摘要
为准确地评定火炮剩余寿命,通过研究火炮内膛径向磨损量和初速下降量的相关关系,提出了基于最小二乘支持向量机的火炮特性模型,引入量子粒子群算法进行模型反演分析,确定最优参数,建立了火炮剩余寿命评定模型.对实弹射击测得的25组试样的实例应用分析表明,预测模型相对误差在±5%以下,显示了最小二乘支持向量机是一种较为有效的非线性建模方法,量子粒子群算法进行模型参数优化能够保证全局最优.验证结果表明,模型的精度较高,有一定的实用价值.
To evaluate the remaining life of a gun accurately, the relation between gun bore's radial wear and the decreasing quantity of muzzle velocity were researched, then a characteristic model was suggested using the least square support vector machine method. Furthermore, the quantum-behaved particle swarm optimization algorithm was introduced to carry out the model inverse analysis so as to determine the optimal parameters. A real sample application analysis of 25 groups from service practice indicates that the least square support vector machine is a kind of effective non-linear model method, and that the quantum-behaved particle swarm algorithm to optimize model parameters is able to guarantee the whole optimization, the relative error of prediction model is within ±5%. The testing results show that the model is of high accuracy and practical in use.
出处
《弹道学报》
EI
CSCD
北大核心
2009年第3期65-68,共4页
Journal of Ballistics
关键词
炮膛
径向磨损
初速
最小二乘支持向量机
量子粒子群算法
gun bore
radial wear
muzzle velocity
least square support vector machine
quan turn-behaved particle swarm algorithm