Genetic Algorithm Based Closed Loop Control for

International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia

Genetic Algorithm Based Closed Loop Control for Blood Glucose Insulin System N.Sivaramakrishnan

S.K Lakshmanaprabu

M.Vamsikrishna muvvala

Assistant Professor Research scholar Assistant Professor Department of Electronics & Instrumentation engineering B.S Abdur Rahman Crescent University, Chennai, India

Abstract Blood glucose level is regulated by production of insulin by the pancreas. Type 1 diabetes mellitus patients cannot produce any insulin and they must administer insulin shots several times a day to regulate their blood glucose level. Hence, continuous monitoring and control of glucose level is necessary for Type 1 diabetes patients. In this paper, Bergeman minimal model is consider and then closed loop system is developed with insulin pump. This closed loop system essentially acts as an artificial pancreas. The closed loop control parameters are optimally tuned to control the blood glucose level most effectively. The Internal model control and proportional integral controller are implemented and the control parameters are tuned optimally. The comparative simulation results are compared and analysed. Keywords: IMC control, PID, Genetic Algorithm, Blood Glucose level. 1. Introduction: Type 1 diabetics disease mostly affecting the health of children. The continuous monitoring of glucose level is required to control the glucose level in the blood. There are some advanced techniques are introduced to monitor and control of glucose level. Insulin pump device and glucose monitoring devices are integrated to form an automatic insulin pump device. The digestion process considered as slow process (dead time process) while consume the food, automatically the blood glucose level in the blood is changed, for type 1 diabetic to injection of insulin from insulin pump in exact quantity. The blood glucose sensor monitoring the continuous value of glucose and Insulin pump injects the insulin based on lag of blood glucose. A genetic algorithm based PID control algorithm is proposed to control blood glucose level effectively in type 1 diabetics. The artificial pancreas (a closed loop control system) enables diabetic patient to maintain normal glucose level by providing the right amount of insulin at the right time with no need for human interaction when it comes to decision making.[1] A controller has been proposed for decision making. A PID controller regulates the glucose level in type -1 diabetic patients , an active sliding table is used to prescribe insulin infusion rates[2] The bio-inspired method for in-vivo is used to model a pancreatic-cell and this system implemented using analogue integrated circuits[3]. 430

In some advanced control techniques to combined the two types of controllers like proportional integral derivative (PID), and fuzzy logic controllers (FLC) and control the blood glucose level.[3] a closed-loop system which utilizes modified glucose insulin interaction model is considered. The modified model was derived by adding an exogenous insulin infusion term. Two control algorithms are used for exogenous insulin infusion: a Mamdani type fuzzy logic controller (FLC), and a fuzzy-PID controller [5] The tuning is complex in PID controller and the optimum tuning value is not achieved in fuzzy logic controller. In this paper a genetic algorithm based fuzzy PID controller proposed for closed loop blood glucose system to increase the performance. 2. Mathematical model: Various mathematical models have been proposed to understand the Diabetes dynamics and to correlate the relation between glucose and insulin distribution models that helps in designing of model for control of diabetics. All these models are taken considerations at specific rules and procedures to design these models. Our body system is entirely the non linear system and this model can be compared with the Brownie motion which has various stochastic processes in our body system for various factors. So by taking certain parameters with assumption of some steady state condition, the appropriate model can be

Journal of Engineering Technological Research Half Yearly Vol: 8 March 2017 ISSN: 2229-9262

International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia designed. These models have many limitations in predicting blood glucose in real time clinical situation because of the essential requirement of frequently updated information about the models like glucose loads, insulin availability, and necessary standatise glucose sensor based system [4]. Reference model charts of change of biosignals with amount of insulin injected. Consider a mathematical model comprised of glucose level G, glucose uptake activity X and insulin level I. Many parameters has to be taken into account. Based on these values a mathematical model can be formed using these parameters. This model includes the basal values also i.e. Gb and b I .

uptake ability ( 1 min− ). 𝑚3 - The insulin independent increase in glucose uptake ability in tissue per unit of insulin concentration Ib ( 2 min− (μU/ml)). 𝑚4 - The rate of the pancreatic β-cells‘ release of insulin after the glucose injection and with glucose concentration above h [(μU/ml) min-2 (mg/dl)-1]. 𝑚5 - The threshold value of glucose above which the pancreatic β-cells release insulin. 𝑚 6 - The first order decay. Consider a diabetic that is modelled using following set of parameters ( Lynch and Bequette 2001) process Transfer function,

Gp( s) 

3.79 (40s  1)(10.8s  1)

Meal disturbance transfer function is considered as

Gd( s) 

8.44 s(20s  1)

The meal disturbance is given as pulse and assume that 1 50 g glucose meal is consumed over a 15- minute period; the pulse then has a magnitude of 3.33g/minute of a duration of 15 minutes.

Figure 1 Bergmann’s Minimal Model

3. Controller Design 3.1. PD control

The modal is defined as: 𝑑𝐺 𝑑𝑡 𝑑𝑋 𝑑𝑡 𝑑𝐼 𝑑𝑡

=-𝑚1 𝐺 + 𝑚2 𝐼 + 𝑚1 𝐺𝑏.

= −𝑚2 𝑋 + 𝑚3 𝐼 − 𝑚3 𝐼𝑏 + 𝑚6 𝐼𝑏 .

= −𝑚3 𝐼 + 𝑚4 𝐺 + 𝑚4 𝑚5 − 𝑚6 𝐼 + 𝑚6 𝐼6 .

The mathematical models variables are described as: G (t) The plasma glucose concentration at time t (mg/dl). X (t) The generalized insulin variable for the remote compartment (min-1). I (t) - The plasma insulin concentration at time t (μU/ml) Gb - This is the basal preinjection level of plasma glucose (mg/dl). Ib - This is the basal preinjection level of plasma insulin (μU/ml). 𝑚1 - Insulin independent rate constant of glucose rate uptake in muscles, liver and adipose tissue ( 1 min− ), 𝑚2 - The rate of decrease in tissue glucose

The PD controller inherent the advantages of proportional and Derivative controller the Proportional controller produces an instantaneous response to control the error and its magnitude is proportional to output and error signal. A derivative controller differentiates the error signal and produces the controller signal. It used to find the future error signal and it reduces the peak overshoot to achieve the transient response of output with in small settling time. The equation of PD controller as, 𝑢 𝑡 = 𝐾𝑝 𝑒 𝑡 + 𝐾𝐷

𝑑𝑒 𝑑𝑡

𝑒 𝑡 − 𝐸𝑟𝑟𝑜𝑟 𝑠𝑖𝑔𝑛𝑎𝑙 𝑑𝑒 − 𝐹𝑢𝑡𝑢𝑟𝑒 𝐸𝑟𝑟𝑜𝑟 𝑆𝑖𝑔𝑛𝑎𝑙 𝑑𝑡 𝐾𝑝 − 𝑃𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛𝑎𝑙 𝐶𝑜𝑠𝑛𝑡𝑎𝑛𝑡

431 Journal of Engineering Technological Research Half Yearly Vol: 8 March 2017 ISSN: 2229-9262

International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia 𝐾𝐷 − 𝐷𝑒𝑟𝑖𝑣𝑎𝑡𝑖𝑣𝑒 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑢 𝑡 − 𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑙𝑒𝑟 𝑜𝑢𝑡𝑝𝑢𝑡 3.2. Internal Model Controller (IMC) The IMC controller has good tracing performance to track and regulate the blood glucose level It has good set-point tracking in many process control applications and rejection the disturbance for the stable processes to include a filter with this controller to act as a good disturbances rejection for unstable processes. Thus, an appropriate IMC filter to design an IMC controller for better set-point tracking in unstable processes. Normally in PD controller its is complex to tune the controller parameters Kp and KI and PD controller is not suitable for unstable process to overcome this tuning complexity IMC Controller is proposed for blood glucose system. This IMC controller contains a single tuning parameter λ, so tuning is simple compare with PD controller. Glucose Model is,

p( s ) 

K

 s 1

e s

The IMC controller q(s) is, 𝑞 𝑠 = 𝑃(𝑠)− ∗ 𝑓 where p(s)- is a invertible model without dead time and right hand zero.

p( s) 

 s 1 K

The filter f(s) is

Figure 2 Closed loop GA based IMC control of blood Glucose system 4. Genetic algorithm The Genetic algorithm (GA) is used to solve highly complex problems, the solution of a problem is expressed as set of parameters and this parameters are regarded as genes of a chromosome and it‘s structured by a string of concatenated values. This is called as encoding scheme, based on data values these variables are expressed as binary values, real number values and some other codes values. Initially chromosomes are randomly generated this chromosome consist of Kp ,Ki, Kd parameters with value bounds varied depend on the delay and objective functions used. The value bounds are set using Ziegler-Nichols rule value The fitness of each chromosome is assessed and a survival of the fittest strategy is applied. In this work, the error value is used to assess the fitness of each chromosome. There are three main operations in a genetic algorithm: reproduction, crossover, and mutation. Genetic Algorithm Steps Step 1. Initialize the parameter with a population of random

1 𝜆𝑠+1 The Filter for blood glucose model is selected as, 1 f (s)    s  13

solutions, such as crossover rate, mutation rate, number of

The filter time constant λ is the only tuning parameter of IMC controller. The genetic algorithm based tuning optimization concept is proposed to tune the parameter λ to minimization of ISE( Integral Square Error). The objective function of optimization (J) is given as,

Step 3. Proceed with crossover and mutation operation and

𝑓(𝑠) =

clusters, and number of generations. Determine the coding mode. Step 2. Compute and evaluate the value of the fitness function. make up the new cluster. Step 4. Repeat Step 2, till the best value is obtained. 4.1 Coding and Decoding:


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia Genetic algorithms work with a population of strings or

operator mutation is introduced in genetic algorithm with

chromosomes, instead of considering parameters directly.

appropriate

Hence, to solve our problem, the controller parameter vector

alternation of the gene from zero to one or from one to zero

should be coded to a string called chromosome. For

with the mutation point determined uniformly at random.

probability.

