基于改进遗传算法的装备保障效能评估指标约简  

Reduction of Equipment Support Effectiveness Evaluation Index Based on Improved Genetic Algorithm

作　　者：潘成胜[1] 周敏[1] 杜秀丽[1] 吕亚娜 PAN Cheng-sheng;ZHOU Min;DU Xiu-li;LV Ya-na(Communication and Network Key Laboratory,Dalian University,Dalian Liaoning 116622,China)

机构地区：[1]大连大学通信与网络重点实验室,辽宁大连116622

出　　处：《计算机仿真》2021年第7期223-229,共7页Computer Simulation

基　　金：中央军委装备发展部领域基金项目(61400010301)。

摘　　要：针对现有装备保障系统效能评估指标体系存在指标冗余而导致效能评估复杂度高的问题,提出了基于改进遗传算法的装备保障系统效能评估指标约简方法。首先,方法通过强精英参与、完备交叉与预变异策略对遗传算法的选择、交叉与变异步骤进行改进;然后,将改进后的遗传算法用于装备保障系统效能评估指标约简,以指标间综合互信息与选择比例构建遗传算法适应度函数。仿真结果表明,改进后的遗传算法具有较高的收敛性与全局寻优能力,与粗糙集属性约简算法和灰色关联分析算法对比表明,本文方法能够在选择更少评估指标的同时,得到更高的评估精度。Aiming at the problem of high complexity of effectiveness evaluation caused by index redundancy in the existing effectiveness evaluation index system of equipment support system, a reduction method of equipment support system effectiveness evaluation index based on improved genetic algorithm is proposed. Firstly, the selection, crossover and mutation steps of genetic algorithms were improved through strong elite participation, complete crossover and pre-mutation strategies. Then, the improved genetic algorithm was used to reduce the efficiency evaluation indexes of the equipment support system, and the fitness function of genetic algorithm was constructed by synthesizing mutual information and selection ratio among the indexes. The simulation results show that the improved genetic algorithm has higher convergence and global optimization ability;Compared with rough set attribute reduction algorithm and gray correlation analysis algorithm, the proposed method can obtain higher evaluation accuracy while selecting fewer evaluation indexes.

关 键 词：装备保障系统 效能评估 遗传算法 指标约简 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]