Modified t-Distribution Stochastic Neighbor Embedding Using Augmented Kernel Mahalanobis-Distance for Dynamic Multimode Chemical Process Monitoring

The traditional data-driven process monitoring methods may not be applicable for the system which has dynamic and multimode characteristics. In this paper, a novel scheme named modified t-distribution stochastic neighbor embedding using augmented Mahalanobis-distance chemical (AKMD-t-SNE) is proposed to realize multimodal monitoring. First, matrix strategy utilized ensure sample contains autocorrelation of process. Moreover, AKMD-t-SNE method eliminates scale spatial distribution differences among multiple modes by calculating kernel Mahalanobis distance between samples establish global model. features extracted via are obviously non-Gaussian, there will deviation in construction statistics. Then, support vector data description (SVDD) used construct statistics deal with problem. addition, hybrid correlation coefficient (HCC) achieve fault isolation improve accuracy results. advantages illustrated numerical case Multimode Tennessee Eastman Process (MTEP) benchmark.