is the elephant in the room of technology driven
manufacturing industry , improving preventive and predictive (Muhando, et al., 2010) maintenance to reduce failure rates often are the key issues .At present the majority of maintenance of a wind farm are conducted as a reaction to a failure which suppose not to be so rather increasing the capabilities and responsiveness of the maintenance organization to reducing the gap between the
scheduled and
unscheduled timelines should be the target. The development of higher power rated WTGs to produce more power will be necessary to capture the full potential of the wind if a little more achievements can be done on the reliability. 3.14.2 The WTGs Reliability Growth Pattern These analyses show differences between the reliability characteristics of the
selected components, gearboxes, generators and others over a period of one year though the machines have been commissioned at different dates. In the past, specific interest shown in the industry had been about differences in cost and performance achieved by different WTG models rather than reliability [59], as wind energy industry grows, the turbine performance becomes a very important issue to consider in the many areas of the inclement operations.
- 50 Implementation of condition monitoring for optimized operations will give a clear understanding on the variations in energy capture, and combining it with control to impact on reliability, improve efficiency and long sustainability of wind energy. The information gathered during successive condition monitoring (Ault et al., 2007pp.66) can help define preventive and preventive maintenance in line with reliability. Reliability growth is the positive improvement of reliability parameters of WTG over time (hence the term, growth) due to changes in WTG model or the manufacturing process through learning about the deficiencies of the model and taking action to eliminate or minimize the effect of these deficiencies. In planning reliability growth, the major role of this is to quantify the overall development effort so that the growth pattern can be evaluated relative to the objectives and resources of the particular program under consideration (MIL-HDBK-338B, 1998).
The initial prototypes for a complex system like the WTG with major technological advances will invariably have significant reliability and performance deficiencies that could not be foreseen in the early model stage. The ensuing system reliability and performance characteristics will depend on the number and effectiveness of these fixes. The ultimate goal of reliability growth is to meet the system reliability and performance requirements. Time variations of reliability present problems only in the early stages of development. Once any specific equipment model has been fully developed and used for some time in service, its reliability stabilizes at a relatively fixed value. However, during performance and initial application, deficiencies are often detected which require model changes to improve reliability. These changes are often the source of the problem of underperformance when the other technical issues were not observed and improved alongside. They complicate the preparation of statistically valid analysis of equipment performance. They introduce conflict between the original concept and the estimated reliability.
- 51 4.0
DISCUSSION AND CONCLUSION
4.1 Direct Drives and Geared Systems It has been argued previously in many publications (El-Fouly, et al., 2008pp166)
that geared turbines are inherently less reliable, and it has been seen simply to do with the gearbox (Li & Chen, 2008pp125; Polinder, et al., 2006pp.734) and in fact many WTG failures have been down to other thing like electrical systems as shown from the analysis. And because gearless technology is low-maintenance does not maximize returns rather a probable forecast of the system (Dubois, et al., 2009pp.3) predictive maintenance routine is what is needed. The failure intensities for WTGs were relatively low (Spinato, et al., 2009pp354) in terms of gearbox in our case. Material choice and design are high on the list of what can be done to improve future WTGs. Advances in materials for electrical components, blades, and gearboxes will incrementally improve the performance of WTG over time. Coupled with solving grid (Kasmas & Papathanassiou, 2008pp105) and transmission issues could contribute to reducing the failure rate. Voltage rise on rural 11kV networks is often a constraining factor for the installation of WTGs (Thomson & Infield, 2007pp36) and it is important to consider as transmission line expansions unlock new wind areas, more low to medium wind sites, the evolution of advanced technologies that maximize the energy production at low wind speed conditions. This would mean WTG further develop aspects such as wind forecasting, energy storage (Papaefthimiou, et al., 2009p.296) and smart control system in order to achieve grid stability (Haesen, et al., 2007pp29).In the future, the algorithm will be applied to planning in MV grids to improve reliability concerning dispatchable units, long term planning issues and grid stability.
- 52 4.2
Summary While carrying out these comparisons, MTBF often was a key decision
criterion. The failure mode and effect analysis is most commonly used and well qualitative reliability method in the area of reliability methodology. It is dynamic preventative reliability method used in the modification of components. The overall aim was to analyze and modify components in the light of experience to achieve an optimum criterion of reliability assessment. However, much care was taken when putting these values side by side. First, the method of estimating the MTBF values was the same. In addition, many variables and assumptions were used during the process of collecting and analyzing field data and each can have a significant impact on the result. A fair comparison of MTBF was possible when these variables and assumptions were observed. With the foundation provided in this study, MTBF can now be more fairly compared and reliability of the Suzlon WTGs evenly compared because similar assumptions and variables were used, and the definitions of failure were the same, hence a reasonable degree of confidence in the comparison was achieved. But it should be noted when MTBF is very low, the WTG is entering growth testing too soon. Another important factor to consider is to have a tradable regulated wind farm power presumes that there will be a reliable forecast predictions of low turbulence intensity (measured by variance) in the one hour period ahead (Li, et. al., 2009, pp188). Another potential added value can come from providing damping power to the AC grid as ancillary service. Nevertheless, it has been severally reported in many research results (Chiang, et. al., 2010, pp.87), that the grid interface of the WTG/PV hybrid system equipped with battery storage (Brunetto and Tina, 2007,pp223) unit can greatly improve the system reliability and operating flexibility. This may also be what SEL can be looking at in other to preserve the expected lifetime of the WT, because WTG/PV hybrid systems suppresses (Datta, et. al., 2010, pp.156) rapid changes in the output power of
single source such as WT system. Based on field data
measurement method, it has been shown that many parameters impact the system reliability (Bhuiyan and Yazdani, 2010, pp. 213). Among those, the grid instability and the wind strength plays a significant role; the impact of all other parameters become negligible seasonally. However, the component rating (Jayanti, et. al.
- 53 pp.133) and expected WTG reliability determined by proposed method should be viewed as reasonable estimates when a wind power system is planned; they may need to be refined for an installed windfarm.
