ABSTRACT
ABSTRACT: In the process of monitoring the running state of shortwave transmitting system, it is necessary to realize the intelligent discrimination and analysis of the monitoring data and enable users to obtain the real-time health status of the system quickly and accurately. Considering the large amount of information in monitoring data and the highly difficult analysis, the initial structure of BP neural network is optimized with genetic algorithm based on the calculation evaluation with traditional BP neural network algorithm, and then the optimal initial weight value is obtained by the optimizations such as improvement of chromosome coding, selection of fitness function and design of related operators. Based on the simulation training of a large number of monitoring data, the improvement effect of the algorithm is experimentally verified and analyzed. After this improvement, the algorithm optimization evaluation in the monitoring and diagnosis of shortwave transmitting system gets strong convergence ability and high diagnostic accuracy, which greatly increases the use efficiency of the monitoring system and lays a good practical foundation.
