基于遗传神经网络算法后勤指挥决策能力预测

摘要

分析了后勤指挥的主要影响因素，运用信息论中的信息压缩原理，建立和优化了后勤机关指挥决策能力预测指标体系。建立了遗传算法优化的BP神经网络后勤机关指挥能力预测模型，消除了BP神经网络存在的初始权值和阈值难以确定的局限性。给出了预测实例，证明了该算法的合理性和有效性。%Key link and main influencing factors of the command capability of logistics command organs are analyzed. Using the theory of information compress,the evaluating index system of logistics command are established and optimized. The neural network is optimized by genetic algorithm,and it is used to predict the command capability of logistics command organs. A prediction example is presented. The characteristics of generalization ,practicability and validity of the prediction model is verified,and it strongly support command,decision making and system optimum design of logistics command.