i

MODELING OF PHOTOVOLTAIC (PV) MODULE TEMPERATURE BASED

ON AMBIENT FACTOR IN MALAYSIA USING ANFIS

NUR FARHANAH BT WAKIMAN

This project report presented in partial fulfilment of the requirements

for the degree of

Bachelor of Electrical Engineering

Faculty of Electrical and Electronic Engineering

University Tun Hussein Onn Malaysia

JULY 2012

v

ABSTRACT

This paper introduces a model build using Adaptive Neuro-Fuzzy Inference System

(ANFIS) for evaluation of temperature for PV modules. The input of this model were

taken from meteorological data which are ambient temperature,Ta, solar

irradiation,GT, wind speed,Vw and humidity,RH. These parameters were evaluated

from outdoor exposure data measured at Malaysia Green Technology Corporation

(MGTC), Bandar Baru Bangi, Malaysia. The model was validated based on low

training error and accepted validation error.

Keywords— PV Module Operating Temperature, Meteorological data, ANFIS.

vi

ABSTRAK

Penulisan ini memperkenalkan pembinaan model menggunakan ‘Adaptive Neuro-

Fuzzy Inference System (ANFIS) model untuk menilai suhu pada panel

fotovoltaik(PV). Data masukan diambil dari data meteorologi yang mana antaranya

ialah suhu persekitaran, Ta, pancaran cahaya matahari, GT, kelajuan angin, Vw, dan

kelembapan, RH. Parameter yang terlibat telah dinilai dari pendedahanluar yang

diukur pada PV panel yang telah dipasang di bumbung Mlaysia Gree Technology

Corporation (MGTC), Bandar Baru Bangi, Malaysia. Model ini dinilai berdasarkan

kerendahan kesalahan ketika latihan dan nilai kesalahan yang boleh diterima ketikan

penyemakkan.

Keywords— PV Module Operating Temperature, Meteorological data, ANFIS.

