Event-triggered-based subspace identification fault detection with an optimized moving window

Abstract

In this paper, a fault detection (FD) method with moving window subspace identification method (MW-SIM) is proposed for the problem of difficult detection of incipient faults. Since the size of the window length has a direct relationship with the fault detection rate, the optimal window length is found by the sparrow search algorithm (SSA) to obtain the maximum fault detection rate. Furthermore, applying event-triggered strategy to subspace identification algorithms can effectively reduce data transmission. Finally, the effectiveness of the designed strategy is verified by the Tennessee Eastman (TE) process simulation.