Research on the Sample Training of BP Neural Network in Effectiveness Evaluation

Abstract

According to the WSEIAC (Weapon System Effectiveness Industry Advisory Committee) model, an index hierarchy of ground antiaircraft missile weapon system's effectiveness has been developed, and corresponding three hierarchy BP neural network was established. It is briefly concerned with the analysis of the BP algorithm, then through Delphi technique and the FAHP (fuzzy analytical hierarchy process), several groups of training samples are chosen to train the BP neural networks until the precision meet requirements. It is shown that this BP neural network limits the artificial factors when it is used to evaluate the ground antiaircraft missile weapon system's effectiveness. It was concluded that this method is scientific and creditable.