Improved ANN based impedance estimator for Phase to ground faults in UHV Transmission line

目录

声明

ABSTRACT

TABLE OF CONTENT

LIST OF FIGURES

LIST OF TABLES

CHAPTER 1 INTRODUCTION

1．1 NECESSITY AND IMPORTANCE OF RESEARCH ON THE TOPIC

1．1．1 AREA OF APPLICATION

1．1．2 Technical Advances

1．1．3 Numerical Distance Protection

1．2 RESEARCH STATUS ABOUT THE TOPIC

1．3 MAIN POINTS OF THE RESEARCH

CHAPTER 2 DIGITAL DISTANCE PROTECTION METHODS

2．1 INTRODUCTION

2．2 LINE TO GROUND FAULTS ON DOUBLE FED SYSTEM

2．3 CONVENTIONAL GORUND DISTANCE RELAYING METHOD

2．4 COMPREHENSIVE NEGATIVE-ZERO SEQUENCE CURRENT METHOD

2．5 MODIFIED CONVENTIONAL SCHEME

CHAPTER 3 ARTIFICIAL NEURAL NETWORKS

3．1 INTRODUCTION

3．2 ANN IN POWER SYSTEM PROTECTION DOMAIN

3．3 APPROPRIATE ARCHITECTURE FOR ANN BLOCK

3．4 TRAINING PATTERN

3．4．1 IMPROVED LEARNING(BACKPROPAGATION ALGORITHM)

3．4．2 BATCH GRADIENT DESCENT

3．4．3 LEVENBERG-MARQUARDT ALGORITHM

3．4．4 SPEED AND MEMORY COMPARISON

3．5 STEPS OF DESIGNING AND TRAINING

3．5．1 MACHINE LEARNING ASSISTANCE FOR APPROXIMATE IMPEDANCE PREDICTION

3．5．2 INPUTS AND OUTPUTS SELECTION CRITERIA

3．5．3 ANN BLOCK ARCHITECTURE

CHAPTER 4:SIMULATION RESULTS AND DISCUSSION

4．1 INTRODUCTION

4．2 TRANSMISSION LINE MODEL

4．3 PSCAD MODEL

4．4 FLOW CHART

4．5 SINGLE PHASE TO GROUND FAULT

4．5．1 VARIATION IN POWER TRANSFER ANGLE FOR DIFFERENT FAULT LOCATIONS

4．5．2 Variation In Fault Resistance Rf

4．5．3 Simultaneous Open Conductor Fault Impedance

CHAPTER 5 CONCLUSION

APPENDIX

REFERENCES

ACKNOWLEDGEMENT