- 54 5.0
1
REFERENCES
Ummels, B. C., Pelgrum, E., Gibescu, M. & Kling,
W. L., 2009.
Comparison of Integration Solutions for Wind Power in the Netherlands. IET Renew. Power Gener. Vol. 3, Issue 3, pp.279-292[1] 2
Papaefthimiou, S., Karamanou, E., Papathanassiou, S., & Papadopoulos, M.2009.
Operating Policies For Wind-Pumped Storage Hybrid Power
Station In Island Grids.IET Renew. Power Gener. Vol. 3, Issue 3, pp.293307. 3
Kornelakis, A. & Koutroulis, E., 2009. Methodology for the Design Optimisation & the Economic Analysis of Grid-Connected Photovoltaic Systems. IET Renew. Power Gener. Vol. 3, Issue 4, pp.476-492.[2]
4
Piegari, L. & Rizzo, R. 2010. Adaptive Perturb & Observe Algorithm for Photovoltaic Maximum Power Point Tracking. IET Renew. Power Gener. Vol. 4, Issue. 4, pp317-328.
5
Alarcon-Rodriguez, A, Haesen, E., Ault,G., Driesen, J., & Belmans, R. , 2009. Multi-Objective Planning Framework for Stochastic & Controllable Distributed Energy Resources. IET Renew. Power Gener. Vol. 3, Issue 2, pp.227-238
6
Brendan, F. et al., 2007. Wind Power Integration, Connection & system operational aspects. Published by The Institution of Engineering & Technology, London, United Kingdom
7
Hajizadeh , A & Golkar, M.A., 2009. Fuzzy Neural Control of A Hybrid Fuel Cell / Battery Distributed Power Generation IET Renew. Power Gener. Vol. 3, Issue 4, pp 402-414.
8
Karki, R & Billinton, R; Cost-effective wind energy utilization for reliable power supply, IEEE Transactions on Energy Conversion 19 (2) (June 2004), pp. 435–440
9
Kibble, TWB; Classical Mechanics 2nd edition, McGraw-HILL Book Company (UK) Limited.
10
Editorial, 2008. Selected Papers from the European Wind Energy Conference (EWEC), IET Renew. Power Gener. Vol. 2, No. 1, pp. 1-2
- 55 11
Morales, A., Robe, X., Sala M., Prats , P., Aguerri C., 2008. Advanced Grid Requirement for the Integration of Wind Farms into the Spanish Transmission System. IET Renew. Power Gener. Vol. 2, No. 1, pp.47-59
12
Conroy, J.F. & Watson, R.2007. Low-Voltage Ride- Through of a Full Converter Wind Turbine with Permanent Magnet Generator. IET Renew. Power Gener. Vol. 1, No. 3, pp.182-189.
13
Poller, M., 2008. Grid Compatibility of Wind Generators with HdyroDynamically Controlled Gearbox with German Grid Codes. DigSilient GmgH.
14
Teninge, A., Jecu, C., Roye, D., Bacha, S., Duval, J. , & Belhomme, R., 2009.Contribution to Frequency Control through Wind Turbine inertial Energy Storage .IET Renew. Power Gener. Vol. 3, Issue 3, pp.358-370.
15
Katsigiannis, Y. A., Georgilakis, P.S., Tsinarakis, G.J., 2008. Introducing a Coloured Fluid Stochastic Petri Net-Based Methodology for Reliability & Performance Evaluation of Small Isolated Power System Including Wind Turbines. IET Renew. Power Gener. Vol. 2, No. 2, pp.75-88.
16
Jupe, S. C. E., & Taylor, P.C, 2009. Distributed Generation Output For Network Power Flow Management. IET Renew. Power Gener. Vol. 3, Issue 4, pp 371-386.
17
McMillan, D.
& Ault, G.W., 2008. Condition Monitoring Benefit for
Onshore Wind Turbines: Sensitivity to Operational Parameters. IET Renew. Power Gener. Vol. 2, No. 1, pp.60-72. 18
Tavner, P.J.; Review of condition monitoring of rotating electrical machines,
IET Electrical Applications 2 (4) (2008), pp. 215–247. 19
Yang, W., Tavner, P.J. & Wilkinson, M.R, 2009. Condition Monitoring & Fault Diagnosis of A Wind Turbine Synchronous Generator Drive Train
.
IET Renew. Power Gener. Vol. 3, No. 1, pp1-11. 20
Hameed , Z., Hong , Y.S., Cho, Y.M., Ahn, S.H., Song, C.K., 2009. Condition Monitoring & Fault Detection Of Wind Turbines & Related Algorithms: A Review. ScienceDirect; Renewable & sustainable energy reviews 13, pp1-39.
- 56 21
Quinonez-Varela, G., Ault, G.W., Anaya-Lara, O., & McDonald, J.R., 2007. Electrical Collector System Options Large Offshore Wind Farms. IET Renew. Power Gener. Vol. 1, No. 2, pp.107-114.
22
Pearmine, R. ,Song, Y.H., & Chebbo, A., 2007. Influence of Wind Turbine behaviour on the Primary Frequency Control of the British Grid IET Renew. Power Gener. Vol. 1, No. 2, pp.142-150.
23
Lee, T.-Y, & Chen, C.-L, 2009. Wind-Photovoltaic Capacity Coordination for a Time-of-Use Rate Industrial User. IET Renew. Power Gener. Vol. 3, Issue 2, pp152-167.
24
Caramia, et al., 2007. Probabilistic Three-Phase Load Flow for Unbalanced Electrical Distribution Systems with Wind farms. IET Renew. Power Gener. Vol. 1, No. 2, pp.115-122
25
Massoud, A.M. , Ahmed, K.H., Finney, S.J. & Williams, B.W. 2009. Harmonic Distortion-Based Island Detection Technique For Inverter- Based Distributed Generation . IET Renew. Power Gener. Vol. 3, Issue 4, pp 493507.
