Multimode Process Monitoring Based on Sparse Principal Component Selection and Bayesian Inference-Based Probability

Abstract EN

According to the demand for diversified products, modern industrialprocesses typically have multiple operating modes.

At the same time,variables within the same mode often follow a mixture of Gaussiandistributions.

In this paper, a novel algorithm based on sparseprincipal component selection (SPCS) and Bayesian inference-basedprobability (BIP) is proposed for multimode process monitoring.

SPCScan be formulated as a just-in-time regression between all PCs andeach sample.

SPCS selects PCs according to the nonzero regressioncoefficients which indicate the compact expression of the sample.

This expression is necessarily discriminative: amongst allsubset of PCs, SPCS selects the PCs which most compactly express thesample and rejects all other possible but less compact expressions.

BIP is utilized to compute the posterior probabilities of each monitoredsample belonging to the multiple components and derive an integratedglobal probabilistic index for fault detection of multimode processes.

Finally, to verify its superiority, the SPCS-BIP algorithm is appliedto the Tennessee Eastman (TE) benchmark process and a continuous stirred-tankreactor (CSTR) process.