Optimization Design of Distribution Line Lightning Protection Based on BP Neural Network Algorithm

At present, the lightning protection measures for 10 kV distribution lines mainly include installing lightning arresters, protecting gaps, strengthening insulation, erecting lightning wires, etc. To overcome the shortcomings of existing research, this paper proposes a new method of lightning protection optimization design of distribution lines based on CGA_BPNN (Cloud Genetic Algorithm_BP Neural Network) algorithm. By applying CGA to BPNN, the original BP algorithm is improved to try to achieve the optimization goal, so as to achieve the goal of improving the fault location performance of the distribution network. The research results show that it has a high recognition rate of 91.35% for lightning arrester action caused by lightning strike, 88.36% for sample judgment of lightning arrester inaction after lightning strike, and 87.01% for lightning arrester action caused by lightning strike. From the whole test sample, the overall recognition rate of CGA_BPNN model for whether the lightning arrester acts after lightning strike is 84.38%, and its overall accuracy rate is good.