Research on Self-Organizing Mapping Neural Network Algorithm for Edge Security Access of Electric UAV

In this paper, an optimal network structure based on self-organizing mapping neural network algorithm is proposed to solve the problem of secure access and resource allocation of UAV power inspection system in edge computing environment. Firstly, this study designed an improved self-organizing mapping neural network algorithm to enhance the network's adaptability and learning ability by optimizing the network structure, to cope with the dynamic and uncertain environment in the process of power inspection. The core of the algorithm is to dynamically adjust the weight and node distribution of the network, to map the association between UAVs and edge computing nodes more accurately, and ensure the security and efficiency of data transmission. In order to verify the effectiveness of the algorithm, a system simulation model is constructed to simulate the actual scene of the UAV in the inspection process of power facilities. Simulation results show that the optimized self-organizing mapping neural network algorithm reduces the data transmission delay between UAVs and edge computing nodes by 23%, reduces the packet loss rate by 15%, and increases the overall throughput by 18%. These data prove that the algorithm has a significant effect in improving the performance of power inspection system and enhancing the safe access ability of UAV.