基于小波去噪与KPCA的TE过程故障检测研究

Fault Detection of Tennessee Eastman Process Based on Wavelet Denoising and KPCA

针对化工过程复杂非线性,并且含有噪声和随机干扰的特点,提出利用小波去噪与核主元分析(KPCA)相结合的方法来进行故障检测,既可以达到去噪、抗干扰的目的,又可以将输入空间中复杂的非线性问题转化为特征空间中的线性问题,从而解决了主元分析(PCA)方法在非线性过程中性能差的问题。并将该方法应用于Tennessee Eastman(TE)化工过程模型,仿真结果表明其在故障检测方面明显优越于普通的PCA方法。

Aiming at the complex and nonlinear chemical process,an fault detection method which combing wavelet denoising with kernel principal component analysis（KPCA） was presented to solve the noise and disturbance and to compute principal components in higher dimensional feature space by means of nonlinear kernel functions.The simulations in Tennessee Eastman（TE） chemical process demonstrate that this method outperforms the principal component analysis（PCA） in fault detection.