Conference Full Paper template

Palestine Polytechnic University Third International Conference on Energy and Environmental Protection in Sustainable Development ( ICEEP III), October 9-10, 2013, Hebron, West bank, State of Palestine Paper Code. No. ENE162

Comparison of Bus Voltage Configurations for a Residential Wind/PV/Battery Hybrid System Architectures

ABSTRACT This paper investigates three bus voltage configurations of a stand-alone hybrid power system based on photovoltaic generator (PVG), wind energy generator (WEG) and battery storage system (BSS). The configurations include high voltage DC bus (HV-DC), low voltage DC bus (LV-DC) and high voltage AC bus (HV-AC). Dynamic models are designed in MatLab/Simulink/SimPowerSys TM environment in order to study and compare the performance of each configuration. To this purpose some performance indexes are considered, such as: the global efficiency, fraction of energy driven from battery storage system to cover the load demand and fraction of energy delivered to battery storage system from renewable energy sources. The global efficiency of HVDC bus configuration is found the best while the global efficiency of LVDC bus configurations is the lowest. The fraction of energy required from battery energy system is the lowest in LVDC while the HVAC is the highest. The fraction of energy driven to battery energy system in LVDC bus configuration is the highest among the other configurations. Sensitivity of performance indexes to seasonal load variation is investigated using three patterns of load profiles.

ENE162-1


1

INTRODUCTION

The Energy decision makers worldwide are paying great attention for renewable energy sources to mitigate the future energy crisis resulting from the rapid depletion of fossil fuel resources and the accelerating concern over global warming. Photovoltaic energy and wind energy are the most examined and developed renewable energy sources all over the world, since they are clean, sustainable, generated on-site, pollution free and inexhaustible [1,2,3]. Neither Photovoltaic energy nor wind energy can supply separately a continuous energy due to seasonal and daily weather variations. Therefore, in order to adapt with those conditions in a stand-alone systems, PhotoVoltaic generator (PVG) and wind energy generator (WEG) are proposed in this paper. An efficient energy storage system (BSS) is exploited due to the overlap of the availability of the two primary sources. The stand-alone hybrid system plant discussed in this paper is composed of three main elements, namely: PVG, WEG, and BSS. These elements can be connected together according to different configurations and approaches in order to provide energy to load or storage system, such as: DC-coupling, AC-coupling, or even more sophisticated multiple bus coupling. The optimal selection of the internal structure of a stand-alone hybrid system is not an easy task as there is no suitable design tool for comparing different candidate structure. The optimum configuration is identified among different structures on the basis of environmental conditions and load diagrams through a procedure computing the Loss of Power Supply Probability index (LPSP) for standalone plants [4], and the Loss of Produced Power Probability index (LPPP) for grid connected plants [5]. In this paper a more effective comparison approach has been followed based on specific control strategies of each structure, real distribution of solar radiation, wind speed and load as well as taking into account the efficiency of each power converter. To do this dynamic models of each architecture are designed using MatLab/Simulink/SimPowerSys TM as well as different performance indexes are presented. 2

BUS VOLTAGE CONFIGURATIONS

In this paper three possible bus configurations have been selected to build a stand-alone generator with BSS, namely:   

High voltage DC bus configuration (HVDC). Low voltage DC bus configuration (LVDC). High voltage AC bus configuration (HVAC).

Each element in the above configurations are controlled by three independent control systems, also accounting for the Maximum Power Point Tracking (MPPT) [6]. Suitable mathematical models of all the elements of the system have been previously modelled and a detailed description of the models is beyond the objective of this paper [6]. Details about parameters of the main subsystems of the considered plant are given in Table 1.

ENE162-2


Table 2 PVG and WEG main data PVG Module model

Solyndra® - SL001-157

Module unit

157Wp at STC

Module open circuit voltage

92.5 V

Module number

3 × 4 = 12

Power rating

1.88kWp WEG

2.1

Model

TN-1.5 Nozzi Nord

Rated power

1.5kW

Cut in/Cut out speed

4m/s & 20m/s

Generator type and ratings

1.5kW PMSG @ 50 Hz

High Voltage DC Bus Configuration (HVDC)

In this configuration, the power generated from the different sources is delivered to the load through a high voltage DC bus. This configuration requires the installation of an ad hoc network operating in DC. Figure 1 shows a schematic of a stand-alone hybrid system including WEG, PVG and BSS devices. The number of conversion stages is relatively low; this reduces the power conversion losses as well as the cost. 2.2

Low Voltage DC Bus Configuration (LV-DC)

Figure 2 shows a schematic diagram of a standalone hybrid system comprising WEG, PVG and BSS systems. The generated power is delivered to the load via a low voltage DC bus (36100)V. In a preliminary simulation study, the low DC bus voltage is rated at 96V. In this configuration, the BSS is attached directly to the low voltage DC bus; this allows to store energy without conversion losses.

