A sparse dissimilarity analysis algorithm for incipient fault isolation with no priori fault information

The conventional multivariate statistical process control (MSPC) methods in general quantify the distance between the new sample and the modelling samples for fault detection and diagnosis, which, however, do not check the changes of data distribution as long as monitoring statistics stay inside normal region enclosed by control limit and thus are not sensitive to incipient changes. In the present work, a sparse dissimilarity (SDISSIM) algorithm is developed which can isolate the incipient abnormal variables that change the data distribution structure and does not need any priori fault knowledge. First, the distribution dissimilarity is decomposed deeply and significant dissimilarity is extracted to integrate the critical difference of variable covariance structure between the reference normal operation distribution and the actual distribution. Second, a sparse regression-based optimization problem is formulated to isolate abnormal variables associated with changes of distribution structure. Sparse coefficients are obtained with only a small fraction of variables' coefficients nonzeros, pointing to abnormal variables. As illustrations, SDISSIM is applied to both simulated and real industrial process data with encouraging results to figure out the slight distortions.