26
Burton , T, Sharpe, D, Jenkins N, Bossanyi, E, 2001. “Wind Energy
Handbook” John 27
Wiley & Sons Ltd,pp377-468
Chen, S.-S., Wang , L. , Lee, W.J., & Chen, Z., 2009.Power Flow Control & Damping Enhancement Of A Large Wind Farm Using A Superconducting Magnetic Energy Storage Unit.IET Renew. Power Gener. Vol. 3, No. 1, pp.23-38
28
Aguglia D., Viarouge, P., Wamkeue R., & Cros, J, 2008.Analytical Determination of Steady-State Converter Control Laws for Wind Turbines Equipped with Double Fed Induction Generators. IET Renew. Power Gener. Vol. 2, No. 1, pp.16-25
29
Li, H. & Chen, Z., 2008. Overview of Different Wind Generator Systems & Their Comparisons IET Renew. Power Gener. Vol. 2, No. 2, pp.123-138
30
Wu, J.C, 2009. AC /DC Power Conversion Interface For Self-Excited Induction Generator. IET Renew. Power Gener. Vol. 3, Issue 2, pp144-151.
- 57 31
Li,R., Bozhko, S. , Asher, G.M., Clare, J.C. , Yao,L, & Sasse,C.2006. Grid Frequency Control Design for Offshore Wind Farms with Naturally Commutated HVDC Link Connection. IEEE ISIE, vol. 5,pp1595 – 1600.
32
Majumder, et al., 2009. Load Sharing & Power Quality Enhance Operation Of A Distributed Micro grid. IET Renew. Power Gener. Vol. 3, Issue 2, pp109-119.
33
Zhang, S., Tseng, K.J., & Choi, S.S., 2009.Statistical Voltage Quality Assessment Method For With Wind Power Generation.IET Renew. Power Gener. Vol. 4, Issue 1, pp 43-54
34
Zhao, et al., 2009. Load Flow Analysis For Variable Speed Offshore Wind Farms. IET Renew. Power Gener. Vol. 3, Issue 2, pp120-132.
35
Siemes, P.S, Haubrich, H.J, Vennegeerts, H., Ohrem, S., 2008. Concepts for the Improved Integration of Wind Power into the German Interconnected System. IET Renew. Power Gener. Vol. 2, No. 1, pp.26-33.
36
Cipcigan, L.M., & Taylor, P.C.2007.Investigation of the Reverse Power Flow Requirements of High Penetrations of Small-Scale Embedded Generation. IET Renew. Power Gener. Vol. 1, No. 3, pp.160-166.
37
Bowtell, L. & Ahfock, A., 2010. Direct Current Offset Controller For Transformerless Single-Phase Photovoltaic Grid-Connected Inverters. IET Renew. Power Gener. Vol. 4, Issue 5, pp.428- 437.
38
Ma, T.T., 2010. Novel Voltage Stability Constrained Positive Feedback AntiIslanding Algorithms for the Inverter-Based Distributed Generator Systems. IET Renew. Power Gener. Vol. 4, No. 2, pp176-185
39
Zhao, M., Chen, Z., & Blaabjerg, F., 2009. Optimisation Of Electrical System For Offshore Wind Farms Via Genetic Algorithm. IET Renew. Power Gener. Vol. 3, Issue 2, pp205-216
40
Chondroginnis, S. & Barnes, M. , 2008. Stability of Doubly-Fed Induction Generator Under Stator Voltage Orientated Vector Control
IET Renew.
Power Gener. Vol. 2, No. 3, Pp.170-180. 41
Bleijs, J.A.M. 2007. Wind Turbine Dynamic Response-Difference Between Connection to Large Utility Network & Isolated Diesel Micro-Grid. Renew. Power Gener. Vol. 1, No. 2, pp.95-106.
IET
- 58 42
Ramtharan, G., Ekanayaka , J.B. & Jenkins, N., 2007. Frequency Support from Doubly Fed Induction Generator Wind Turbines. IET Renew. Power Gener. Vol. 1, No. 1, pp.3-9.
43
Yang, T.C. , 2008.Initial Study of Using Rechargeable Batteries in Wind Power Generation with Variable Speed Induction Generators .IET Renew. Power Gener. Vol. 2, No. 2, pp.89-101.
44
Kasmas, N.A & Papathanassiou, S.A., 2008. Evaluation Of The Voltage Change Factor Ku For Dg Equipped With Synchronous Generators. IET Renew. Power Gener. Vol. 2, No. 2, pp.102-112
45
Ma, H.T. &
Chowdhury, B. H., 2010. Working Towards Frequency
Regulation with Wind Plants: Combines Control Approaches. IET Renew. Power Gener. Vol. 4, Issue. 4, pp308-316 46
Stannard, N., Bumby, J.R., Taylor, P., & Cipcigan, L.M., 2007. AC & DC Aggregation Effects of Small-Scale Wind Generators. IET Renew. Power Gener. Vol. 1, No. 2, pp.123-130.
47
Sae-Kok , W. , Grant,
D.M. & Williams,
B.W.,2010. System
Reconfiguration Under Open-Switch Fault In A Doubly Fed Induction Machine. IET Renew. Power Gener. Vol. 4, Issue 5, pp 458-470 48
Ramtharan, G., Arulampalam , A., Ekanayake, J.B. , Hughes,
F.M. &
Jenkins, N., 2009.Fault Ride Through Of Fully Rated Converter Wind Turbines With AC & DC Transmission Systems
. IET Renew. Power
Gener. Vol. 3, Issue 4, pp 426-438 49
Hazra, S. & Sensarma, P.S., 2010. Self-Excitation & Control of a StandAlone Wind Energy Conversion System. IET Renew. Power Gener. Vol. 4, Issue. 4, pp383-393.
50
Delfino, F., Procopio, R., Rossi,
M. & Ronda,
G., 2010. Integration of
Large-Size Photovoltaic Systems into the Distribution Grids: A P-Q Chart Approach To Assess Reactive Support Capability. IET Renew. Power Gener. Vol. 4, Issue. 4, pp329-340 52
Bevrani , H. , Ghosh, A. & Ledwich, G., 2010. Renewable Energy Source & Frequency Regulation: Survey & New Perspectives Gener. Vol. 4, Issue 5, pp.438-457.