ENE162-3


HV - DC Bus

PMSG

DC DC

AC DC

Filter

DC

AC Load

AC

DC DC

PVG

MPPT Control

DC DC

BSS

Figure 1: Schematic of HVDC bus hybrid system

PMSG

DC DC

AC DC

LV - DC Bus

Filter

Filter

DC

DC DC

AC

AC Load

Full Bridge (HF trans.)

PVG

DC DC MPPT Control

BSS

Figure 2: Schematic of LVDC bus configuration 2.3

High voltage AC Bus configuration (HVAC)

Figure 3 shows a schematic diagram of the traditional stand-alone hybrid system including WEG, PVG and BSS systems. The generated power is delivered to the load through an AC bus. The great advantage of this configuration is that it can likely use existing infrastructures and facilities. This allows an easily allocation of different elements of the system even at considerable distance. In this context, the possibility of achieving ‘plug and play’ modules designed to be directly connected to the AC single phase domestic network is an interesting perspective. The WEG, PVG and ESS elements in the previous configurations are controlled by independent control systems, also accounting for the Maximum Power Point Tracking (MPPT). Suitable mathematical models of all the elements of the system have been previously modelled and a detailed description of the models is beyond the objective of this paper [ 6-8].

ENE162-4

Palestine Polytechnic University Third International Conference on Energy and Environmental Protection in Sustainable Development ( ICEEP III), October 9-10, 2013, Hebron, West bank, State of Palestine Paper Code. No. ENE162 HV-DC Bus

PMSG

AC DC

DC DC

DC AC

Filter

PVG

DC

DC

AC

DC

AC Load

MPPT Control

BSS

DC

DC DC

AC

Figure 3: Schematic of HVAC bus configuration 3

COMPARATIVE AND PERFORMANCE INDEXES

In fact no standard techniques are available for the selection of the optimum configuration of hybrid power system plants that in practice relies on the experience and the skill of the designer [6]. An efficient comparison is made on the basis of specific control strategies, real distribution of solar radiation, wind speed and load as well as taking into account the efficiency of each power converter. To do this a dynamic models of each architecture of the above presented configurations are designed in MatLab/Simulink/SimPowerSysTM[6]. The comparison has been accomplished on the basis of three performance indexes:  Global efficiency  Efficiency in transferring energy from RES to BSS,  Efficiency in transferring energy from BSS to load, The global efficiency ( 𝜂𝑔𝑙𝑜𝑏𝑎𝑙 ) of an hybrid RES power plant is defined as shown in Eq. (1): 𝜂𝑔𝑙𝑜𝑏𝑎𝑙 =

𝐸𝑜𝑢𝑡

(1)

𝐸𝑖𝑛

𝐸𝑐𝑜𝑛𝑣._𝑙𝑜𝑠𝑠𝑒𝑠 = 𝐸𝑖𝑛 − 𝐸𝑜𝑢𝑡 𝜂𝑔𝑙𝑜𝑏𝑎𝑙 = 1 −

(2)

𝐸𝑐𝑜𝑛𝑣._𝑙𝑜𝑠𝑠𝑒𝑠

(3)

𝐸𝑖𝑛

Where Econv._losses is the total energy conversion losses, Ein is the input energy from all the different energy sources which are PVG, WEG and BSS in this study. The calculation of global efficiency based on daily input weather data and load using dynamic MatLab/Simulink/SimPowerSysTM models of the above mentioned configurations. A typical day is divided into 24 intervals each one hour long. The energy conversion losses in one interval are computed based on a typical efficiency curve of each power converter using Eq. (4).

