基于量子遗传的机械故障盲源分离方法研究

Blind Source Separation of Mechanical Fault Based on Quantum Genetic Algorithm

针对基于遗传算法的机械故障源分离(GA-BSS)方法存在的不足和量子遗传的独特优势,提出了基于量子遗传的机械故障盲源分离(QGA-BSS)方法,并与传统的GA-BSS方法进行了比较。仿真结果表明,提出的方法优于GA-BSS方法,尤其是在快速收敛性方面,避免了GA-BSS方法早熟收敛,同时也大幅度地减少了计算量。将提出的方法应用到轴承故障分离中,能很好地提纯出轴承故障特征。实验结果证明,提出的QGA-BSS方法是有效的。

For the deficiency in the blind separation method of mechanical fault sources based on the genetic algorithm,which is named as GA-BSS method,and the unique advantages of quantum genetic algorithm,a blind separation method of mechanical fault sources based on the quantum genetic algorithm,which is named as QGA-BSS method,is proposed. The proposed method is compared with the traditional GA-BSS method. The simulation results show that the QGA-BSS method is superior to the traditional GABSS method,especially in the convergence speed. The proposed method avoids the premature convergence in the GA-BSS method and greatly reduces the amount of calculation. Finally,The proposed method is applied to the separation of bearing fault,and can extract the bearing fault features from the mixture signals successfully. The experimental results prove that the proposed QGA-BSS method is effective.