Mutation

is

an

occasional

convenience and simplicity, the binary coding method is chosen. Based on the binary coding method, every element of the parameter vector is coded as a string of length, which 4.2 Fitness Fitness is a measure to evaluate the suitability of a chromosome. By the principle of survival of the fittest, a chromosome with higher fitness value has a higher probability of contributing one or more offspring in the next By

employing

genetic

algorithm,

iteration is set as 100. 5. Simulation results & Discussion:

consists of zeros and ones for the desired resolution.

generation.

The population size for optimisation is set as 20 and

the

performance criterion is related to fitness function and optimal PID parameters are derived by minimizing an objective, which inculcates a weighted combination of ISE.

The following simulation results show that the desired objectives of achieving minimum ISE are achieved. The simulation started with time t=0 , without any disturbance. The controller tracking the set point of 80 mg/decilitre level and then the meal disturbance with magnitude of 3.33g/minute is applied at 400th minute as a pulse which raises the glucose level of patient. The GA-IMC based controller rejects the meal disturbance effectively than GAPID control. The GA-IMC regulates the glucose level within 40 minutes after having meal whereas GA-PD takes longer time to regulate the glucose level.

4.3 Crossover Reproduction is a basic operator of genetic algorithm. It is operated on the basis of the survival of the fittest. In each generation, the chromosome of the current population is reproduced or copied into the next generation, according to the reproduction probability, which is defined in where is the population size. 4.4 Reproduction

The optimised control value for minimum ISE is tabulated for GA-PD and GA-IMC in table 1,2.

Reproduction directs the search of genetic algorithm towards the best individuals. Crossover operation is performed to exchange the information between any two chromosomes via probabilistic decision in the mating pool

GA-PD

Table.1 GA-PD Controller kp ki ISE -0.0527 -2.37 1666

and to provide a mechanism to mix chromosomes with the splice point. 4.5 Mutation

GA-IMC

Table.2 GA-IMC controller  ISE -0.12 45810

In genetic algorithms, however, the gene pool tends to become more and more homogeneous as one better gene begins to dominate after several generations and leads to premature convergence of non optimal solution. To overcome this undesirable convergence, the third genetic

The blood glucose level has to maintain a particular level while it increases, naturally pancreas inject the insulin to our body to maintain the glucose level whereas for type-1 diabetic patients pancreas doesn‘t inject the insulin so insulin pump injects the insulin The glucose sensor


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia continuously monitoring the blood glucose level after a meal the glucose level is increases drastically it has refers to input disturbances to the process. The Process is first order with dead time process. The genetic algorithm based single loop PD controller is used for control the blood glucose system. Insulin pump there is a stepper motor it coupled with a injection needle shaft. If the glucose level increases the ID controller produces a control signal in terms of voltage to drive a stepper motor. The displacement of shaft is equivalent to injection of insulin into human body. The PD controller tuned using GA to achieve minimum ISE. The PD control signal affects the insulin system due to its oscillatory response; the time take to regulate the blood glucose is high. To overcome this complexity genetic algorithm based (GA) IMC controller is used. In practice, the glucose level does not enter the blood stream immediately. Meals take some time to reach it into blood stream. Assume, the patient is taking 50g meal over 15 minutes. Hence 3.3g/minutes pulse is applied in simulation for 15 minutes. The glucose level increases suddenly due to the meal, and it is gradually reduced by applying insulin. 3.5 3 Meal

meal(g/minute)

2.5 2 1.5 1 0.5 0 -0.5

0

100

200

300

400

500

600

700

800

900

1000

200

glucose(mg/deciliter)

Glucose Level setpoint PD IMC

150

100

50

0

0

100

200

300

400

500 time(min)

600

700

800

900

1000

6. Conclusion: In this paper GA based PD controller proposed for blood glucose system The simulation part was done by taking Bergmann‘s minimal model as a reference model from which glucose and insulin kinetics where referred and then it is simulated. In the simulation, Glucose (meal) is taken as a disturbance which has to be controlled by infusing the insulin in blood of the patient which maintains the glucose

level in blood as normal. It is inferred that GA-IMC and GA-PD controllers are compared. The superiority of GAIMC is shown and discussed. The optimal selection of filter constant in IMC regulates the blood glucose level effectively. There is major research effort to develop implantable glucose sensors and insulin pumps. The IMC is a model based control which requires exact model of blood glucose system, but the modelling varies patient to patient. The modelling has to be done for each patients using his glucose level data. References: [1]. Sawsan M. Gharghory, Mervat Mahmoud ―Low power fuzzy control system for adjusting the blood glucose level ‖ Microelectronics (ICM), 2016 28th International Conference on IEEExplore 2017 [2]. F. Chee; T. L. Fernando; A. V. Savkin; ―Expert PID control system for blood glucose control in critically illpatients‖ IEEE transaction [3]. Ilias Pagkalos, Pau Herrero, Christofer Toumazou ―BioInspired Glucose Control in Diabetes Based on an Analogue Implementation of a β-Cell Model‖ IEEE transaction [4]. Sandhya and Deepak Kumar ‖ Mathematical Model for Glucose-Insulin Regulatory System of Diabetes Mellitus‖ Advances in Applied Mathematical Biosciences. ISSN 2248-9983 Volume 2, Number 1 (2011), pp. 39-46 [5]. Ahmed Y. Ben Sasi , Mahmud A. Elmalki ― A Fuzzy Controller for Blood Glucose-Insulin System‖ Journal of Signal and Information Processing, 2013, 4, pp no:111-117 [6]. A. Jayachitra and R. Vinodha ―Genetic Algorithm Based PID Controller Tuning Approach for Continuous Stirred Tank Reactor‖advances in Artificial Intelligence , Volume 2014 [7]. Bhawna Tandon, Randeep Kaur ―Genetic Algorithm based parameter tuning of PID controller for composition control system‖ International Journal of Engineering Science and Technology (IJEST). [8]. Jinhua Zhang, , Jian Zhuang, Haifeng Du, Sun‘an Wang ―Self-organizing genetic algorithm based tuning of PID controllers‖ journal of Information Sciences Elseiverpublications Volume 179, Issue 7, 15 March 2009, Pages 1007–1018 [9]. Blood glucose control algorithms for type 1 diabetic patients:A methodological review,Katrin Lunze a, Tarunraj Singh b, Marian Walter a, Mathias D. Brendel, Steffen Leonhardt a Department of Electrical Engineering, RWTH Aachen University, Aachen 52074, Germany [10]. A new approach to diabetic control: Fuzzy logic and insulin pump technology, Paul Grant Department of Medicine, Royal Sussex County Hospital, Eastern Road, Brighton, East Sussex, and United Kingdom.



Talent Retention – An Assessment of Psychological Contract of Indian Information Technology Professionals Mr. T. Naresh Babu

Dr. N.Suhasini Dr. G. Krishna Mohan

Assistant Professor, KSRM College of Management Studies, Kadapa-516003. Andhra Pradesh. Assistant Professor, KSRM College of Management Studies, Kadapa-516003. Andhra Pradesh. Professor, Principal, KSRM College of Management Studies, Kadapa-516003.Andhra Pradesh. Abstract Retention of talented employees has become a strategic business challenge and critical driving force of corporate performance. The changing dimensions of corporate due to impact of globalization, liberalization, and privatization are influencing the human resource practices across the globe. Corporate require adopting strategies to attract, motivate and retain talented employees in the organization.. In order to retain critical talent the psychological contract plays vital role which an organization need to build. Psychological contact refers to employee‘s subjective interpretations and evaluations of their deal with the organization. Building a positive psychological contract increases employee retention and commitment. The research examines the variables of psychological contract in retaining Indian IT professionals. A sample size of 100 IT professionals was considered for study with the help of structured questionnaire. Factor analysis was performed to explore the factors influencing psychological contract. The results of the study revealed that the factors like employee engagement, employee motives intentions, job sculpting, positive work environment, employee empowerment, motivation, fair appraisal and effective communication build psychological contract.. IT firms need to develop effective talent management policies to develop positive psychological contract for talent retention. Keywords:- Talent retention, Economic turmoil, Psychological contract, Attrition. Introduction:Over the past decades, the economic environment has been changed dramatically due to factors like globalization of competition, the tightening of skilled labor market, advancements in technology, the growth of knowledge economy, organizations are required to be more flexible and to increase their productivity. This has reduced the job security of employees at all levels in the organization. (King 2000). HR managers are pressed to attract and retain talented employees who have competencies that are critical for organizational survival. It became very difficult to retain due to the tendency of employees to attach more importance to their own career path rather than to organizational loyalty which results in increased rates of voluntary turnover (Capelli 2001). Retention management has become a popular concept to examine the portfolio of HR practices put into place by organizations in order to reduce voluntary turnover rates. Another important concept that gained interest as a construct relevant for understanding and managing contemporary employment relationship is the psychological contract which refers to employee subjective interpretations & evaluations of their deal with the organization (Turnley & Feldman 1998). Talent retention is effective when employee‘s expectations relating to HR practices are managed. Both researchers and HR practitioners argue that the employment relationship is under going fundamental changes that have impact on attraction, motivation & retention of talent. In view of large costs associated with employee turnover even in a global economic downturn, HR managers need to focus on

psychological contract which is helpful to develop positive attitude of employees towards organizations. Understanding and effectively managing psychological contract will help organizations thrive, but these needs to be clear agreement on the contributions that the employees make and employers expect. The effective management of psychological contract can thus result in increased job performance, lower staff turnover and higher job satisfaction for both the employee & employer. (Dell Campo, 2007) Literature Review:The importance of psychological contract in talent retention can be studied through various theoretical and empirical studies as follows: Munden, Mandl, and Solley (1962) and Schein (1965) coined the term, psychological contract. Rousseau (1989) conceptualized the psychological contract as individual-level cognition focusing on employee‘s perceptions. She defines the psychological contract in terms of individual beliefs regarding the terms and conditions of a reciprocal agreement between that focal person and another party or as ―individual beliefs in a reciprocal obligation between the individual and the organization‖. The types of promises contained in an employee's psychological contract can be explicitly or implicitly communicated in a written document, oral discussion, or organizational practices and procedures. Morrison and Robinson (1997) define the psychological contract as the employee‘s beliefs about reciprocal and promissory obligations between himself and the organization. Although the psychological contract is an individual-level construct, it reflects beliefs about tangible and intangible items that are to be exchanged in the context


