An improved mixture of probabilistic PCA for nonlinear data-driven process monitoring

TL;DR

This paper introduces an improved mixture of probabilistic PCA models for nonlinear process monitoring, combining local PPCA models with a novel composite statistic to enhance fault detection accuracy in complex systems.

Contribution

The paper presents a new composite monitoring statistic based on a mixture of local PPCA models, improving nonlinear process fault detection over existing unsupervised algorithms.

Findings

Effective fault detection demonstrated on Tennessee Eastman process

Enhanced monitoring accuracy compared to traditional methods

Validated on autosuspension model with promising results

Abstract

An improved mixture of probabilistic principal component analysis (PPCA) has been introduced for nonlinear data-driven process monitoring in this paper. To realize this purpose, the technique of a mixture of probabilistic principal component analysers is utilized to establish the model of the underlying nonlinear process with local PPCA models, where a novel composite monitoring statistic is proposed based on the integration of two monitoring statistics in modified PPCA-based fault detection approach. Besides, the weighted mean of the monitoring statistics aforementioned is utilised as a metrics to detect potential abnormalities. The virtues of the proposed algorithm have been discussed in comparison with several unsupervised algorithms. Finally, Tennessee Eastman process and an autosuspension model are employed to demonstrate the effectiveness of the proposed scheme further.