基于核递推最小二乘的自适应均衡的弱故障提取方法研究

Adaptive equalization method based on kernel recursive least square and its application in weak fault extraction

针对传统的自适应均衡方法存在的不足,提出了一种基于核递推最小二乘(KRLS)的非线性系统自适应均衡方法。该方法通过引入核函数,将原始的非线性数据映射到高维特征空间,然后在高维特征空间中实施标准最小二乘算法。提出的方法并与传统的非线性系统均衡方法进行了对比分析,仿真研究表明,提出的方法优于传统的均衡方法,能很好的消除传递通道的影响,有效地提取出弱冲击性成分。最后,将提出的方法应用到转子系统的弱冲击性故障提取中,实验结果进一步验证了提出的方法的有效性。

Aiming at defects of the traditional self-adaptive equalization,a new self-adaptive equalization method based on kernel recursive least-square（KRLS）was proposed.With the proposed method,the original nonlinear data were mapped to a high-dimensional feature space by introducing a kernel function.Then,the standard least-square algorithm was implemented in the high-dimensional feature space.the proposed method was compared with the traditional self-adaptive equalization one.The simulation results showed that the proposed method is supiror to the raditional self-adaptive equalization one;the new method can effectively eliminate the influence of transfer channels.Finally,the proposed method was applied to extraction of weak impact faults in a rotor system,the test results verified the validity of the proposed method.