Application of CART-Based Variable Ranking for Faulty Variable Isolation in Tennessee Eastman Benchmark Process

By analyzing multivariate data collected from complex industrial processes via data-driven techniques, potential process faults can be accurately detected and isolated in a timely manner; this is essential for ensuring safety, availability, and reliability of them. In this paper, we apply the fault isolation (FI) method via classification and regression tree based variable ranking (proposed by Yu et al. [12]) to Tennessee Eastman (TE) benchmark process; TE process has been widely used in academic fields of fault detection and isolation. The purpose of this paper is to verify the performance of the FI method through TE benchmark process. As described in experimental results, the method can isolate faulty variables more clearly than comparison methods without fault smearing.