Many industrial processes are operated in multiple modes due to different manufacturing strategies. Multimodality of process data is often accompanied with nonlinear and non-Gaussian characteristics, which makes data-driven monitoring more complicated. In this paper, statistics pattern analysis (SPA) is introduced to extract low- and high-order statistics from raw process data. Support vector data description (SVDD), which can deal with nonlinear and non-Gaussian problems, is applied to monitor multimode process in this paper. To improve detection performance of SVDD for training multimode data with outliers, modified local reachability density ratio (mLRDR) is proposed as a weight factor to be embedded in the weighted-SVDD (wSVDD) model, in which the local neighbors in terms of both space and time are considered. Finally, the effectiveness and superiority of our proposed method are demonstrated by the Tennessee-Eastman (TE) process and wastewater treatment process (WWTP).
Industrial processes with high dimensional data are generally operated with mixed normal/faulty states in different modes which are difficult to be automatically and accurately identified. In this paper, a state identification framework is proposed for multimode processes. First, a key variable selection approach is presented based on sparse representation to eliminate redundant variables. Then, the modified density peak clustering (MDPC) is proposed to identify different states, in which a distance measurement with a time factor is constructed to select out all the possible cluster centers; and the sum of squared error(SSE) based approach is developed to determine the optimal cluster centers automatically. Further, considering that the mode attributes may be mixed with the fault attributes, a two-step “coarse-to-fine identification” strategy is designed to precisely identify the mode and the faults in each mode. Finally, three cases including a numerical simulation, Tennessee Eastman (TE) benchmark process and an actual semiconductor manufacturing process are given to show the feasibility of the proposed method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.