IET Renew. Power
- 59 53
Chiang, H.C., Ma, T.T., Cheng,
Y.H., Chang, J.M. & Chang, W.N,
2010.Design & Implementation of a Hybrid Regenerative Power System Combining Grid-Tie & Uninterruptible Power Supply Functions. IET Renew. Power Gener. Vol. 4, Issue 1, pp 85-99. 53
Chondrogiannis, S., & Barnes, M., 2008. Specification Of Rotor Side Voltage Source Inverter of a Doubly-Fed Induction Generator for Achieving Ride- Through Capability. IET Renew. Power Gener. Vol. 2, No. 3, Pp.139150
54
Zeineldin, H.H., El-Foul, T. H. M., El-Saadany, E.F., & Salama, M.M.A., 2009. Impact of Wind Farm Integration on Electricity Market Prices. IET Renew. Power Gener. Vol. 3, No. 1, pp84-95.
55
Spinato, F., Tavner, P. J., Van Bussel, G.J.W. & Koutoulakos, E., 2009, Reliability of wind Turbine subassemblies. IET Renew. Power Gener. Vol. 3, Issue 4, pp 387-401.
56
Billinton, R. & Allan, R.N., 1996.Reliability evaluation of engineering
systems: concepts & techniques (2nd ed.), Plenum Press 57
Blumenthal, S., Greenwood, J. A. & Herbach, L. H., 1976. A Comparison of the Bad As Old and Superimposed Renewal Models. Management Science, vol. 23, No. 3, pp. 280-285
58
Pudjianto, et al., 2007. Virtual Power Plant & System Integration of Distribution Energy Resources, IET Renew. Power Gener. Vol. 1, No. 1, pp.10-16.
59
Ault, et al., 2007.Calculation of Economic Transmission Connection Capacity for Wind Power Generation. IET Renew. Power Gener. Vol. 1, No. 1, pp.61-69.
60
MIL-HDBK-338B, Electronic Reliability Design Handbook, October 1, 1998
61
Lamarre, B. G., Mathematical Modelling, Reliability and Maintainability of Electronic Systems, Edited by: J.E. Arsenault and J.A. Roberts,Computer Science Press, p372 - 373.
62
Billinton, R & Chowdhury, AA; Incorporation of wind energy conversion systems in conventional generating capacity adequacy assessment, IEE
- 60 -
Proceedings Generation, Transmission & Distribution, Part C 139 (1) (Jan. 1992), pp. 47–56. 63
Billinton, R & Hua Chen, Assessment of risk-based capacity benefit factors associated with wind energy conversion systems, IEEE Transactions on
Power Systems 13 (3) (Aug. 1998), pp. 1191–1196 65
Caralis, G. & Zervos, A., 2007. Analysis of the Combined Use of Wind & Pumped Storage System in Autonomous Greek Islands. IET Renew. Power Gener. Vol. 1, No. 1, pp.49-60.
66
Caralis, G., & Zervos, A., 2010. Value of Wind Energy on the Reliability of Autonomous Power Systems. IET Renew. Power Gener. Vol. 4, No. 2, pp186-197.
67
Bianchi, F.D., De Battista, H. & Mantz , R.J, 2008. Optimal Gain-Scheduled Control of Fixed-Speed Active Stall Wind Turbines.IET Renew. Power Gener. Vol. 2, No. 4, pp228-238
68
Di Fazio, AR & Russo, M; “Wind farm modelling for reliability assessment,” IET Renew. Power Gener., 2008, Vol.2, No. 4, pp. 239-248
69
Mcdonald, A.S., Mueller, M.A., & Polinder, H., 2008. Structural Mass in Direct-Drive Permanent Magnet Electrical Generators .IET Renew. Power Gener. Vol. 2, No. 1, pp.3-15.
70
Muyeen, et al., 2007. Comparative Study on Transient Stability Analysis of Wind Turbine Generator System Using Different Drive Trains Models. IET Renew. Power Gener. Vol. 1, No. 2, pp.131-141.
71
Elghali Ben, S.E., Benbouzid, M.E.H. & Charpentier, J.F., 2010.Modelling & Control of a Marine Current Turbine-Driven Double Fed Induction Generator. IET Renew. Power Gener. Vol. 4, Issue 1, pp.1-11.
72
Geng, H. , & Yang, G., 2009.Robust Pitch Controller For Output Power Level Of Variable-Speed Variable-Pitch Wind Turbine Generator Systems. IET Renew. Power Gener. Vol. 3, Issue 2, pp168-179.
73
Manwell, J.F., et al, 2002. Wind Energy Explained. John Wiley & Sons Ltd
74
Tsili, M., &
Papathanassiou, S. 2009. A Review Of Grid Code Technical
Requirements for Wind Farms. IET Renew. Power Gener. Vol. 3, Issue 3, pp.308-332
- 61 75
Khan, M. J.
Iqbal, M. T. & Quaicoe, J.E., 2010. Dynamics of a Vertical
Axis Hydrokinetic Energy Conversion System with a Rectifier Coupled Multi –Pole Permanent Generator.IET Renew. Power Gener. Vol. 4, No. 2, pp116-127. 76
Foster, S., Xu, L and Fox, B. 2010. Coordinated Reactive Power Control for Facilitating Fault Ride through of Doubly Fed Induction Generator and Fixed Speed Induction Generator-Based Wind Farms. IET Renew. Power Gener. Vol. 4, No. 2, pp128-138.
77
Wang, Y.J & Hsu, P.C., 2010. Analytical Modelling Of Partial Shading & Different Orientation Of Photovoltaic Modules. IET Renew. Power Gener. Vol. 4, Issue 3, pp 272-282.
78
Conroy, J., & Watson,
R., 2009. Aggregate Modelling Of Wind Farm
Containing Full- Converter Wind Turbine Generators with Permanent Magnet Synchronous Machines: Transient Stability Studies. IET Renew. Power Gener. Vol. 3, No. 1, pp.39-52 79
Petersson A. Analysis, modeling & control of doubly-fed induction generators for wind turbines, PhD thesis. Chalmers University of technology, Gutenberg, Sweden; 2005.