vii

TABLE OF CONTENTS

CHAPTER TITLE PAGE

1

2

SUPERVISOR VALIDATION

TITLE

DECLARATION

DEDICATION

ACKNOWLEDGEMENT

ABSTRACT

ABSTRAK

TABLE OF CONTENTS

LIST OF FIGURES

LIST OF TABLES

LIST OF ABBREVIATIONS

LIST OF APPENDICES

INTRODUCTION

1.1 Introduction

1.2 Problem Statement

1.3 Objectives

1.4 ProjectScope

1.5 Thesis Outlines

LITERATURE REVIEW

i

ii

iii

iv

v

vi

vii

viii

xii

xiii

xiv

1

3

4

5

5

viii

3

4

2.1 Introduction

2.2 Technology Developments

2.2.1 Modeling of The Nominal Operating Cell

Temperature Based on Outdoor Weather

2.2.2 Assessment of PV Cell Performance under

Actual Operating Condition

2.2.3 Operating Temperature of PV modules: A

Survey of Pertinent Correlations

2.3 Project Review

2.3.1 Photovoltaic System in Malaysia

2.3.1.1 Operating Temperature

2.3.1.2 Ambient Temperature

2.3.1.3 In-plane Irradiation

2.3.1.4 Wind Speed

2.3.1.5 Dew-Point Temperature

2.4 Conclusion

METHODOLOGY

3.1 Introduction

3.2 Project Review

3.3 Artificial Neuro-Fuzzy Inference System

3.3.1 Layer 1

3.3.2 Layer 2

3.3.3 Layer 3

3.3.4 Layer 4

3.3.5 Layer 5

3.4 Application of ANFIS

3.5 Conclusion

RESULTS AND ANALYSIS

7

8

8

9

10

11

11

12

13

13

14

15

15

16

18

20

22

23

23

24

24

25

29

ix

5

4.1 Introduction

4.2 ANFIS Method

4.3 Trained Module

4.4 Data Training without Normalized Process

4.4.1 Loading Data

4.4.2 Generate FIS

4.4.3 Evaluate FIS System

4.4.4 Training Data

4.4.5 Performance

4.4.6 Plotting

4.5 Data Training with Normalized Process

4.5.1 Loading Data

4.5.2 Normalized Process

4.5.3 Generate FIS

4.5.4 Data Training

4.5.5 Performance

4.5.6 Plotting

4.5.7 De-normalized Process

4.6 Conclusion

CONCLUSION

5.1 Conclusion

5.2 Recommendations

REFERENCES

VITA

APPENDIXES

30

31

31

32

32

33

36

37

39

41

44

44

45

48

51

56

58

61

64

65

66

67

70

71

x

LIST OF FIGURES

FIGURE. NO TITLE PAGE

1.1

3.1

3.2

3.3

3.4

3.5

3.6

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

4.14

Solar panels that contain solar cells

Flowchart for overall process

Location of investigated PV

Input Data need to be train

Rule view of two rule Sugeno system

An ANFIS architecture for a two rule Sugeno system

ANFIS major step flowchart

Command to load data

Generating FIS

Ruleview for fisa network

Evaluate command

Training data command

Command used to check error performance

Plotting command

Graph of real and training output

Graph of real and checking output

Command to load data

Command to normalized data

Flowchart for normalized process

Training and checking input

Generating FIS

3

17

18

20

21

22

25

32

33

36

36

37

39

41

41

42

44

46

47

48

49

xi

4.15

4.16

4.17

4.18

4.19

4.20

Command for training data

Evaluated network performance

Plotting command

Graph of real and training output

Graph of real and checking output

De-normalized command

51

57

59

60

60

61

xii

LIST OF TABLES

TABLE. NO TITLE PAGE

3.1

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

System Specifications and Site Descriptions

Structure of fisa network

Structure for fisa1 network

Performance results

Real target and training output data

Structure of fisa network

Structure of fisa1 network

ANFIS info for fisa1

Structure of fisa2 network

Structure of fisa3 network

ANFIS info for fisa2 and fisa3

Performance results

Real target and training output data

Real target and checking output data

19

35

38

30

43

50

52

53

54

55

56

58

62

63

xiii

LIST OF ABBREVIATIONS

PV - Photovoltaic

BIPV - Building Integrated Photovoltaic

Tc - Operating Temperature

Ta - Ambient Temperature

GT - In-plane Irradiation

RH - Relative Humidity

Vw - Wind Speed

Td - Dew Point Temperature

FIS - Fuzzy Inference System

ANFIS - Artificial Neuro Fuzzy Inference System

kWh - kilo Watt hour

PHANTASM Photovoltaic Analysis and Transient Simulation method

MECM Ministry of Energy Communication and Multimedia

Voc Open Circuit Voltage

xiv

LIST OF APPENDICES

APPENDIX TITLE PAGE

A Gant chart 71

B

C

Rawdata

Programming

Result data

72

83

D 89

REFERENCES

[I] J. Arrillaga, High Voltage Direct Current Transmission, 1998.

[2] N. Yahaya, Building Integrated Photovoltaic (BZPV) Technology

Application Project, 2004.

[3] L. C. Haw, "In Building for Malaysia : Prototype Solar House,".

[4] T.Erge (on behalf of NLCC Architects) C. Reise, Solar Irradiation and

Energy Yieldsfor Photovoltaic Systems in Kuala Lumpur, Januasy 2002.

[5] J.A Palyvos E. Skoplaki, "Solar Energy," On the temperature dependence

ofphotovoltaic module electrical performance : A review of eficiency power

correlations, 2009.

[6] A.G.Boudouvis, J.A Palyvos E.Skoplaki, "Solar Energy Materials & Solar Cells,

" A simple correlation for the operating temperature o~photovoltaic modules of

arbitrary mounting, 2008.