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia of a dyadic relationship. Shore and Tetrick (1994) stressed that psychological contract was a set of beliefs shaped by multiple sources of input and by cognitive and perceptual biases. The psychological contract is especially important in today‘s global economy where employees have increased mobility (Hytrek and Zentgraf 2008). Employees, on average, no longer stay with one organization for their entire career. To increase intention to stay, job satisfaction, and job involvement, it is the responsibility of the organization to fulfill the psychological contracts of their employees (Stone and Gallagher 2010). Pffeffer,(1998); Woodruffe, (1999) highlighted the importance of financial rewards for employee retention. Provision of attractive rewards and recognition motivate employees to build psychological contract which leads to talent retention. Bevan (1997) in his study revealed that high percent of people who had left their employees gave dissatisfaction with pay as the main reason for leaving. Hall& Moss (1998), Klein&Tang, (2003) suggested that a company wants to strengthen its bond with its employees must invest in the development of these employees. Roehling et.al (2000) Provision of mentoring or coaching to employees, organization of career management workshops, and the set up of career management workshops and the set up of competency management programs reduced the intentions of employees to leave the organization. Cappelli (2001) suggested that loyalty to the organization is a thing of the past but loyalty to one‘s own colleagues acts as an effective means of retention. Organizations can contribute to the creation of positive social atmosphere by stimulating interaction and mutual cooperation among employees and through open and honest communication between management and employees. According to Turnley & Feldman(2000) psychological contract reflects accumulating evidence for its influence on diverse work related outcomes. These studies show that employees evaluate the inducements they receive from their organization in view of previously made promises and that this evaluation leads to feeling of psychological contract or breach. Robinson et.al (1994) reported that contract violations by their employer two years after organizational entry. Study reveals that psychological contract violation is relatively common and this could explain the difficulties organizations are currently experiencing in retaining their employees. Effective employee retention is possible in line with employees subjective views and expectations. Research Methodology:Objectives of the study:To examine and investigate the various factors of psychological contract. To assess the role of individual and organizational related variables in building psychological contract. Methodology and Sample Design:The study explores the underlying factors for psychological contract in the IT sector by conducting online survey to the

employees. The researcher targeted online survey by administering a structured questionnaire for 100 respondents. Each of the attribute in questionnaire was measured on 5 point scale and questionnaire was tested with Cronobach‘s alpha reliability test, factor analysis technique was used to analyze the primary data. Results and Discussion:The data was analyzed using SPSS 16.0 version. First the data was tested for reliability by determining Cronobach‘s alpha which is higher than 0.6 and also Kaiser-Mayer-Olkin measure of sample adequacy and Bartlett‘s test of sphericity was used to examine the appropriateness of using factor analysis. Table 1: Reliability Statistics Cronbach's Alpha

N of Items

.672

21

Table 2: KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett's Test of Sphericity Approx. Chi-Square

.622 352.069

Df

210

Sig.

.000

The respondents were asked to respond for various individual and organizational variables. Individual variables included employee‘s self-efficacy, self-identity, controlling, sense of achievement and association with organization. Organizational variables includes employees motivating factors like job security, fair policies related to career growth and compensation, employers intentions, reward system, empowerment, employee long term well being, recognition and conducive work environment, effective leadership, open communication, transparent appraisal. Table 3: Communalities


Initial Extraction

International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia Marital status

1.000

.741

Career and promotional opportunities 1.000

.627

Honesty and faith

1.000

.609

Job security

1.000

.637

Reward system

1.000

.619

Effective leadership

1.000

.574

Sense of belongingness

1.000

.546

Open and honest communication

1.000

.698

Trust and respect

1.000

.528

Conducive environment

1.000

.762

Life interest

1.000

.760

Adequate and fair compensation

1.000

.702

Autonomy in goal setting

1.000

.670

Sense of achievement

1.000

.725

Participation in decision making

1.000

.687

Challenging assignment

1.000

.730

Constructive feedback

1.000

.745

Recognition and appreciation

1.000

.680

Extraction Method: Principal Component Analysis.

On the basis of Varimax rotation all the variables that have factors loading around 0.5 are extracted. Extraction has been limited to nine factors. All the 9 factors contributed to 67% of variance. 9.52% Variance is explained by factor 1, 9.52

Creativity and innovation

1.000

.608

Employee long term well being

1.000

.743

Employers intentions

1.000

.734

% explained by factor 2, 8.29% explained by factor 3, 7.14% explained by factor 4, 7.12% explained by factor 5, 6.68% explained by factor 6 , 6.59% explained by factor 7, 6.26% explained by factor 8 and 6.11% explained by factor 9.

Table 4 : Total Variance Explained

Initial Eigenvalues

Extraction Sums of Squared

Rotation Sums of Squared

Loadings

Loadings

Component Total

% of

Cumulative

Variance

%

Total

% of

Cumulative

Variance

%

Total

% of

Cumulative

Variance

%

1

3.354

15.973

15.973

3.354

15.973

15.973

2.001

9.527

9.527

2

1.917

9.130

25.103

1.917

9.130

25.103

1.999

9.520

19.046

3

1.646

7.839

32.943

1.646

7.839

32.943

1.741

8.290

27.337


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia 4

1.382

6.579

39.522

1.382

6.579

39.522

1.500

7.142

34.479

5

1.253

5.965

45.487

1.253

5.965

45.487

1.496

7.122

41.601

6

1.201

5.718

51.205

1.201

5.718

51.205

1.403

6.683

48.284

7

1.178

5.611

56.816

1.178

5.611

56.816

1.386

6.599

54.882

8

1.130

5.382

62.198

1.130

5.382

62.198

1.316

6.267

61.150

9

1.063

5.062

67.260

1.063

5.062

67.260

1.283

6.110

67.260

10

.905

4.309

71.569

11

.789

3.759

75.328

12

.764

3.638

78.966

13

.690

3.288

82.254

14

.646

3.077

85.331

15

.587

2.793

88.124

16

.521

2.480

90.604

17

.470

2.239

92.843

18

.433

2.060

94.903

19

.400

1.905

96.809

20

.363

1.728

98.537

21

.307

1.463

100.000

Extraction Method: Principal Component Analysis.



Table 5 : Component Matrixa Component 1

2

3

4

5

6

8

.363 .102

Honesty and faith

.664

-.254

Reward system

.618

-.106 -.224 .182 .256 -.127

Sense of belongingness

.480 .231 -.283

Trust and respect

.498

Life interest

.303 -.336 .452

Autonomy in goal setting

.725


.455 -.157 -.229


.259 -.137 -.345


.289 .104

Recognition and appreciation

.255

Creativity and innovation

.506 -.275 .226 -.188

-.263 .130 -.305

Employee long term wellbeing

.453 -.316

.120 .230 -.177 .355

Employers intentions

.383 .155 .230 .445 .277 -.256 .241 -.276 -.190

-.180

.278 -.122 .212

-.241 -.275 -.433 .181

-.243 -.162 -.136

.234 .556

.176

.248 -.154 -.194

.114 -.123 .139 .489 -.349

-.305 -.121 .279 .569 .352

.296 -.508

.388 .477 .275

Career and promotional opportunities .266 Effective leadership

-.601

9

Marital status

Job security

-.321 -.224

7

.452

-.417

-.147 .359 .104

.475 -.276 .296 -.363 .443 .477

-.170

-.363

.236 .509 -.291 .211

.211 .152 .218 .141 -.238


.336 -.144 -.263 .403 .455 .340


.492 .242

-.456 .220 .223 -.112 -.368

Adequate and fair compensation

-.123 .695 .226

.140 -.168

Sense of achievement

.358 .426 -.348

.364

.242 .201 -.209

.327

Extraction Method: Principal Component Analysis. a. 9 components extracted.



Table 6 : Varimax Rotated Component Matrixa Component Employers Positive Employee motives Job Fair Marital work Empowerment Motivation Communication engagement and Sculpting appraisal status environment intentions Marital status

.791

Honesty and faith

.584

Reward system

.514

Sense of belongingness Trust and respect

.593 .558

Life interest Autonomy in goal setting

.693 .492


.725


.810


.822

Recognition and appreciation Creativity and innovation

.747 .629


.763

Employers intentions Career and promotional opportunities

.713

Job security Effective leadership

.660 .642


.824


.823

Adequate and fair compensation Sense of achievement

.508 .615

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia Table 6 : Varimax Rotated Component Matrixa Component Employers Positive Employee motives Job Fair Marital work Empowerment Motivation Communication engagement and Sculpting appraisal status environment intentions Marital status

.791

Honesty and faith

.584

Reward system

.514

Sense of belongingness Trust and respect

.593 .558

Life interest Autonomy in goal setting

.693 .492


.725


.810


.822

Recognition and appreciation Creativity and innovation

.747 .629


.763

Employers intentions Career and promotional opportunities

.713

Job security Effective leadership

.660 .642


.824


.823

Adequate and fair compensation Sense of achievement

.508 .615

a. Rotation converged in 19 iterations.


