基于小波去噪的FVS-KPCA故障检测方法

对于复杂非线性系统,实际得到的数据不可避免地带有噪声、随机干扰,而传统的核主元分析(KP-CA)方法应用于大样本集的故障检测,要计算核矩阵K很困难。为此,提出一种小波去噪与特征矢量选择-核主元分析(FVS-KPCA)相结合的故障检测方法,首先对数据进行小波去噪,再采用特征矢量选择(FVS)与KPCA结合的方法能有效降低故障检测计算的复杂性。把上述方法应用到Tennessee Eastman(TE)化工过程,仿真结果表明该方法能有效地提高故障检测的速度。

A FVS-KPCA Method of Fault Detection Based on Wavelet Denoising

For complicated nonlinear systems,the received data inevitably had noise,random disturbance,and the traditional kernel principal component analysis（KPCA） methods were very difficult to calculate the kernel matrix K for fault detection with large sample sets.An integrated fault detection method based on wavelet denoising and feature vector selection-KPCA（FVS-KPCA） was developed.First,wavelet denoising method was used for data processing,and then FVS-KPCA method could evidently reduce calculation complexity of fault detection.Finally,the proposed method was applied to the benchmark of Tennessee Eastman（TE） processes.The simulation results show that the proposed method can effectively improve the speed of fault detection.