80
Prasad, R.D., Bansal, R.C. & Sauturaga, M., 2009. Some of the Design & Methodology Considerations in Wind Resource Assessment. IET Renew. Power Gener. Vol. 3, No. 1, pp.53-64.
81
Sayas, FC & Allan, RN; Generation availability assessment of wind farms,
IEE Proceedings Generation, Transmission & Distribution, Part C 143 (5) (Sept. 1996), pp. 507–518. 82
Chen, et al., 2007. Stability Improvement of Induction Generation-Based Wind Turbine Systems.IET Renew. Power Gener. Vol. 1, No. 1, pp. 81-93
84
Arabian, H, Hoseynabadi, HO & Tavner, PJ. Wind turbine productivity considering electrical subassembly reliability, Science direct, Renewable Energy Volume 35, Issue 1, January 2010, Pages 190-197
84
Hamid AT & Gerald, BK; Handbook of Electric Motors: Reliability, Second
Edition, Revised & Expanded; CRC Press 2004, pp.655-691
- 62 85
Mostafaeipour, A & Abarghooei, H; Harnessing wind energy at Manjil area located in north of Iran, Renewable & Sustainable Energy Reviews 12 (2008), pp. 1758–1766 Elsevier.
86
Shek, et al., 2007. Reaction Force Control of a Linear Electrical Generator for Direct Wave Energy Conversion, IET Renew. Power Gener. Vol. 1, No. 1, pp.17-24.
87
Liu, X., McSwiggan, D., Litter, T. B., & Kennedy, J., 2010. MeasurementBased Method For Wind Farm Power System Oscillations Monitoring.IET Renew. Power Gener. Vol. 4, No. 2, pp198-209.
88
Faried, S.O., Billinton, R., & Aborsehaid, S., 2010. Probabilistic Evaluation of Transient Stability of a Power System Incorporating Wind Farms. IET Renew. Power Gener. Vol. 4, Issue. 4, pp299-301.
88
Tabesh, A., & Iravani, R, 2008. Small-Signal Model & Dynamic Analysis of Variable Speed Induction Wind Farms. IET Renew. Power Gener. Vol. 2, No. 4, Pp215-227
89
Klobasa, M. 2010. Analysis of Demand Response & Wind Integration in Germany’s Electricity Market.IET Renew. Power Gener. Vol. 4, Issue 1, pp 55-63
90
Boehme, T., Harrison, G.P., & Wallace, A.R. ,2010. Assessment of Distribution Network Limits for Non-Firm Connection of Renewable Generation .IET Renew. Power Gener. Vol. 4, Issue 1, pp.64-74.
91
Melicio, R., Mendes, V .M .F, & Catalao,
J. P .S.,2010 Harmonic
Assessment of Variable-Speed Wind Turbines Considering a Converter Control Malfunction. IET Renew. Power Gener. Vol. 4, No. 2, pp139-152. 92
Kasem, A.H., El-Saadany, E.F., El-Tamaly, H.H. & Wahab, M.A.A, 2008. An Improved Fault Ride-Through Strategy for Doubly Fed Induction Generator-Based Wind Turbines. IET Renew. Power Gener. Vol. 2, No. 4, pp201-214.
93
Helander, A. Holttinen, H., & Paatero. J. 2010. Impact of Wind Power on the Power System Imbalances in Finland &. IET Renew. Power Gener. Vol. 4, Issue 1, pp 75-84.
- 63 94
Trichakis, P., Taylor, P.C, Lyons, P.F., & Hair, R., 2008. Predicting The Technical Impacts Of High Levels Of Small-Scale Embedded Generators On Low-Voltage Networks.IET Renew. Power Gener. Vol. 2, No. 4, Pp249-262
95
Bakirtzis, AGA; Probabilistic method for the evaluation of the reliability of standalone wind energy systems, IEEE Transactions on Energy Conversion 7 (1) (March 1992), pp. 99–107.
96
Gao, Y & Billinton,R., 2009. Adequacy Assessment of Generating Systems Containing Wind Power Considering Wind Speed Correlation. IET Renew. Power Gener. Vol. 3, Issue 2, pp.217-226
96
IEEE 90 – Institute of Electrical & Electronics Engineers, IEEE Standard Computer Dictionary: A Compilation of IEEE Standard Computer Glossaries. New York, NY: 1990
98
Miller, I., 2007. Probability and Statistics for Engineers, Prentice Hall Inc., Englewood Cliffs, pp. 552-563.
99
Military Handbook, "Reliability Growth Management" MIL-HDBK-189, 13 February, 1981.
100
Hahn, B., Durstewitz, M., & Rohrig, K., (2007).Reliability of Wind Turbines. Springer,pp324 - 332
101
IEC International Standard, Reliability Growth- Statistical test and estimation methods, IEC 1164 International Electrotechnical Commission, 1995
102
El-Fouly, T.H.M., Zeineldin, H.H., El-Saadany, E.F, 2008. Impact of Wind Generation Control Strategies, Penetration Level & Installation On Electricity Market Prices. IET Renew. Power Gener. Vol. 2, No. 3, Pp.162169.
103
Thomson, M. & Infield, D.G., 2007. Impact of Widespread Photovoltaic’s Generation on Distribution Systems. IET Renew. Power Gener. Vol. 1, No. 1, pp.33-40
105
Haesen, et al., 2007. Robust Planning Methodology for Integration of Stochastic Generators in Distribution Grids. IET Renew. Power Gener. Vol. 1, No. 1, pp.25-32.
- 64 106
Dysko, et al.,2007. Reducing Unnecessary Disconnection of Renewable Generation from the Power System. IET Renew. Power Gener. Vol. 1, No. 1, pp.41-48.
107
Mathiesen, B. V., & Lund,
H., 2009. Comparative Analyses Of Seven
Technologies To Facilitate The Integration Of Fluctuating Renewable Energy Sources. IET Renew. Power Gener. Vol. 3, Issue 2, pp190-204 108
Hassanain, N.E.A.M., & Fletcher, J.E. , 2010. Steady-State Performance Assessment of Three-and Five-Phase Permanent Magnet Generators Connected To A Diode Bridge Rectifier Under Open-Circuit Faults. IET Renew. Power Gener. Vol. 4, Issue 5, Pp 420-427.