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia FACTORS INFLUENCING PSYCHOLOGICAL CONTRACT OF IT PROFESSIONALS: Factor 1: Employee engagementTrust and respect, Autonomy in goal setting, Creativity and innovation, Career and promotional opportunities. In IT firm‘s employee engagement is a key to the retention of talent. Organizations seeking to build competitive advantage through engaged employees are often blocked by the challenges of engaging an entire workforce of individuals with unique values, culture, interests, and needs. Employees need to be treated with respect and allow them in goal setting. Creative ideas to be encouraged. Career and promotional opportunities enhance talent retention. Factor 2: Employers motives and intentions - Honest and faithful, Sense of belongingness, Effective leadership, Sense of achievement, Talented people need to feel valued and that their motives should give adequate importance.. By feeling appreciated, recognized and valued, the identified talent will not only be motivated, but highly engaged and aligned to the organization‘s goals and objectives. Effective leadership style helps to promote honest and faithful transactions with employees. Factor 3: Job Sculpting – Life interests, Employee long term well being, Effective leadership. The important aspect in IT firms is job-person fit. We need to make sure jobs are meaningful; people have the right tools and autonomy to succeed. Selection of right people for the right job based on employee potential. Factor 4: Positive work environment – Job security, Conducive environment, Work environment in organization shows both positive and negative impact on psychological well being of employees. It plays major role in providing autonomy in work, participation in decision making, good working conditions .Conducive work relationships between employer and employees develop psychological contract of employees. Factor 5: Empowerment- Participation in decision making, Challenging assignments, The technology updating and pressures for increased productivity necessitates the change of existing HR practices. Job enrichment helps to enrich jobs, giving people more decision-making power to increase employee empowerment. Factor 6: Motivation – Recognition and appreciation, adequate compensation, Talented people can be motivated with reward system and equitable compensation structure. Intrinsic and extrinsic rewards carry both surface and symbolic value. It indicates how much employees valued by the organization. Most organizations use rewards like incentives, benefits,

perquisites and awards combined to create an individual‘s compensation package. Factor 7: Fair appraisal – Reward system, Constructive feed back, Fair appraisal system acknowledges the importance in high performing organizations. Performance appraisal helps to fulfill psychological contract which describes expectations between employer and employee. Transparent appraisal system with continuous feedback maximizing the firms performance and to develop the employees in the environment of trust. Factor 8: Communication - Open and honest communication, Effective two way communication promotes good relation and trust between superiors and subordinates. Factor 9: Marital status - Married people are high concern about work life balance. They demand flexible work schedules which allow for success in personal as well as professional life. Provision of flexible work environment helps to build psychological contract. Conclusion and Recommendations:Organizations try their best to apply psychological contract to employees as a form of motivation and management tool. In today‘s work environment, employees are no longer receptive to command and control management style. Employees should be treated with trust, decency and respect, providing employees with challenging and interesting work in the form of job enrichment and job rotation. Fair performance appraisal system should act as a link between performance and rewards must be clearly communicated to employees. In order to motivate employees and manage their psychological contract in the employment relationship organizations need to implement fair and transparent appraisal system, recognize achievements, provide challenges through career advancements and job enrichments and create environment of trust and mutual respect. Organizations need to create the feeling of belongingness among employees to satisfy their socioemotional needs. Management has to develop the culture of employee empowerment within the organization. This leads to high performance and lower turnover. Organizations should focus on making sure that the people they hire are good match for the job and the work culture. Engaged employees act as an ambassadors and produce better results. References:1. Cappelli, P. (2001). A market-driven approach to retaining talent. Harvard Business Review on finding and keeping the best people (pp. 27-50). Boston : Harvard Business School Press.


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia 2. Hall, D. T., & Moss, J. E. (1998). The new protean career contract: Helping organizations and employees adapt. Organizational Dynamics, 26(3), 22-37. 3. Hancer, M., & George Thomas, R. (2003). Psychological empowerment of non-supervisory employees working in fullservice restaurants. International Journal of Hospitality management, 22(1), 3–16. 4. Herriot, P., Manning, W. E. G., & Kidd, J. M. (1997). The content of the psychological contract. British Journal of Management, 8, 151–162. 5. Hytrek, Gary, and Kristine Zentgraf (2008), America transformed: Globalization, inequality, and power New York, New York: Oxford University Press, Inc. 6. Pfeffer J., 1998, Six myths about pay, Harvard Business Review, May-June, 38-57. 7. Robinson, S. L., Kraatz, M. S., & Rousseau, D. M. (1994). Changing obligations and the psychological contract: A

longitudinal study. Academy of Management Journal, 37(1), 137- 152. 8. Roehling, M. V., Cavanaugh, M. A., Moynihan, L. M., & Boswell, W. (2000). The nature of the new employment relationship: A content analysis of the practitioner and academic literatures. Human Resource Management, 39(4), 305-320. 9. Rousseau, D. M. (1989). Psychological and implied contracts in organizations. Employee Responsibilities and Rights Journal, 2, 121–139. 10. Stoner, Jason S., and Vickie C. Gallagher (2010), "Who Cares? The Role of Job Involvement in Psychological Contract Violation," Journal of Applied Social Psychology, 40 (6), 14901514. 11. Turnley, W. H., & Feldman, D. C. (2000). Reexamining the effects of psychological contract violations: unmet expectations and job dissatisfaction as mediators. Journal of Organizational Behavior, 1- 21, 25-42.



Spatial Image Processing Sensor for Traffic Queue Length Measurement Rajaambika.S1, Dhivyabharathi.S2 Assistant Professor1,2 /Computer Science &Engineering1,2, Kathir College Of Engineering1,2, Coimbatore-641062

ABSTRACT As the social economic activities grow variously in recent days, an advanced traffic control system corresponding to the changes of road traffic circumstances becomes an urgent matter. In order to develop and realize a more optimum signal flow algorithm which corresponds to every moment changing traffic information, an advanced sensor which can collect more precise and more detail traffic flow information is essential. In this paper we present about an image processing sensor called SPITS (Spatial Image processing Traffic flow sensor), which is used to measure the traffic queue length very efficiently. An exclusive hardware unit is used in SPITS which has a simple configuration and high-speed processing. A traffic queue length measurement is done using a queue and delay length measurement algorithm. The basic idea of this algorithm is "Vehicle groupings of moving vehicles" from "vehicle groupings ofexisting vehicles" leaves "vehicle groupings of delay vehicles‖ This algorithm is realized using two processes, image preprocess and queue length measurement process. By using this algorithm, precise and stable queue length measurement is achieved. Further, the efficiency of the imaging sensor SPITS is established by the field test which was conducted under various timings (day, twilight, night) and different environmental conditions (fine, rainy, etc.). Thus the ‗high speed simple hardware and the high degree of accuracy‘ of SPITS makes it indispensable in the field of traffic queue length measurement. INTRODUCTION Systems where the process of applying any operation to image data for a given purpose are called image processing systems. Examples of operations include scene analysis, image compression, image restoration, image enhancement, image preprocessing, quantizing, spatial filtering, and construction of two- and three-dimensional models of objects. In this paper, we deal about the application of image processing sensor in measurement of traffic queue length. Basically, an ultrasonic vehicle detector is used to detect vehicle presence by the time difference of the reflection of ultrasonic wave fired from above the road surface to just under it. But especially queue and delay length are measured indirectly by the number of passed vehicles in a unit time. So a sensor which can collect more precise traffic flow information is needed. Also each Ultrasonic Vehicle Detector has to be installed above the road surface per a measurement lane and so there is a fear of spoiling the beauty of the city.

directly measure a spatial area like shown in Figure 1. And the Image Sensor has the following characteristics. 

It is expected to collect more precise and more detail queue and delay length information because of the direct measurement of them.



It does not spoil the beauty of the city so much because one image sensor installed aside and above the approach lane can measure more than one lane at the same time.

IMAGE SENSOR: On the other hand, the Image Sensor processes an image received from the ITV camera installed aside and above the approach lane at the traffic signal intersection, so it is able to

While the function of direct measurement of queue and delay length as well as the number of number of passed vehicles,


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia speed, vehicle type are the advantages of an image sensor, it is hard to measure them precisely in the distance from the camera position (which is usually placed above the approach lane at the traffic signal intersection) because of poor visibility. New measurement algorithms and a new hardware, led to the development of a new image sensor, “Spatial Image processing Traffic flow Sensor (SPITS)” to measure queue and delay length precisely even in the distance with a sophisticated method and in real time processing. The hardware configuration of SPITS as well as the queue and delay length measurement applied for accurate measurement of maximum queue length under various conditions (day, night, twilight, rain, etc.,) are discussed here.

Control Center or Consol unit where the measurement results are stored. HARDWARE CONFIGURATION:

SYSTEM CONFIGURATION: The outline and the configuration of SPITS is shown in the following. SPITS consists of Camera Unit and Control Unit. Camera Unit is installed above the approach lane on an arm attached to a high pole settled roadside. And Control Unit is installed at the lower end of the pole. SPITS consists of Camera Unit and Control Unit. Camera unit is installed above the approach lane on an arm attached to a high pole settled roadside. And Control Unit is installed at the lower part of the pole.

It has an exclusive hardware in Control Unit which has a simple configuration and can run high-speed processing. Figure 3 shows the hardware configuration. The NTSC image signal inputted from Camera Unit is amplified and analog-digital converted. Preprocessing Unit makes a Sampled-Points Image from the input image, and to this Sampled-Points Image executes the image processing‘s such as Time Difference, Background Difference, or Spatial Difference in high-speed. By using the executed results of Preprocessing Unit, CPU calculates traffic flow information such as queue length. CPU also executes controlling Preprocessing Unit, transfer processing of the measurement results to the TrafficControlCenter and moreover the man machine processing. In the case of initial adjustment or maintenance, a display board can be easily attached, and by using a monitor TV or a mouse an initial setting of measurement area can be performed and so on. QUEUE AND DELAY LENGTH MEASUREMENT ALGORITHM:

A monochrome CCD camera and an EE-lens is attached to correspond to brightness changes in outdoors. And Camera Unit and Control Unit are tightly closed to endure the outdoor uses, so we are thinking of circumstances endurance. The image signal from Camera Unit is inputted to Control Unit. And by processing the inside of measurement area set in advance, traffic flow information is measured. Then measurement results will be transferred to the Traffic

The basic idea of queue and delay length measurement algorithm is as following: "moving vehicles" from "existing vehicles" leaves "delay vehicles". In the distance from the camera position, it is hard to measure queue and delay length precisely because of poor visibility. But it is just noticed that each vehicle need not be recognized in the case of measuring queue and delay length and that it is effective to detect the vehicle groupings consisting of more than one vehicle in that case. Then the basic idea can be considered as, "Vehicle groupings of moving vehicles" from "vehicle groupings of existing vehicle" leaves "vehicle groupings of delay


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia vehicles". The algorithm is realized mainly by two processes, image preprocess and queue length measurement process.

which is strong in the noise by smoothing the calculated delay range by referring to the output at the time of back and forth. This process is composed of main five processes.

