面向装备作战的弹药消耗智能预测方法  

Method of Ammunition Consumption Intelligent Prediction Oriented on Equipment Combat

作　　者：樊胜利[1] 柏彦奇[2] 张耀坤[1] 姚涛[1] 

机构地区：[1]军械工程学院基础部,河北石家庄050003 [2]军械工程学院装备指挥与管理系,河北石家庄050003

出　　处：《装甲兵工程学院学报》2011年第1期22-27,共6页Journal of Academy of Armored Force Engineering

基　　金：军队科研计划项目

摘　　要：针对装备作战仿真环境下,弹药消耗预测计算时间长、实时仿真速度慢的问题,以战役战术火力毁伤弹药需求预测计算方法为主线,以人工神经网络为技术支持,提出了智能化的弹药消耗预测方法。通过遗传算法对网络连接权值的训练及优化,以及误差目标函数的改进与隐含层神经元数目的智能选取,提高弹药消耗预测的实时仿真速度,为下一步的弹药消耗预测研究提供了理论参考与方法支持。Aiming at the problem of longer time computing on ammunition consumption prediction and lower speed on real-time simulation in the circumstance of equipment combat simulation,the intelligent prediction method of ammunition consumption is brought forward with the fire destroy ammunition demand prediction computing method as masterstroke and artificial neutral network as technical support.And it improves the real time simulation speed of ammunition consumption prediction by means of genetic algorithm training and optimizing on network linking weight,improving error target function and intelligent selection of nerve cell quantity so that it provides theoretical reference and method support for future research on ammunition consumption prediction.

关 键 词：装备作战仿真 弹药消耗预测 遗传算法 智能 

分 类 号：E932[兵器科学与技术—火炮、自动武器与弹药工程] TP183[军事—军事装备学]