109
Wang, Y., Xu, L. & Williams, B.W., 2009. Compensation Of Network Voltage Unbalance Using Doubly Fed Induction Generator-Based Wind Farms. IET Renew. Power Gener. Vol. 3, No. 1, pp12-22
110
Guo, X. Q., & Wu, W. Y., 2010. Improvement Current Regulation of ThreePhase Grid-Connected Voltage-Source Inverters for Distributed Generation Systems. IET Renew. Power Gener. Vol. 4, No. 2, pp101-115
111
El Moursi, M. Joos , G. & Abbey, C., 2007. High-Performance Voltage Control Scheme for Wind Park Integration. IET Renew. Power Gener. Vol. 1, No 3, pp.151-159.
112
Emblemsvåg, J., 2003. Life-cycle costing using activity-based costing &
Monte Carlo methods to manage future costs & risks. John Wiley & Sons, Inc., Hoboken, New Jersey, published simultaneously in Canada. 112
Scanff, E. et al., 2007. Life cycle cost impact of using prognostic health management (PHM) for Helicopter Avionics. Microelectronics Reliability, Vol. 47, No.12, pp. 1857 – 1864.
113
El-Khattam, W. & Sidhu, T. S., 2009. Resolving the Impact Of Distributed Renewable Generation On Directional Overcurrent Relay Coordination: A Case Study. IET Renew. Power Gener. Vol. 3, Issue 4, pp 415-425.
114
Avelar, V., 2010. Avoiding AC Capacitor Failures in Large UPS systems. APC by Schneider Electric, White paper 60, pg1-11
- 65 115
Diego D.M., Ricardo Q.M., & Felix A.F., 2007. Interaction between PEM Fuel Cells & Converters for AC Integration.IEEE PowerEng,1-4244-08954/07,pp359-361,pp19-21.
116
Rubira, S. D. & McCulloch, M.D., 2000. Control method Comparison of Doubly Fed Wind Generators Connected to the Grid by asymmetric Transmission Lines. IEEE Transactions on Industry Applications, Vol. 36. No. 4 , July/August, 2000,pp.986 – 991.
117
IEEE Std. 929-2000 – Institute of Electrical & Electronics Engineers, IEEE Standard recommended practice for utility interface of photovoltaic(PV) systems
118
Muyeen, S.M, Takahashi, R.,
Murata, T., Tamura,
J., Ali,
M.H.,
Matsumura, Y., Kuwayama, A. & Matsumoto, T., 2009. Low Voltage Ride through Capability Enhancement of Wind Turbine Generator System during Network Disturbance. IET Renew. Power Gener. Vol. 3, No. 1, Pp65-74 119
Rahim, Y., Fletcher, H.A & Hassanain,
N.E.A.M. , 2010. Performance
Analysis Of Salient-Pole Self –Excited Reluctance Generators Using A Simplifier Model. IET Renew. Power Gener. Vol. 4, Issue 3, pp 253-260. 120
Syafaruddin, Karatepe, E. & Hiyama, T., 2009. Artificial Neural NetworkPolar Coordinated Fuzzy Controller Based Maximum Power Point Tracking Control under Partially Shaded Conditions IET Renew. Power Gener. Vol. 3, Issue 2, pp.239-253
121
Hansen, A. D. & Michalke,
G. 2009. Multi-Pole Permanent Magnet
Synchronous Generator Wind Turbines’ Grid Support Capability in Uninterrupted Operation during Grid Faults. IET Renew. Power Gener. Vol. 3, Issue 3, pp.333-348 122
Ronner, B., Maibach, P. & Thurnherr, T., 2009. Operational Experiences of Statcoms for Wind Parks. IET Renew. Power Gener. Vol. 3, Issue 3, pp.349357
123
Todeschini G. & Emanuel, A.E., 2010. Wind Energy Conversion System as Active Filter: Design & Comparison of Three Control Methods.IET Renew. Power Gener. Vol. 4, Issue. 4, pp341-353
- 66 124
Karatepe,E.,
H Iyama, Syafaruddin
T., 2010. Simple & High-Efficiency
Photovoltaic System under Non-Operating Conditions.IET Renew. Power Gener. Vol. 4, Issue. 4, pp354-368. 125
Roscoe, A. J., & Ault, G., 2010. Supporting High Penetrations of Renewable Generation Via Implementation of Real-Time Electricity Pricing & Demand Response.IET Renew. Power Gener. Vol. 4, Issue. 4, pp369-382.
126
Liu, F., Kang, Y., Zhang, Y. & Duan, S, 2010. Improved SMS Islanding Detection Method For Grid-Connected Converters. IET Renew. Power Gener. Vol. 4, Issue 1, pp 36-42.
127
Kasem, A.H, El-Saadany,
E.F., El-Tamaly,
H.H., Wahab, Mohamed
A.A.,2010. Power Ramp Rate Control & Flicker Mitigation For Directly Grid Connected Wind Turbines IET Renew. Power Gener. Vol. 4, Issue 3, pp.261-271. 128
Kanellos, F.D, & Hatzi, N.D., 2009. Control of Variable Speed Wind Turbines Equipped with Synchronous or Doubly Fed Induction Generators Supplying Islanded Power Systems.IET Renew. Power Gener. Vol. 3, No. 1, pp96-108
129
Tavner, PJ, Xiang, J. & Spinato, F; 2006. Reliability Analysis for Wind Turbines. Wiley Interscience wind energ. 2007; 10: 1-18.
130
Tavner PJ; Predicting the design life of high integrity rotating electrical machines. IEEE 9th International EMD Conference, Canterbury; 1999. p. 286–90.
131
Hicks, T.G. 4th ed., 2004. Standard handbook of engineering calculations. The Mcgraw-Hill Companies.