IMAGE PREPROCESS: They are: In the Image Preprocess, data is extracted in the sample point positions which were beforehand set like Figure 1 from the input image and make a Sampled-Points Image. If these sample points are set elaborately, whole computation quantity can be reduced while securing the data which is necessary to measure in precise. Next the status such as existing or moving of every sample point by the combination and by utilizing each characteristics of basic image processing methods such as time difference, background difference and spatial difference are detected. The outline of each image processing method TIME DIFFERENCE: It is the method to calculate a difference among two images in time t and time t+ alpha and detect change quantity in time alpha. And by binarizing the calculated change quantity of each sample point with appropriate threshold, the "move detected sampled-point" is detected. And by binarizing the calculated change quantity of each sample point with appropriate threshold, the ‗move detected sample-point‘ is detected. BACKGROUND DIFFERENCE: It is the method to calculate a difference between the input image and the background image which is a road surface image including no vehicles. By processing this method, a background difference imagewhich is only of Vehicles are made. SPATIAL DIFFERENCE: It is the method to calculate the change quantity of the brightness data with the neighbor pixel. By processing this method for the background difference image and by binarizing the calculated change quantity of each sample point with appropriate threshold, edges of vehicles can be detected. So the "existence detected sampled-point" is detected

Moving vehicle grouping detection process: In case of the time difference process, for a sampled point with small change quantity, a move detected sampled pointcan not be detected and the crack sometimes occurs in the vehicle body. The detection of the noise is also seen. Therefore, the body range of the moving vehicle must be made clear and the noise must be removed. But to recognize each vehicle as the vehicle grouping in case of queue and delay lengthmeasurement is effective. So we a concept of the vehicle grouping BLOCK is introduced. That is, BLOCK has the size which was beforehand set according to the distance from the camera and so on. And when the number of movingdetected sampled-points in this BLOCK is above the appropriate threshold value, the BLOCK is registered as the Moving BLOCK. By making BLOCK, the body range of the moving vehicle is made clear and the noise is removed. It is also possible to detect the moving vehicle grouping which stride two lanes by setting BLOCK to the optional position. DELAY BLOCK DETECTION PROCESS: Using the technique like the detection of the Moving BLOCK, this process detects the Delay BLOCK out of the delay detected sampled-points. DELAYRANGE ALCULATION PROCESS: This process calculates the delay range of each lane. We regard the detection range of Delay BLOCK every lane as the congestion range at the lane. DELAYRANGE ORRECTION PROCESS: Traffic congestion has the characteristics that it doesn't suddenly emerge or break off, and it never cancels from the queue end. Therefore, this process makes the output of the measurement result stable by smoothing the uneasy change of the delay range or by replacing the impossible change by Referring to the output at the time of back and forth

QUEUE LENGTH MEASUREMENT PROCESS: In this process, moving vehicle grouping is detected and delay vehicle grouping from the results of Image Preprocess, and calculate the delay range. Also, a stable output is made

CHARACTERISTICS METHOD:

OF

MEASUREMENT

Now, the characteristics of the Measurement Method are in the following.


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia 



 



By the processes of making BLOCK or smoothing, the influence of noise is reduced, and the precise and stable queue length measurement result can be obtained as output. By automatically calculating the parameter which depended on the spot according to the distance from the camera position and so on, it secures measurement precision in the distant place and the operation management such as adjustment or maintenance can be simplified. Because this method doesn't depend on the lane position, it is possible to detect the vehicle grouping which stretched for two lanes. With elaborate sampled-points position setting, we could reduce whole computation quantity while securing the data which is necessary to measure in precise. Therefore, it is possible to make a hardware scale small and the cost of the whole system can be reduced. The output delay time is so small that SPITS can provide traffic information to the traffic signal control in real time.

PROOF OF EFFICIENCY: FIELD TEST: A Field test was conducted with SPITS on the busy roads. In this experiment, the measurement items of UTMS specification, number of passed vehicles, speed, vehicle type, existence, queue length and delay length at the same time were measured. MEASUREMENT CONDITION: In this experiment, Camera Unit was installed in the position about 2m aside from the roadside and 10m above from the road surface near the stop line on the approach lane at the intersection. And the measurement area as the Range of 15165m from the stop line was set and 2 lanes. Experimentation was done under various conditions such as daytime, twilight, night, rain and so on. Measurement results were collected to the personal computer. Way of evaluating: A low-speed running vehicle is in congestion is considered. When the way of the congestion was moving and the congestion continued after that, the farthest position being in congestion is considered as the queue length.

The measured queue length is outputted every second and

the measurement precision with the maximum queue length each signal cycle is evaluated. In other words, measurement precision is calculated by the following computation formula. Measurement precision (%) = (The measurement value of the maximum queue length) / (The true value of the Maximum queue length) x 100 Here, the true value of the maximum queue length was measured by the direct watching in the spot. EVALUATION RESULT: Table 4 shows the average of the measurement precision in each measurement time and it shows the stable measurement precision within 10% error. Also, Figure 5 shows the measurement result of the queue length every second and the true value of the maximum queue length every signal cycle in 15:45-16:15, and confirms that SPITS has measured the queue length precisely. As the main error factor, the detection failure of a ashy black car whose brightness difference from the road surface is small and the lightless car in night were seen, and also the detection failure when a small car is hidden in the rear of the large-sized vehicle was seen, too. And the factor that the precision decline in case of the rainy weather on august 2nd is thought to be the decline of visibility because of the headlight reflection at the wet road surface CONCLUSION: Thus it is observed that, SPITS is able to measure queue and delay length precisely even in the distance where visibility is poor with a sophisticated method and processes in real time with a high-speed hardware. Also, the experimental results of field tests show that SPITS has achieved accurate


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia measurement within 10% error of maximum queue length under various conditions (day, night, twilight, rain etc.). Also, in the other measurement items, too, the high measurement precision can be achieved. And simplification in case of practice use is realized, so SPITS is considered to reach a practical use step. REFERENCES: Digital signal processing: Principles, algorithms and applications by J.G.Proaxis and D.G REFERENCES [1] Y. Rui and T. S. Huang, ―Image retrieval: Current techniques, promising directions and open issues,‖ J.. Vis. Commun. Image Represent., vol. 10, pp. 39–62, 1999. [2] A. W.M. Smeulders, M. Worring, S. Santini, A. Gupta, and R. Jain, ―Content-based image retrieval at the end of the early years,‖ IEEE Trans. Pattern Anal. Mach. Intell., vol.22, no.12, pp.1349–1380, Dec.2000. [3] S. S. Durbha and R. L. King. Knowledge mining in earth observation data archives: a domain ontology perspective. In Proceedings of IEEE International Geoscience and Remote Sensing Symposium, volume 1, September 2004. [4] D. A. Landgrebe. Signal Theory Methods in Multispectral Remote Sensing. John Wiley &Sons, Inc., 2003. [5] G. G. Wilkinson. Results and implications of a study of fifteen years of satellite image classification experiments. IEEE Transactions on Geoscience and Remote Sensing, 43(3):433– 440, March 2005. [6] M. Datcu, H. Daschiel, A. Pelizzari, M. Quartulli, A. Galoppo, A. Colapicchioni, M. Pastori,K. Seidel, P. G. Marchetti, and S. D‘Elia. Information mining in remote sensing image archives: system concepts. IEEE Transactions on Geoscience and Remote Sensing, 41(12):2923–2936, December 2003. [7] R. M. Haralick and L. G. Shapiro. Computer and Robot Vision. Addison-Wesley, 1992. [8] R. L. Kettig and D. A. Landgrebe. Classification of multispectral image data by extraction and classification of homogeneous objects. IEEE Transactions on Geoscience Electronics, GE- 14(1):19–26, January 1976 [9] B. S. Manjunath and W. Y. Ma. Texture features for browsing and retrieval of image data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(8):837–842, August 1996.

[10] S. Aksoy and R. M. Haralick. Feature normalization and likelihood-based similarity measures for image retrieval. Pattern Recognition Letters, 22(5):563–582, May 2001. [11] M. Schroder, H. Rehrauer, K. Siedel, and M. Datcu. Interactive learning and probabilistic retrieval in remote sensing image archives. IEEE Transactions on Geoscience and Remote Sensing, 38(5):2288–2298, September 2000. [12] C. M. Bishop. Neural Networks for Pattern Recognition. Oxford University Press,1995. [13] A. J. Cand, U. de Mendon Braga Neto, and E. C. de Barros CarvalhoFilho, ―A mathematical morphology approach to the star/galaxy characterization,‖ Journal of the Brazilian Computer Society 3, April 1997. [14] H. Heijmans and J. Goutsias, ―Nonlinear multiresolution signal decomposition schemes. ii. Morphological wavelets,‖ IEEE Transactions on Image Processing 9, pp. 1897 – 1913, Nov. 2000. [15]P. A. Maragos, ―A representation theory for morphological image and signal processing,‖ IEEE Transactions on Pattern Analysis and Machine Intelligence II(6), 1989. [16]T. Ikenaga and T. Ogura, ―Real-time morphology processing using highly parallel 2-d cellular automata cam/sup 2/,‖ IEEE Transactions on Image Processing 9, pp. 2018–2026, Dec. 2000. [17] A. J. Cand, U. de Mendon Braga Neto, and E. C. de Barros CarvalhoFilho, ―A mathematical morphology approach to the star/galaxy characterization,‖ Journal of the Brazilian Computer Society 3, April 1997. [18]H. Heijmans and J. Goutsias, ―Nonlinear multiresolution signal decomposition schemes. ii. Morphological wavelets,‖ IEEE Transactions on Image Processing 9, pp. 1897 – 1913, Nov. 2000. [19]P.Dhivya and M.Rajakani “Location Based Anonymous Routing Protocol for MANETS‖, IJIRSET, Volume 3, Issue 3, March 2014. [20]Taehong Kim, SeongHoon Kim, Jinyoung Yang, SeongeunYoo, Member, IEEE, and Daeyoung Kim, ―Neighbor Table Based Shortcut Tree Routing in ZigBee Wireless Networks‖, IEEE, Vol 5, Issue 3, March 2014.



