An Analysis System of Fault Diagnosis and Classification in Electrical Energy System Distribution Networks Based on Convolutional Neural Network

One of the primary causes contributing to the disruption of dependability and the termination of energy supply is the frequency of faults in distribution networks, making solar energy one of the most dependable renewable energy sources. Consequently, effective and rapid problem detection and prediction in distribution networks are critical for enhancing overall dependability, boosting customer happiness, and optimizing electrical energy efficiency. Rapid improvements in communication and automation technologies for distribution power networks allow a distributed generation (D.G.) protection coordination and reclosing strategy based on information exchange. Several artificial intelligence (A.I.) methodologies are used in modern classification, defect detection, and optimization for solar photovoltaic (P. V.) panels. Many algorithms are used to convert solar energy into electricity. During our investigation, we began searching for possible faults in the power system. The grid is a hybrid system that uses fossil fuels and solar energy to create electricity. The CNN approach was utilized and trained using a dataset comprising the highest and lowest voltage values that could be detected in a particular electrical network.