Mixed kernel canonical variate dissimilarity analysis for incipient fault monitoring in nonlinear dynamic processes

Abstract

Incipient fault monitoring is becoming very important in large industrial plants, as the early detection of incipient faults can help avoid major plant failures. Recently, Canonical Variate Dissimilarity Analysis (CVDA) has been shown to be an efficient technique for incipient fault detection, especially under dynamic process conditions. CVDA can be extended to nonlinear processes by introducing kernel-based learning. Incipient fault monitoring requires kernels with both good interpolation and extrapolation abilities. However, conventional single kernels only exhibit one ability or the other, but not both. To overcome this drawback, this study presents a Mixed Kernel CVDA method for incipient fault monitoring in nonlinear dynamic processes. Due to the use of mixed kernels, both enhanced detection sensitivity and a better depiction of the growing fault severity in the monitoring charts are achieved. Looking ahead, this work takes a step towards understanding the impact of kernel behavior in process monitoring performance.

A Review of Kernel Methods for Feature Extraction in Nonlinear Process Monitoring

“…For instance, kernel ICA is equivalent to kernel PCA + ICA [66]. Likewise, kernel CVA can be performed as kernel PCA + CVA [67]. Hence, kernel PCA was cited more frequently than other techniques.…”

“…In addition, some MSPM tools are inherently capable of extracting dynamic features effectively, such as canonical variate analysis (CVA) [299], slow feature analysis (SFA) [300], and dynamic latent variable models (DLV). Kernel CVA is the kernelized version of CVA and is used in many works [67,166,172,177,178,223,224,281,290,291]. Meanwhile, kernel slow feature analysis has appeared in [174,215,216,259], and more recently, the kernel dynamic latent variable model was proposed in [225].…”

“…A variant of kernel ICA that avoids the usual KPCA-ICA combination is also proposed by Feng et al [262]. Aside from kernel ICA, the non-Gaussianity issue can also be handled using a kernel Gaussian mixture model [136], the use of statistical local approach for building the statistical index [112], and kernel density estimation (KDE) for threshold setting [67,194,251].…”

“…This is beneficial in particular for detecting incipient faults, which are small-magnitude faults with a drifting behavior. These faults are difficult to detect at the initial stage because they are masked by noise and process control [67]. Yet because they are drifting, they can seriously escalate if no action takes place.…”

Application of fault monitoring and diagnosis in process industry based on fourth order moment and singular value decomposition

“…ICA‐based methods were shown to be superior to conventional PCA‐based methods . As for various application scenarios, ICA has been extended to dynamic ICA (DICA), in order to monitor processes with strong dynamic features, and kernel ICA (KICA), for non‐linear process. Zhang and Zhao introduced both high‐order and second‐order statistics to enhance process monitoring performance.…”