Design and Mode Analysis of Patch Antenna for Radar Applications T.Mary Neebha Assistant Professor Dept. of ECE Karunya University, Coimbatore M.Nesasudha Associate Professor Dept. of ECE Karunya University Coimbatore

Abstract In this paper three different antennas exhibiting two different modes i.e. TM120, and ZOR (zero order resonance) are analyzed and a hybrid mode antenna by combining two modes is introduced. Hybrid patch slot antenna gives TM120 mode and the mushroom antenna gives ZOR mode. The modes combined are TM120 and ZOR. The resulting antenna is low profile and multi band. It works in the frequency range of 5 to 25GHz.The TM120 mode antenna, gives low profile due to a CELC (Complementary Electric Resonator) structure etched on the ground. The mushroom antenna exhibits ZOR mode and multi band. So for the hybrid antenna the radiating element is mushroom structure and on the ground CELC structure is etched. A major drawback of this design is that backward radiation is induced by the CELC structure etched, which reduces the far field gain, so a cavity is added under the CELC structure to overcome this drawback. The hybrid mode antenna can be used for Radar applications. Radars demand low profile and light weight antenna subsystems. Also, here, a single line fed square microstrip antenna with truncated corners is designed with TM120 mode. The software used is CADFEKO 5.5. Index terms- low profile, TM mode, multi band.

I.INTRODUCTION Microstrip antennas offers many advantages like low profile, light weight, easy integration with microwave devices in planar or non planar surfaces, inexpensive to manufacture. These advantages make them quiet popular among large number of users. Dual band or multi band operations are also required for satellite communication and many wirelesses communication systems [1]. To meet these requirements microstrip antennas can be used. Meandering the patch gives miniaturization [2] - [3] but it makes the process of manufacturing more complex and conformability with shaped surfaces is lost in this case. Another approach for size reduction is to print the radiating patch on a substrate with high dielectric constant [4], or using artificial materials [5]. It degrades the substrate integrity. It also results in antenna with large radiation loss and narrowed band width. It is found that meandering slots in the antenna ground plane give miniaturized antenna [6]. CML (complementary

meander line) structure etched on the ground beneath the patch, this antenna shows TM120 mode. The meander slot used in this paper is a complementary electric resonator (CELC) structure [7]. Which is a periodic structure etched on the metallic surface. The periodic structure behaves like a metamaterial. Metamaterial is a material whose property depends on its structure. The etching of CELC structure in the ground results in antenna with low profile and high gain. The CML structure induces the TM120 mode. ZOR mode is a unique property of metamaterial [8]. The mushroom structure gives the ZOR mode [9]. The placement of mushroom structure at the center of the geometry gives ZOR mode [10].ZOR antennas are available at single resonance, but it can also give multiple or dual [11] resonance. An equivalent horizontal magnetic The mushroom structure gives the ZOR mode [9]. The placement of mushroom structure at the center of the geometry gives ZOR mode [10].ZOR antennas are available at single resonance, but it can also give multiple or dual [11] resonance. An equivalent horizontal magnetic



patch introduces different modes. The newly proposed antenna is called hybrid mode antenna because it is a combination of both TM120 and ZOR mode. In this paper first of all the design of hybrid patch/slot antenna which gives TM120 mode and mushroom antenna which gives ZOR mode are analyzed. For the TM120 mode antenna the radiating element is a rectangular patch and on the ground CML structure is etched. For the ZOR mode antenna the radiating element is the mushroom structure. In order to combine the two modes and design hybrid mode antenna, the rectangular radiating element is replaced with the mushroom structure. The substrate thickness is kept the same as that of the TM120 mode, also the CML structure etched on the ground is retained. The CML structure etched on the ground below the patch results in backward radiation. As a result the gain of the designed antenna is very small. In order to reduce backward radiation and to further improve the gain. The antenna is loaded with a cavity, i.e. a cavity of size equal to that of the CML structure and height .5mm is added under the CML structure.

Fig.2. The geometry of the patch antenna with etched meander line slots at the ground. The second antenna is a mushroom antenna. The geometry of the antenna is shown in Fig.3.The dimensions of the antenna are Wm=6.4mm, Lm=3.4mm, gm=.2mm and the radius of the circle is 2.2mm. The dielectric constant is 2.2.the feed used is line feed at the center of the circle. Meshing is done with the suggested value

II.ANTENNA DESIGN The first antenna is a hybrid patch slot antenna with etched CML structure at the ground. The CML structure is shown in Fig.1.

Fig.1.Complementry meander line(CML) structure etched on the ground. thickness of .8mm is used. It has a dielectric constant of 3.38, and tanδ=0.0007 respectively. The feed used is a line feed at 3mm in the y-axis. Meshing is done with the suggested value.

Fig.3.The geometry of mushroom antenna Fig.4 shows the hybrid mode antenna. The dimension of the radiating element i.e. the mushroom structure is 24 Χ 3 mm2 for the rectangle and circle is of radius 3mm.The substrate dimension and thickness is same as that of the hybrid patch slot antenna, 60 Χ 55 mm2 and thickness of .8mm dielectric constant is

450



3.38, and tanδ=0.0007 The dielectric constant is also the same. The feed used is a line feed. Meshing is done with the suggested value. The resultant antenna is a low profile multi band antenna.

where Lt is the dimension of the truncation, Qt being the total quality factor, f0 the operation frequency and fr the resonant frequency of the square microstrip antenna. The formula for calculating the resonant frequency for a square microstrip antenna is given as : fr = c / 2Le √ εr (3)

where c is the velocity of electromagnetic waves in free space, and Le the effective length of the square patch. Given next in Figure 6, is the CADFEKO model of a conventional square patch antenna with specifications: length of square patch = 31mm, length of the substrate = 50mm, height of substrate =

Fig.4. The geometry of the patch antenna with etched meander line slots at the ground.

2.87mm, εr = 2.2, and the frequency = 6GHz. In Figure 6, is given the CADFEKO model of a square microstrip antenna with truncated corners with specifications: length of square patch = 31mm, length of the substrate = 50mm, length of truncated corner = 7mm, height of substrate = 2.87mm, εr = 2.55,and frequency = 6GHz.

The configuration of single feed circularly polarized square microstrip antenna, with truncated corners is exhibited in Fig. 5.The square patch has a dimension of L × L which is etched into a dielectric substrate with a thickness of H, loss tangent of tan δ, and relative permittivity of εr .The pair of patch corners is symmetrically truncated in order to produce circularly polarized radiations. The conditions of circularly polarized operation of the single-feed microstrip antenna with truncated corners, can be arranged as follows:

Lt = L / √ 2Qt (1) f0 = fr ( 1+ 1 / 4Qt ) (2)

451

Fig.5 CADFEKO model for square microstrip antenna



Fig.8. S-parameter graph Fig. 6. CADFEKO model for square microstrip antenna with truncated corners

The S-parameter of hybrid patch/slot antenna is shown in Fig.8.A dip of -9.86dB is observed at 2.44GHz.

III.RESULTS

A. Hybrid patch slot antenna The hybrid patch/slot antenna (TM120) gives the following results. The plots of far field gain (6.1dB), S-Parameter (-9.86dB), electric and magnetic current distributions (TM120) mode are given.

Fig.9. Electric current distribution (TM120 mode) The electric current distribution of TM120 mode in hybrid patch/slot antenna is shown in Fig.9.

Fig.7. Far field gain pattern The far field gain pattern of hybrid patch/slot antenna is shown in Fig.7. A gain of 6.1dB is observed at 2.44GHz. 452



Fig.10. Magnetic current distribution (TM120 mode)

The magnetic current distribution of TM120 mode in hybrid patch/slot antenna is shown in Fig.10. B. Mushroom antenna The designed mushroom antenna (ZOR) gives the following results. The plots of far field gain (4.6dB), S-Parameter (-4.7dB), vertical and horizontal polarization and electric and magnetic current distributions (ZOR) mode are given below.

Fig.12. S-parameter graph The Fig.12 shows the S-Parameter graph of mushroom antenna. A dip of -4.7dB is observed at 2.44GHz.

Fig.11. Far field gain pattern

Fig.13. Electric current distribution (ZOR mode)

The Fig.11 shows the far field gain pattern of mushroom antenna. The figure shows a value of 4.6dB at 2.44GHz.\

453

The mushroom antenna gives an electric current distribution as shown in Fig.13 for the ZOR mode



The Fig.15 shows the far field gain pattern of mushroom antenna. The figure shows a maximum gain of 0dB.

Fig.14. Magnetic current distribution (ZOR mode) The Fig.14 shows the magnetic current distribution of mushroom antenna.

C. Hybrid mode antenna The designed hybrid mode antenna, combination of TM120 mode and ZOR mode gives the following results. The plots of far field gain (4.6dB), S-Parameter (-32.277dB at 5.48GHz, -8.8dB at 9.28GHz,-9.85dB at 14.31GHz, -6.1dB at 20.34 GHz) electric and magnetic current distributions of hybrid mode are given below.

Fig.16. S-parameter graph The Fig.16 shows the S-Parameter graph of hybrid mode antenna. A dip of -32.277dB at 5.48GHz, -8.8dB at 9.28GHz,-9.85dB at 14.31GHz, 6.1dB at 20.34 GHz is observed.

Fig.15. Far field gain pattern

454

Fig.17. Electric current distribution (ZOR mode)



The designed hybrid mode antenna, combination of TM120 mode and ZOR mode gives the following results. The plots of far field gain (3 dB), S-Parameter (-33.589dB at 5.48GHz, -6.36 dB at 9.43 GHz,-6.37 dB at 14.49GHz, -22.8 dB at 21.08GHz) electric and magnetic current distributions of hybrid mode are given below.

Fig.18. Magnetic current distribution (ZOR mode)

For the cavity loaded hybrid mode antenna, the following results are observed

The designed hybrid mode antenna, combination of TM120 mode and ZOR mode gives the following results. The plots of far field gain (3 dB), S-Parameter (-33.589dB at 5.48GHz, -6.36 dB at 9.43 GHz,-6.37 dB at 14.49GHz, -22.8 dB at

Fig.20. S-parameter graph

21.08GHz) electric and magnetic current distributions of hybrid mode are given below.

The Fig.20 shows the S-Parameter graph of hybrid mode antenna. A dip of -33.589dB at 5.48GHz, -6.36 dB at 9.43 GHz,-6.37 dB at 14.49GHz, -22.8 dB at 21.08GHz. The designed hybrid mode antenna, combination of TM120 mode and ZOR mode gives the following results. The plots of far field gain (3 dB), S-Parameter (-33.589dB at 5.48GHz, -6.36 dB at 9.43 GHz,-6.37 dB at 14.49GHz, -22.8 dB at 21.08GHz) electric and magnetic current distributions of hybrid mode are given below.