ENE162-5

Palestine Polytechnic University Third International Conference on Energy and Environmental Protection in Sustainable Development ( ICEEP III), October 9-10, 2013, Hebron, West bank, State of Palestine Paper Code. No. ENE162 95

100 90

60 50 40

85

80

Efficiency(%)

Efficiency (%)

70

Efficiency(%)

80

Efficiency (%)

90

90

90 85

80

30

10

70 0 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 100

0

70

100

80

75

75

75

20

85

70 0 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 100

0

100

Pin/Prated Pin/Prated

Pin/Prated Pin/Prated

Figure 4: Typical efficiency of a 2kW DC/DC converter (left) and of a 2kW rectifier (right) 𝐸𝑐𝑜𝑛𝑣−𝑙𝑜𝑠𝑠𝑒𝑠 = 𝑃𝑖𝑛−𝑐𝑜𝑛𝑣 (1 − 𝜂𝑡𝑦𝑝𝑖𝑐𝑎𝑙−𝑐𝑜𝑛𝑣 ) ∗ 𝛥𝑡

95

96

92

96

93 92

94 93

85

88 80 86

85

95

90 80 70 10 20 030 40 50 60 70 80 100 90

Pin/Prated

90 20 30 40 50 60 70 80 100 90 110

Pin/Pratedi n

80

92 75 84

91 91

90 20

90

Efficiency(%)

94

Efficiency(%) Efficiency (%)

95

Efficiency (%)

Efficiency (%)

90

90

95

Efficiency (%)

Where Pin-conv is the input power of a converter, ηtypical-conv is the typical efficiency of the converter at measured Pin-conv. Each amount of produced and consumed energy has been normalized in reference to the total energy consumed by the load. The typical efficiency curves of models’ power converters are shown in Figures 4 and 5.

Efficiency(%)

(4)

30 50 60 70 80 90 100 110 90 40 20 30 40Pin/Prated 50 60n 70 80 90 100 i

110

70 82 0

75

10 10

Pin/Prated

20 20

30 30

40 50 6070 70 80 90 100 40 50 60 40 50 60 070 1080 2090 30 100 Pin/Pratedc Pin/Pratedr Pin/Prated

Figure 5: Typical efficiency of a 5kW PWM inverter (left) and of a 5kW full bridge DC/DC converter (right) The flow chart shown in Figure 6 illustrates the global efficiency calculation procedure.

ENE162-6

70

80

90


Figure 6 : Flow chart of global efficiency calculation in each configurations 5

SIMULATION RESULTS

Simulations have been accomplished using MatLab/Simulink/SimPowSysTM dynamic models. Figure 7 shows the MatLab/Simulink/SimPowSysTM simulation model for the HV-DC schematic diagram shown in Figure 1. The schematic diagrams of HVAC and LVDC are modelled in the same way [6]. The average residential load demand profile for one home in Italy in the cold season considered in this study . The average wind speed, solar irradiance and air temperature data used are those of a typical day of December recorded at CNR/ITAE-Italy /Messina [6].

ENE162-7

Palestine Polytechnic University Third International Conference on Energy and Environmental Protection in Sustainable Development ( ICEEP III), October 9-10, 2013, Hebron, West bank, State of Palestine Paper Code. No. ENE162 mpp_cont Inv_cont L2

T (C)

+

G (W/m2)

PV+

D1

Vdc

-

PV-

-

T em(C)

DC bus

PVG

Load

g

Vac

A

+ - v

DC/DC

G(W/m2)

+

C -

B

+

+

+ v -

Vac

-

mppt

Vw(m/s)

Weather Data

A

A

Vw(m/s) B

B

C

C

+

WEG

PWM IGBT Inverter

WT + +

C1 -

WT -

AC/DC

-

D2

DC/DC2 Bid_cont bid_cont

+

SOC_Batt

Batt_sig

+

bat+ bat-

-

-

BSS

Bidirectional DC/DC converter

Figure 7 : MatLab/Simulink/SimPowSysTM of HV-DC bus configuration The results of the simulation studies are shown as following: 5.1

Global Efficiency Figure 8 shows the global efficiency of each configuration.

Figure 8: Global efficiency of different bus voltage configurations It is clear that the HVDC bus configuration obtains the highest global efficiency. The global efficiency doesn’t depend only on the number of conversion steps as generally known. In fact other conditions may play significant roles in determining the effectiveness of energy flow from different energy sources and load such as the control strategy and the efficiency of power converters. 5.2

Efficiency of energy transfer from RES to BSS (ηRES-->BSS)

Figures 9 and 10 show the maximum and average ηRES-->BSS index, respectively. It is clear that HVDC and LVDC achieve the highest levels of efficiency.

