研究目的
To propose a method for modeling, detection, and classification of faults in photovoltaic systems using Artificial Neural Networks (ANN) to ensure the system operates at expected performance and reliability levels.
研究成果
The proposed ANN-based method effectively detects and classifies faults in PV systems, such as short circuits and partial shading, with high accuracy. Future work aims to extend this research to grid-connected PV systems.
研究不足
The study focuses on specific types of faults (short circuit and partial shading) and may not cover all possible faults in PV systems. The ANN model requires a rich database for training, which may not always be available.
1：Experimental Design and Method Selection:
The study uses ANN for fault detection in PV systems, focusing on short circuit faults and partial shading with and without bypass diodes.
2：Sample Selection and Data Sources:
A photovoltaic module PV TDG Holding T 250 P606 is used for simulation.
3：List of Experimental Equipment and Materials:
Matlab / Simulink software for simulation.
4：Experimental Procedures and Operational Workflow:
Simulation of PV system under normal and fault conditions, followed by ANN training and testing.
5：Data Analysis Methods:
ANN performance is evaluated based on the accuracy of fault detection and classification.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