Fig.19. Far field gain pattern The Fig.19 shows the far field gain pattern of cavity loaded hybrid mode antenna .The figure shows a maximum gain of 3dB.

455

The Fig.19 shows the far field gain pattern of cavity loaded hybrid mode antenna .The figure shows a maximum gain of 3dB.



Fig.21. Electric current distribution

Fig. 24.VSWR graph of a square microstrip antenna with truncated corners

IV. CONCLUSION

Fig.22. Magnetic current distribution

Fig.21 and Fig.22 shows the electric and magnetic current distributions respectively. There is no difference between the electric and magnetic current distribution of hybrid patch slot antenna without cavity loading.

Fig.23 and Fig. 24 shows the S11 parameter and VSWR graph for truncated corner microstrip patch antenna. The results show that this antenna can also be used for Radar applications. 456

In this paper a hybrid patch slot antenna with TM120 mode and a mushroom antenna with ZOR mode is analyzed and a hybrid mode antenna is designed, which is a combination of TM120 mode and ZOR mode. The CML structure etched on the ground gives the TM120 mode and the mushroom structured radiating element produces the ZOR mode. The advantage of TM120 mode antenna is that it is low profile. The advantage of ZOR mode antenna is that it is multiband. The hybrid mode antenna combines both these advantages. The CML structure etched on the ground results in backward radiation thus reducing the gain of the antenna. So a cavity is added right behind the CML structure to reduce backward radiation and increase gain. From the simulation results it is found that the cavity loaded hybrid mode antenna has low reflection coefficient, higher gain and higher far field directivity compared to hybrid mode antenna without cavity. The hybrid mode is retained even when the cavity is added. The hybrid antenna operates in the frequency range 5GHz


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia to 25GHz with multiple bands of operation. Also a truncated corner microstrip patch antenna is also analyzed. These antennas are suitable for Radar communication. REFERENCES [1] Jiang Xiong, Hui Li, Yi Jin, and Sailing He." Modified TM0δ0 mode of a rectangular ratch antenna partially loaded with metamaterial for dual-band applications", IEEE Trans. Antennas Propag., Vol. 8, no. 8, pp. 1536-1225, 2009. [2] K. L. Wong and H. C. Tung, ―An inverted U- shaped patch antenna for compact operation,‖ IEEE Trans. Antennas Propag., Vol. 51, no. 7, pp.1647–1648, Jul. 2003. [3] J. S. Kuo and K. L. Wong, ―A dual-frequency L- shaped patch antenna,‖ Microw. Opt. Technol. Lett., Vol. 27, pp. 177–179, Nov. 2000. [4] B. Lee and F. J. Harackiewicz, ―Miniature microstrip antenna with a partially filled high- permittivity substrate,‖ IEEE Trans. Antennas Propag., Vol. 50, no. 8, pp. 1160– 1162, Aug. 2002. [5] H. Mosallaei and K. Sarabandi, ―Design and modeling of patch antenna printed on magneto- dielectric embeddedcircuit metasubstrate,‖ IEEE Trans. Antennas Propag., Vol. 55, no. 1, pp. 45–52, Jan. 2007. [6] J. S. Kuo and K. L. Wong, ―A compact microstrip antenna with meandering slots in the ground plane,‖ Microw. Opt. Technol. Lett., Vol. 29, no. 2, pp. 95– 97, Apr. 2001. [7] M.C Tang, S. Xiao, Yan-Ying Bai, Tianwei Deng, Changrong Liu, Yuping Shang, Chao Lei Wei, and BingZhong Wang," Design of hybrid patch/slot antenna operating in induced ―TM120 Mode‖, IEEE Trans. Antennas Propag., Vol 60. .no. 5.pp. 2157-2165, May 2012.

457

[8] A. Lai, C. Caloz, and T. Itoh, ―Composite right/lefthanded transmission line metamaterials,‖ IEEE Microwave Mag., vol. 5, no. 3, pp.34–50, Sep. 2004. [9] Seung-Tae Ko and Jeong-Hae Lee," Hybrid ZerothOrder Resonance Patch Antenna With Broad -Plane Beamwidth," IEEE Trans. Antennas Propag., Vol 60.no. 1.pp. 19-25,Jan 2013. [10] A. Sanada, C. Caloz, and T. Itoh, ―Planar distributed structures with negative refractive properties,‖ IEEE Trans. Microwave Theory Tech., vol. 52, no. 4, pp. 1252–1263, Apr. 2004. [11] Krushna A Munde and Veeresh G," A Zeroth- Order Resonance Patch Antenna for Dual Band Operations with Broad E-plane Beam Widths," International Journal of Computer Applications (0975 – 8887) Volume 103 – No.7, October 2014. [12] A. Lai, K. M. K. H. Leong, and T. Itoh, ―Infinite wavelength resonant antennas withmonopolar radiation pattern based on periodic structures,‖ IEEE Trans. Antennas Propag., pp. 868–876, Mar. 2007. [13] J. G. Lee and J. H. Lee, ―Zeroth order resonance loop antenna,‖ IEEE Trans. Antennas Propag., vol. 55, no. 3, pp. 994–997, Mar. 2007. [14] J. H. Park, Y.H.Ryu, J. G. Lee, and J. H. Lee, ―Epsilon negative zerothorder resonator antenna,‖ IEEE Trans. Antennas Propag., vol.n55, no. 12, pp. 3710–3712, Dec. 2007. [15] S. T. Ko and J. H. Lee, ―Miniaturized ENG ZOR antenna with high permeability material,‖ in IEEE APS/URSI, Canada, Jul. 2010.



PHOTOVOLTAIC BASED ASYMMETRICAL MULTILEVEL INVERTER P.Jamuna Research Scholar, Department of EEE Pondicherry Engineering College Puducherry, INDIA Dr.C.Christober Asir Rajan Professor, Department of EEE Pondicherry Engineering College Puducherry,INDIA Abstract: The unique structure of proposed multilevel converter circuit based on series/parallel connection of switches significantly reduces the number of switches and increases number of output voltage levels. Photovoltaic source model is simulated and applied as input for multilevel inverter Inverse Sine Pulse Width Modulation (ISPWM) is used to reduce of total harmonic distortion in the proposed multilevel inverter. This multilevel inverter is extended for the three phase power system .Thus the simulation of the proposed multilevel inverter is performed in the MATLAB/SIMULINK environment and the results are presented. Index Terms — AMI, ISPWM, THD

I. INTRODUCTION Power conditioning systems are often designed to supply an AC load from DC source. An inverter should provide constant and ripple free AC voltage to ensure the safety of the equipments .The design of such systems must achieve an output voltage behavior as close as possible to the ideal AC voltage in the sense of fast transient response to load variation and low THD. With the expansion of power electronics towards the medium voltage and high power applications in the past few years, multilevel power conversion is an emerging area. Multilevel voltage source inverter offers several advantages which makes it preferable over the conventional (two level) inverter. These include the possible utilization of higher DC link voltage, improved harmonic performance, reduced power device stress etc. Several multilevel topologies [1-2], namely the diode clamped multilevel inverter, the flying capacitors Multilevel Inverter (MLI) and cascaded multilevel inverter, have evolved and are applied in adjustable speed drives, electric utilities and renewable energy systems. Among the multilevel inverter topologies, cascaded multilevel inverter has more advantages than the other two. Cascaded MLIs use more than one DC voltage source to generate an AC output voltage that resembles a sine wave. Various modulation methods are available for multilevel inverter [3]. In recent years, asymmetric multilevel inverters have received increasing attention because it is possible to synthesize voltage waveforms with reduced harmonic content, even using a few series-connected cells. This advantage is achieved by using distinct voltage levels in different cells, which can create more levels in the output voltage and minimize its total harmonic distortion (THD) without increasing the number of switching devices and isolated

458

sources [4-7] and without using an output filter. When the number of output levels increases, harmonics of the output voltage and current decreases. The harmonics in the output voltage of power electronic inverters can be reduced using Pulse Width Modulation (PWM) switching techniques. Various control techniques are proposed for the cascaded MLI with single reference /multiple carrier and multiple reference /single carrier [8-13]. This paper presents a novel approaches for controlling the harmonics of output voltage in series/ parallel connected multilevel inverter employing PWM switching strategies. Photovoltaic source is a renewable energy source which is obtained from the radiation of sunlight and applied as input for multilevel inverter .Simulations are performed using MATLAB/SIMULINK.

II. MODELLING OF PV CELL The modeling and simulation of photovoltaic (PV) have made a great transition and form an important part of power generation in this present age. The most common model used to predict energy production in photovoltaic cell modeling is the single diode circuit model as shown in Fig. 1. The ideal photovoltaic module consists of a single diode connected in parallel with a light generated current source. The equation or the output current is given by:

…… (1)

I=ISC-ID Where,

ID=ISCref(exp

(qVoc/KAT)

-1)

…… (2)



In this proposed model, a current source ISC which depends on solar radiation and cell temperature; a diode in which the inverse saturation current Io depends mainly on the operating temperature; a series resistance Rs and a shunt resistance Rp which takes into account the resistive losses. The equation that describes the I-V characteristic of the circuit in Fig. 2 is given by;

Fig.1. Solar cell model using single diode

The light current depends on both irradiance and temperature. It is measured at some reference conditions. Thus,

ISC=[ISCref+Ki(Tk-Tref)]*ζ/100

.…. (3)

.….(4)

Isc-Id-Vd/Rp-Ipv=0

Thus, …..(5)

Ipv=Isc-Id-Vd/Rp Where, ISC is the photocurrent in (A) which is the lightgenerated current at the nominal condition (25oC and

And the reverse saturation current Irs Irs=ISCref(exp

is given as:

(qVoc/NsKAT)-

..…( 6)

1)

2

1000W/m ), Ki is the short-circuit current/temperature co-

The module saturation current Io varies with the cell temperature which is given by;

efficient at ISCref (0.0017A/K), Tk and Tref

are the actual and reference

3 qcg/AK

Io=Irs[(T/Tref) e

…..(7)

*(1/Tref-1/T)]

temperature in K, is the irradiation on the device 2 surface, and 1000W/m is the nominal irradiation. Where Io is the diode saturation current (A). The basic equation that describes the current output of the photovoltaic (PV) module Ipv of the single-diode model is as given in equation (8). Thus,

Ipv=NpIsc-NsIo{exp

(q(Vpv+IpvRs)/NsAKT)-

1}

Where , Fig .2.Solar cell using single diode with Rs and Rp

Equation (2) does not adequately represent the behavior of the cell when subjected to environmental variations, especially at low voltage. A more practical model is shown in Fig. 2, where Rs and Rp series and parallel resistance, respectively. 459

k is the Boltzmann constant (1.38 x 10-23 J K1), q is the electronic charge (1.602 x 10-19 C), T is the cell temperature (K);


.….(8)


Where, A is the diode ideality factor, Rs the series resistance (Ω) Rp is the shunt resistance (Ω). Ns is the number of cells connected in series Np is the number of cells connected in parallel, Vpv=V oc=21.06v .