ENE162-8


Figure 9: Maximum ηRES-->BSS index of three bus voltage configurations

Figure 10: Average ηRES-->BSS index of three bus voltage configurations 5.3

Efficiency of energy delivered from BSS to load (ηBSS-->Load)

Figures 11 and 12 show the maximum and average ηBSS-->Load index of each architecture of a stand-alone hybrid power system, it is clear that the efficiency of energy delivery from BSS to load are very satisfactory in all configurations in HVDC and HVAC but LVDC is the lowest.

Figure 11: Maximum ηBSS-->Load index of three bus voltage configurations

ENE162-9


Figure 12: Average ηBSS-->Load index of three bus voltage configurations The sensitivity to seasonal load variations is investigated for each configuration. For this purpose, the simulation runs with three seasonal load profiles [9], namely: 0 or base load profile, 0.27 and -0.27 profiles, obtained multiplying the base profile respectively by a factor 1.27 and 0.73. Figure 13 illustrates global efficiency of the three configurations over seasonal load variation. HVDC configurations is less sensible to load variation. The global efficiency in HVAC and LVDC configurations is sensible to load variations as if the load demand increases the efficiency improves.

Figure 13: Global efficiency of different configurations with seasonal load variation 6

CONCLUSION

1. Three architectures for a stand-alone hybrid system encompassing PVG, WEG and BSS are studied and analysed and suitable dynamic models have been built. 2. Some performance indexes, namely: the global efficiency, the efficiency in transferring energy from RES to BSS and the efficiency in transferring energy from BSS to load have been selected to evaluate the three considered configurations. 3. On the basis of these indexes it is found that (at least for the specific case considered):

ENE162-10




The HVDC bus configuration features the highest global efficiency 79.3%. HVAC and LVDC configurations show lower levels of efficiency: 72.7%, 71.9%, respectively. Under the global efficiency point of view HVDC is less sensible to load variation than HVAC and LVDC.  The LVDC configuration is particularly efficient in storing energy (ηRES-->BSS) while HVAC is less efficient.  In case of energy delivery to load from BSS (ηBSS-->Load), all configurations obtain a quite similar efficiency. 4. Based on the above discussions and understandings for the considered case study the HVDC bus configuration is the most effective among the three different configurations.

ENE162-11


REFERENCES [1] G. Grandi, D. Casadei, C. Rossi, Direct Coupling of Power Active Filters with Photovoltaic Generation System with Improved MPPT Capability, in Proc. 2003 IEEE Bologna Power Tech Conference, vol. 2, June, 2003. [2] Y.-C Kuo, T.-J Liang, and J.-F Chen, "Novel Maximum-Power -Point-Tracking Controller for Photovoltaic Energy Conversion System", IEEE Trans. Ind. Electron., vol. 48, no. 3, June. 2001. [3] W. El-Khattam M.M.A. Salama, Distributed generation technologies, definitions and benefits , Science direct, electric power systems research 71 (2004) 119–128. [4] B. Nelson, M. H. Nehrir, and C. Wang, Unit sizing and cost analysis of stand-alone hybrid wind/PV/fuel cell power generation systems, Renewable Energy, Volume 31, 2006, pages 16411656,. [5] Testa, S. De Caro, R. La Torre, T. Scimone , Optimal size selection of Step-Up Transformers in PV Plants. International Conference on Electrical Machines (ICEM 2010), 6-8 September, Rome, Italy. [6] Aysar Yasin, Distributed Generation Systems Based on Hybrid Wind/Photovoltaic/Fuel Cell Structures. PhD dissertation, University of Catania/Italy , 2011. [7] A.Yasin, G. Napoli, M. Ferraro and V. Antonucci, 2011. Modeling and Control of a Residential Wind/PV/Battery Hybrid Power System with Performance Analysis. Journal of Applied Sciences, 11: 3663-3676. [8] Aysar Yasin, Distributed generation System based on renewable energy sources – case study. In proceeding of Second Emuni ReSouk - The Euro-Mediterranean Student Multi-Research Conference –14 June 2010, Pages: 855-866, ISBN 978-961-6805-02-9.

[9] Ross, Michael M.D. Comparison of AC, DC, and AC/DC Bus Configurations for PV Hybrid Systems. Report submitted to CETC-Varennes in fulfilment of Contract #3-1542SR. Varennes, Québec, Canada: Natural Resources Canada, 2004.

ENE162-12

Conference Full Paper template

Conference Full Paper template

Suggest Documents