BATCH/SEMI-BATCH PROCESS FAULT DETECTION AND DIAGNOSIS USING ORTHOGONAL NONLINEAR MULTI-WAY PCA: Application to an emulsion co-polymerization process

In this paper, a fault detection and diagnosis for batch/semi-batch processes by utilizing the PCA scores subspace is proposed. To develop the diagnosis model, first the multi-way unfolding is utilised to transform 3-dimensional batches data onto 2-dimensional data. The process of extracting linear and nonlinear correlations from process data is performed by sequentially applying a linear PCA and an orthogonal nonlinear PCA. As a result the nonlinear structures become more apparent. In addition, the sequential approach reduces the complexity of nonlinear PCA development and compact the information to a very low dimension. The trajectory-boundary-limit crossing point discriminant analysis is proposed to diagnose the fault at the instance of being detected and to improve the diagnostic performance. The validity of the proposed strategy is demonstrated by application to the emulsion copolymerization of styrene/MMA semi-batch process.