132
Baldor Electric Company, USA. Inverter-Driven Induction Motors Shaft & Bearing Current Solutions. Industry White paper. www.baldor.com [Assessed: March, 22, 2011]
133
Fletcher J., Judendorfer, T.
Mueller, M. Hassanain,
N., & Muhr, M,
2009.Electrical Issues Associated With Sea-Water Immersed Windings In Electrical Generators For Wave-& Tidal Current-Driven Power Generation .IET Renew. Power Gener. Vol. 3, Issue 2, pp. 254-264
- 67 134
Madureira, A.G. & Pecas Lopes, J.A. , 2009. Coordinated Voltage Support In Distribution Network with Distributed Generation & Microgrids. IET Renew. Power Gener. Vol. 3, Issue 4, pp 439-454
135
Keane, A., Zhou, Q. Bialek, J.W., O’ Malley, M, 2009. Planning & Operating Non-Firm Distributed.IET Renew. Power Gener. Vol. 3, Issue 4, pp 455-464
137
Rahimi, M. & Parniani, M.2010 Efficient Control Scheme of Wind Turbines with Doubly Fed Induction Generators for Low-Voltage Ride-Through Capability Enhancement IET Renew. Power Gener. Vol. 4, Issue 3, pp.242252.
138
Meibom,P., Weber., C. & Barth,R., 2009 Operational Costs Induced By Fluctuating Wind Power Production In Germany & Scandinavia. IET Renew. Power Gener. Vol. 3, No. 1, pp75-83.
138
Mendonca, A. , & Pecas Lopes, J.A.2009. Robust Tuning of Power System Stabilisers To Install In Wind Energy Conversion Systems
IET Renew.
Power Gener. Vol. 3, Issue 4, pp 465-475 139
De Moura, P.A., De Moura, A.A.F., De Moura, A.A.F. , 2008. Analysis of Injected Apparent Power & Flicker in a Distribution Network after Wind Power Plant Connection.IET Renew. Power Gener. Vol. 2, No. 2, pp.113122
141
Pal, B.C & Mei, F. 2008. Modelling Adequacy of the Doubly Fed Induction Generator for Small-Signal Stability Studies in Power Systems.IET Renew. Power Gener. Vol. 2, No. 3, Pp.181-190
142
Abbas, A.Y.M. & Fletcher, J.E., 2010. Synthetic Loading Applied To Linear Permanent Magnet Synchronous Machines. IET Renew. Power Gener. Vol. 4, Issue 3, pp 211-220.
144
Li, P., Keung, P.K. & Ooi,
B.-T., 2009. Development & Simulation of
Dynamic Control Strategies for Wind Farms .IET Renew. Power Gener. Vol. 3, Issue 2, pp180-189. 146
Brunetto C., & Tina, G. 2007. Optimal Hydrogen Storage Sizing for Wind Power Plants in Day ahead Electricity Market. IET Renew. Power Gener. Vol. 1, No. 4, pp.220-226.
- 68 147
Datta M., Senjyu, T, Yona, .A & Funabashi T., 2010. Minimal-Order Observer-Based Coordinated Control Method For Isolated Power Utility Connected Multiple Photovoltaic Systems To Reduce Frequency Deviations. IET Renew. Power Gener. Vol. 4, No. 2, Pp153-164.
148
Bhuiyan, F.A., & Yazdani , A. , 2010. Reliability Assessment of A Wind – Power System With Integrated Energy Storage. IET Renew. Power Gener. Vol. 4, Issue 3, pp 211-220.
149
Ameri M, Ghadiri M & Hosseini M; Recent advances in the implementation of wind energy in Iran. In: The 2nd joint international conference on “Sustainable Energy & Environment (SEE 2006)” Bangkok, Thailand; 21–23 November 2006.
150
Ascher, H. & Feingold, H., 1984. Repairable systems Reliability, Modeling,
inference, misconceptions & their causes.Marcel Dekker,Inc.Vol. 7. 151
Abduwahid,
et
al.,
2007.
Development
of
a
Dynamic
Control
Communication System for Hybrid Power Systems. IET Renew. Power Gener. Vol. 1, No. 1, pp.70-80. 152
Bertsche,B.,2008.Reliability
in
Automotive
&
Mechanical
Engineering.Springer – Verlag Berlin Heidelberg 153
Bagen, 2005. Reliability & Cost/Worth Evaluation of Generating Systems
Utilizing Wind & Solar Energy. Ph. D.: University of Saskatchewan, Saskatoon. 154
Crow, Dr. Larry H., "Reliability Growth Projection from Delayed Fixes," Proceedings, Annual Reliability & Maintainability Symposium, 1983, pp.8489.
155
Duane, J.T., "Learning Curve Approach to Reliability Monitoring," IEEE Transactions on Aerospace, Vol. 2, No. 2, 1964.
156
Dubois, M.R, Polinder, H., Ferreira, J.A.,2009. Comparison of Generator Topologies for Direct-Drive Wind Turbines.IET Renew. Power Gener. Vol. 3, Issue 2, pp.1-5.
157
Dukpa, A., Duggal, I., Venkatesh, B., & Chang, L., 2010. Optimal Participation & Risk Mitigation of Wind Generation in an Electricity Market.IET Renew. Power Gener. Vol. 4, No. 2, pp165-175.
- 69 158
Gavanidou, ES ,Bakirtzis, AG & P.S. Dokopoulos, A probabilistic method for the evaluation of the performance of wind–diesel energy systems, IEEE
Transactions on Energy Conversion 7 (3) (Sept. 1992), pp. 418–425 159
Hameeda, Z, Honga, YS ,Choa, YM Ahnb, SH & Songc CK; “Condition monitoring & fault detection of wind turbines & related algorithms: A review”, Science direct Renewable & Sustainable Energy Reviews 13 (2009) 1–39
160
Jayanti, N. G., Basu, M., Conlon, M. F., & Gaughan, K.2009. Rating Requirement of the Unified Power Quality Conditioner to Integrate the Fixed-Speed Induction Generation to The Grid. IET Renew. Power Gener. Vol. 3, Issue 2, pp133 – 143.
