Extracting fault subspaces for fault identification of a polyester film process

Although fault detection has been well studied in process monitoring under stationary conditions, the identification of faults in chemical processes is a difficult task. We present a new approach in the use of a fault identification index to identify faults based on fault directions extracted from abnormal data using singular value decomposition. The proposed method is demonstrated on an industrial polyester film process which is characterized by frequent set-point changes and multiple grade changes. Further, a comparison between the fault identification index and the contribution plot method is given.