[7] J.A Palyvos ESkoplaki, "Renewable Energy," Operating Temperature of Photovoltaic

Modules: A survey ofpertinent correlations, 2009.

[8] M. N. Taib, S. M. S. B. Z. M. Zain, Hot and humid climate: Prospect for thermal comfort in residential building, 2006.

191 M. G. Lawrence, "American Meteorologicak Society," The Relationship

between Relative Humidity and the Dew point Temperature in Moist Air,

A Simple Conversion and Applications, February 2005.

[lo] College of Engineering and Applied Science, University of Colorado at

Boulder Contributed by: Integrated Teaching and Learning Program. (2010, December)

Lesson: The Temperature Effect. [Online]. http://www.teachengineering.org/view-lesson.php~l=h~p:// www. teachengineering.org/collectiodcub~/lessons/cub~veff/cub~ pveff-lesson02.xml#intro

[ll] M. Heck and S. W. M. Koehl, Evaluation of the Accelerated Life Testing Conditions for PV-Modules Based on Measured and Simulated Weathering Stress.

[12] M. N.Khalid , and M. H. Yusri , A. R. Hasimah , Assesment ofPVCell Performance Under Actual Malaysia Operating Condition.

[13] E Lee, H.K. Lim,M. F. Sepikit, M.R.M. Maskum, M. F. Ahrnad, and M. A. Mahmood N.M. Maricar, "Photovoltaic Solar Energy Technology Overview for Malaysia Scenario," in National Power and Energy Conference (PECon), Bangi Malaysia, 2003.

[14] S. V. M., A. M. Dastgheib, J. T. H. N. Afrouzi, Economic Sizing of Solar Array for A Photovoltaic Building in Malaysia with MATLAB.

[15] S. Shaari,A. M. Omar,S. I. S., Z. Hedzlin , "Operating Temperature Correlation with Ambient Factors of Building Integrated Photovoltaic (BIPV) Grid-Connected (GC) Systems in Malaysia," in Praise Worthy Prize S.r. l., August 201 1, pp. Vo1.4, N.4.

1161 S. Shaari, A. M. Omar, S. I. Sulaiman, Z. Hedzlin, "Power Prediction for Grid - Connected Photovoltaic System in Malaysia," in ZEEE, Melaka, Malaysia, 1-3 June 2011, pp. 110-113.

1171 A. Cruz. (2012, May) ANFIS- Adaptive Neuro Fuzzy Inference System. [Online]. eauipe.nce.ufri. br/adrianolfuzz~ltran~~arenciaslanfi~lanfis.~df

[IS] J. Shing, R. Jang, "IEEE Transaction on System," ANFIS- Adaptive Network Based Fuzzy Infireme System, vol. 23, MayIJune 1993,

[19] M. Setak,and M. J. Tarokh Hossein Abbasimehr, "International Journal of Computer Applications," A Neuro-Fuzzy Classifier for Customer Churn Prediction, vol. 19, April 201 1.

[20] Inc The MathWorks. (1984-2012, march) mathworks.com. [Online]. htt~://www.mathworks.com/hel~ltoolbox/h~zy/genfi~2~html

[2 11 Mathwork.Inc. (1 984-20 12) Mathwo1k.com. [Online]. http:Nwww.mathworks.com/help/toolbodh~~~/eva1fi~~html

[22] Mathwork.Inc. (1984-2012) Mathwork.com. [Online]. http://www.mathworks.comkelp/toolbox/f~~~ylanfi~~html

[23] Mathwork.Inc. (1984-2012) Mathwork.com. [Online]. ht tp:~lwww.mathworks.com/help/toolbox/Eul

1241 Mathwork Inc. (1984-2012) Mathwork.com. [Online]. h t tp : l lwww.mathworks .com/he l~ / too lbox/ fuz l

[25] Inc. The MathWorks. (1984-2012, March 2012) Mathworks. [Online]. htt~:llwww.mathworks.com/help/toolbox/fu~yl~enfis2.html