161
Jenkins, D.P, Fletcher, J. & Kane, D., 2008. Lifetime Prediction & Sizing of Lead-Acid Batteries for Micro generation Storage Application. IET Renew. Power Gener. Vol. 2, No. 3, pp191-200.
162
Jou, H.L., Chiang, W.J. & Wu, J.C.2007. Virtual Inductor-Based Islanding Detection Method for Grid-Connected Power Inverter of Distributed Power Generations System. IET Renew. Power Gener.Vol. 1, No. 3, pp.175-181
163
Jupe, S.C.E., Taylor,
P.C.& Michiorri, A., 2010. Coordinated Output
Control of Multiple Distributed Generation Schemes IET Renew. Power Gener. Vol. 4, Issue 3, Pp 283- 297. 164
Katsigiannis, Y.A. ,Georgilakis, P.S, Karapidakis, E.S., 2010. Multi objective Genetic Algorithm Solution to the Optimum Economic & Environmental Performance Problem of Small Autonomous Hybrid Power System with Renewables.IET Renew. Power Gener. Vol. 4, Issue 5, pp.409419.
165
Kennedy, J. Fox, B. & Morrow, D.J.2007. Distributed Generation As A Balancing Resource For Wind Generation IET Renew. Power Gener. Vol. 1, No 3, pp 167-174
166
Leonard, C., “MIL-HDBK-217: It’s Time To Rethink It”, Electronic Design, October 24, 1991
- 70 167
McMahon RA, Wang X, Abdi-Jalebi E, Tavner PJ, Roberts PC & Jagiela M; The BDFM as a generator in wind turbines, European Power Electronic Conference, Portoroz, Slovenia; August 2006.
168
Military Handbook, "Reliability Test Methods, Plans & Environments for Engineering, Development, Qualification & Production," MIL-HDBK-781A, 1 April 1996.
169
Mohammed, H & Nwankpa, CO; Stochastic analysis & simulation of gridconnected wind energy conversion system, IEEE Transactions on Energy
Conversion 15 (1) (March 2000), pp. 85–90 170
Muhando, E.B., Senjyu, T. , Uchida, K. Kinjo, H, & Funabshi, T.,2010. Stochastic Inequality Constrained Closed-Loop Model-Based Predictive Control of MW-Class Wind Generating System in the Electric Power Supply. IET Renew. Power Gener. Vol. 4, Issue 1, pp23-35.
171
Patel, H., & Agarwal, V.,.2009. Investigation into Performance of Photovoltaic –Based Active Filter Configurations & Their Control Schemes under Uniform & Non-Uniform Radiation Conditions. IET Renew. Power Gener. Vol. 4, Issue 1, pp.12-22.
172
Pecht, M.G., & Nash, F.R., 1994. Predicting the Reliability of Electronic Equipment, Proceedings of the IEEE, Vol. 82, No. 7
173
Polinder H, Van der Pijl FFA, De Vilder GJ, & Tavner PJ, September, 2006. Comparison of direct-drive & geared generator concepts for wind turbines. IEEE Transactions on Energy Conversion; 21(3):725–33.
174
Ummels, B.C, Pelgrum, E., Kling, W.L., 2008. Integration of Large-Scale Wind Power & Use of Energy Storage in the Netherlands’ Electricity Supply. IET Renew. Power Gener. Vol. 2, No. 1, pp.34-46.
175
Wang, Z. Sun, Y. , Li, G. & Ooi, B.T, 2010.Magnitude & Frequency Control of Grid-Connected Doubly Fed Induction Generator Based On Synchronized Model For Wind Power Generation .IET Renew. Power Gener. Vol. 4, Issue 3, Pp 232-241.
176
http://europa.eu/legislation_summaries/development/sectoral_ development_policies/r12 008_en.htm Assessed: 15-02-11
177
http://www.markov-model.com.Assessed: 20-03-11
- 71 6.0
ANNEXURE(S)
Original
Error Analysis of LBBE WTG
Number of Incidents
Turbine electrical Mech RpmFR1 CNT DiffStop Pitch system Pitch 3% Substation Elec Drive train Elec FB GearBoxValve ExtPowerSupply24VStopp Drive train Elec 1%FB GearOil Pump CurrentAsymmetry Conv2 0% 0% 1%
Pitch system Pitch EmergencyRun 38%
Grid Elec CurrentAsymmetry 48%
Pitch system PID Grid Rep Elec CurrentAsymmetry PowerLowerThanWindSpeed 1% 7%
Jonathan Okoronkwo
Determination of Reliability of Suzlon WTGs
Manual stop RS BottomServiceStop 1%
Original
Forecast of Maintenance Culture for Y MW Model
Regression Graph showing Present Lifetime of WTGs in Years vs. MTTR in Clusters 0.700
0.600
MTTR (Hrs)
0.500
0.400
MTTR 0.300
Power (MTTR)
0.200
0.100
0
0.5
1
1.5
2 Present Life Time in Years
Jonathan Okoronkwo
Modeling of Reliability of WTGs
2.5
3
3.5
4
Original
Components Failure Rate and their MTTR
6000
MTTR per Each System Component - X MW Model
Sum of MTTR (Hrs)
5000 4000 3000 2000 1000 0
System Components
Jonathan Okoronkwo
Modeling of Reliability of WTGs
Original
Component Failure Rate per State
25000
MTTR Vs Component For ALL States
Sum of MTTR in Hrs
20000
15000
Gujarat Karnataka
10000
Maharashtra Rajasthan Tamilnadu 5000
0
Components
Jonathan Okoronkwo
Modeling of Reliability of WTGs
Original
Scatter Plot of Present Lifetimes of WTGs in Years vs MTTR in Clusters
Scatter Graph showing Present Lifetime of WTGs in Years vs. MTTR in Clusters
All Models 0.700
0.600
MTTR (Hrs)
0.500
0.400
MTTR
0.300
0.200
0.100
0
1
2
3
4
Present Life Time in Years
Jonathan Okoronkwo
Modeling of Reliability of WTGs
5
6
7
