基于量子遗传的机械故障非线性盲源分离方法研究  

NONLINEAR BLIND SOURCE SEPARATION OF MECHANICAL FAULT BASED ON QUANTUM GENETIC ALGORITHM

作　　者：皮海玉 李志农[1] 肖尧先[1] 

机构地区：[1]南昌航空大学无损检测技术教育部重点实验室,南昌330063

出　　处：《机械强度》2015年第3期393-397,共5页Journal of Mechanical Strength

基　　金：国家自然科学基金(51075372;51265039;50775208);江西省教育厅科技计划项目(GJJ12405);湖南科技大学机械设备健康维护湖南省重点实验室开放基金(201204)资助~~

摘　　要：针对传统的机械故障非线性盲分离方法的不足,即将非线性盲源分离中分离矩阵和非线性去混合函数的参数分开来优化,这样容易顾此失彼,学习效率低。将量子遗传引入到机械故障非线性盲分离中,提出一种基于量子遗传的机械故障非线性盲源分离方法(简称QGA-NBSS方法),该方法能同时对分离矩阵和非线性去混合函数的参数进行优化,获得全局最优解并加快了算法的全局收敛性,克服了传统的机械故障源的非线性盲源分离方法的不足。仿真和实验结果验证了提出的方法的有效性。Based on the deficiency in the traditional nonlinear blind separation method of mechanical fault sources, i.e. the separation matrix parameter and nonlinear mixing parameter in the nonlinear blind source separation are usually optimized separately, which easily lead to have one without another and low learning efficiency. Quantum genetic algorithm is introduced into the nonlinear blind source separation of mechanical fault, a nonlinear blind separation method of mechanical fault sources based on the quantum genetic algorithm, which is named as QGA-NBSS method, is proposed. The proposed method can simultaneously optimize all parameters in the nonlinear blind separation, i.e. the separation matrix and nonlinear mixing function, obtain global optimal solution, and greatly improves the global convergence of the algorithm. The simulation and experimental results show that the proposed algorithm is effective.

关 键 词：量子遗传 非线性盲源分离 故障诊断 

分 类 号：TH17[机械工程—机械制造及自动化] H165.3[语言文字—汉语]