Fault detection using CUSUM based techniques with application to the Tennessee Eastman Process

In this paper, a cumulative sum based statistical method is used to detect faults in the Tennessee Eastman Process (TEP). The methodology is focused on three particular faults that could not be observed with other fault detection methodologies previously reported. Hotelling's-T2 charting based on the cumulative sums of the faults' relevant variables was successful in detecting these faults, however, with significant delay. The speed of detection is further enhanced by retuning the fault's relevant controller at the expense of closed loop performance.