Fault causes and its detection in standalone PV system using ANN and GEO technique.

Area of Science:

• Renewable Energy Systems
• Artificial Intelligence in Engineering
• Fault Detection and Diagnosis

Background:

• Photovoltaic (PV) systems are widely adopted for efficient power generation.
• Partial shading is a common defect that significantly reduces PV system performance and energy quality.
• Effective fault detection and diagnosis are crucial for maintaining optimal PV system operation.

Purpose of the Study:

• To propose a novel method for predicting, detecting, and diagnosing faults in standalone PV systems.
• To enhance the performance and reliability of PV power generation.
• To address the common partial shading defect in PV systems.

Main Methods:

• Integration of Artificial Neural Network (ANN) for fault classification and identification.
• Application of Golden Eagle Optimization (GEO) for efficient ANN training, reducing time and improving accuracy.
• Implementation and testing of the proposed technique using MATLAB simulations.

Main Results:

• The proposed ANN and GEO technique effectively classifies and identifies PV system defects.
• GEO optimization significantly reduces ANN training time and enhances model accuracy.
• Comparative analysis shows superior performance over existing methods like Genetic Algorithm (GA), Elephant Herding Optimization (EHO), and Particle Swarm Optimization (PSO).

Conclusions:

• The combined ANN and GEO approach offers a highly accurate and effective solution for PV system fault detection and diagnosis.
• This method successfully recovers optimal performance in PV systems affected by defects.
• The study demonstrates a significant advancement in maintaining the efficiency and reliability of photovoltaic power generation.