The use of the simplified circuit model for this work makes it suitable for power electronics designers to have an easy and effective model for the simulation of Fig. 3 Simulink model of a PV cell photovoltaic devices with power converters. The value of the parallel resistance Rp is generally high and hence neglected to simplify the model. A procedure based on Simulink model to determine the values to these parameters is proposed. The evaluation of these model parameters at real condition of irradiance and temperature of the target PV modules are then determined according to their initial values. The simulation model is shown Fig. 3. The P-V and I-V characteristics are shown in Figures 4(a)–(d) for constant temperature and constant irradiance.

460



m = 2s +1, Where s is the number of dc sources. Each H-bridge requires a dc source and 4 switches. There are 12 switching devices for 7 levels, 16 switching devices for 9 levels, 20 switching devices for 31 levels, which makes an inverter complicated. But in proposed inverter the number of output phase voltage level (m) is given by m = 12n +7

Fig.4(a). P-V characteristic-varying irradiance-constant

temperature

Where n is number of voltage sources in series/parallel circuit. Total number of switches devices (D) required is given by D = 3n + 9 which makes the output voltage of the inverter almost sinusoidal. When the voltage ratio of 0.5:1:6 is assumed in the H-bridges, the inverter requires 14 switching devices for 31 levels, which is shown in Fig. 5. The switching sequence is shown in Table 1.

Fig.4(b). I-V characteristic-varying irradiance- constant

temperature.

Fig.4 ( c ) P-V characteristic-constant irradiance (1000W/sqm) varying temperature. III. PROPOSED ASYMMETRICAL MULTILEVEL INVERTER The topology of the modified multilevel inverter is shown in Fig. 5. It consists of three bridges, that is two upper Hbridges and the lower H-bridge with series/parallel circuit, which are cascaded. DC voltage sources are connected in series when the switches Sa1-San-1 becomes ON and Sb1-

Sbn-1, Sc1-Scn-1 becomes OFF. In parallel when the switches Sb1-Sbn-1 and Sc1- Scn-1 becomes ON, Sa1- San-1 becomes OFF. In series/parallel when the switches, Sa2San-1, Sb1, Sc1 becomes OFF, current flows in switches Sb2-Sbn-1, Sc2-Scn-1 via the switch Sa1. By using this principle high numbers of steps are obtained at output series parallel circuit. The number of output phase voltage levels in a conventional cascade inverter is defined by, 461

Fig.5. Circuit diagram of Asymmetrical series/parallel inverter


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia Table I: Conduction sequence for 31 level Inverter

Amplitude modulation index In amplitude modulation index, modulation index depends on the amplitude of carrier signal and the reference signal used. In general the value of modulation index varies between 0 and 1. Amplitude modulation index (ma) can be defined as

ma=Ar/(m-1)Ac where, Ar -Amplitude of reference signal Ac -Amplitude of carrier signal Frequency modulation index In frequency modulation index, the modulation index depends on the frequency of the reference signal and the carrier signal used. The frequency modulation index of the mf of the multilevel inverter can be defined as

mf=fc/fr

The above Table I shows the conduction sequence for the proposed 31 level inverter. Switching signals are generated based on the conduction sequence.

where, fc- frequency of carrier signal frfrequency of reference signal Based on ISPWM technique, switching pulses are generated for different switches in the inverter circuit. The 31 level output voltage waveform and THD spectrum is shown in Fig. 6 and Fig. 7.

IV.Inverse Sine Pulse Width Modulation (ISPWM) The inverted sine PWM (ISPWM) method uses the conventional sinusoidal as reference signal and an inverted sine as carrier signal that helps to maximize the output voltage for a given modulation index. The proposed control strategy replaces the triangular based waveform by inverted sine wave. The inverted sine PWM has a better spectral quality and a higher fundamental voltage compared to the triangular based PWM. For an ‗m‘ level inverter, (m-1) carrier waves are required. The pulses are generated when the amplitude of the modulating signal is greater than that of the carrier signal. The ISPWM strategy enhances the fundamental output voltage particularly at lower modulation index ranges. There is also a reduction in the total harmonic distortion (THD) and switching losses. The appreciable improvement in the total harmonic distortion attracts drive applications where low speed operation is required. The performance evaluation of inverted sine pulse width modulated inverter is done using MATLAB and the optimum switching frequency with minimized total harmonic distortion is determined.

462


International Convention on Dissemination of Innovative Research in Science, Engineering, Technology & Management Applications 11th-12th March 2017, Kuala Lumpur, Malaysia Fig.8. Simulink circuit of 3 phase multilevel inverter

The Proposed method is extended to the three phase system and the simulation model of a three phase inverter circuit and its output waveform are shown in Fig.8 &9.

Electrical systems are subject to a wide variety of power quality problems which can interrupt production processes, affect sensitive equipment, and cause downtime, scrap, and capacity losses. Momentary voltage fluctuations can disastrously impact production. Extended outages have an even greater impact. Voltage fluctuations can also impact the stability and reliability of utility transmission and distributions.

Fig.9. Output voltage of three phase 31 level inverter

Fig. 6. Output waveform of single phase 31 level inverter

Fig.7. Harmonic spectrum of 31 level inverter

463



Conclusion: This paper has dealt with the new series/parallel connection based asymmetrical multilevel inverter. The main achievement of the proposed PWM control technique is to reduce the total harmonic distortion (THD) and closer to sinusoidal waveform without the usage of an output filter. This inverter reduces the

switching losses by reducing number of switches and provides improved output voltage capability. Photovoltaic source model is modelled and act as input to the inverter .The inverter is simulated in MATLAB/SIMULINK package and the results are obtained.

VI. REFERENCES [1] Fang Zheng Peng, Jih-Sheng Lai, and Rodriguez ―Multilevel inverters: a survey of topologies, controls, and applications‖, Industrial Electronics, IEEE Transactions, Vol 49, issue:4, pp. 724-738, Aug 2002. [2] Peng F.Z., Lai J.S., ―Multilevel converters- A New breed of Power Converters‖, IEEE Transactions on Industry Applications, Vol. 32, No. 3, May-June 1996, pp.509-517.

[3] M.G.H.Aghdam, S.H.Fathi, B.Gharehpetian, ―Analysis of multicarrier PWM methods for asymmetric multilevel inverter‖ in Proc. 3rd IEEE Conference on Industrial Electronics and Applications, ICIEA‘08, June 2008, pp.20572062. J. Rodriguez, J.S. Lai, and F.Z. Zeng, ―Multilevel inverters:A survey of topologies, controls and application,‖IEEE Trans. Ind. Electron.,vol. 49,no. 4,pp. 724-738, Aug.2002. [4] S.Mariethoz and A. Rufer, ―Design and control of asymmetric multilevel inverters,‖ in Proc.IECON Conf., 2002, pp.840-845. [5] K. Corzine and Y. Familiant, ―A new cascaded multilevel H-bridge drive,‖ IEEE Trans. Power Electron., vol. 17, no. 1, pp.125-131, Jan. 2002. [6] R. Lund, M. D. Manjrekar, P. Steimer, and T. A. Lipo, ―Control strategies for a hybrid seven- level inverter,‖ in Proc. EPE Rec., 1999. [7]

J. Rodriguez, B.Wu, S. Bernet, J. Pontt, and S. Kouro,

―Multilevel voltage source converter topologies for industrial medium voltage drives,‖ IEEE Trans. Ind. Electron., vol. 54, no. 6, pp. 2930–2945, Dec. 2007. [8] H. Abu-Rub, J. Holtz, J. Rodriguez, and G. Baoming, ―Medium voltage multilevel converters—State of the art, challenges and requirements in industrial applications,‖ IEEE Trans. Ind. Electron, vol. 57, no. 8, pp. 2581–2596, Aug. 2010. [9]

M. T. Haque, ―Series sub-multilevel voltage source inverters

(MLVSIs) as a high quality MLVSI,‖ in Proc. SPEEDAM, 2004, pp. F1B-1–F1B-4.

464



465


Genetic Algorithm Based Closed Loop Control for

Genetic Algorithm Based Closed Loop Control for

Suggest Documents

Crowdsourcing for closed-loop control

A fuzzy logic based closed-loop control system for ...

Identification-Based Closed-Loop Control Strategies for ... - IEEE Xplore

Dynamic and Closed-Loop Control

Vision-based closed-loop control of mobile microrobots for ... - CiteSeerX

Vector Based Kinematic Closed-Loop Attitude Control-System for ...

Lidar-based wake tracking for closed-loop wind farm control

Closed-loop Control for Precision Antimicrobial ...

PREDICTIVE CLOSED LOOP POWER CONTROL FOR MOBILE ...

Microcontroller based closed-loop automatic speed control of ... - wseas

Genetic Algorithm Based High Performance Control ...

Microcontroller based Closed Loop Speed Control of DC Motor using ...

Wireless Magnetic-Based Closed-Loop Control of Self - PLOS

Physics-Based Closed-Loop Control of Phasing, Peak Pressure and ...

Closed-loop separation control using machine learning

Supplementary material: Closed-loop control of

Mining Requirements from Closed-Loop Control Models

Closed-Loop Learning of Visual Control Policies

Closed-Loop Control of Carbon Dioxide ...

Process automation and closed-loop control technology

Command Shaped Closed Loop Control of Flexible

Closed Loop Control of Soft Switched Interleaved

CLOSED-LOOP IDENTIFICATION AND CONTROL ...

Title A closed-loop type algorithm for determination of ... - CiteSeerX