Fault detection for multi-rate sampling systems based on dynamic principal component analysis

Abstract

For the multi-rate sampling systems with time series correlation data, a multi-rate fault detection algorithm based dynamic principal component analysis is proposed. The same sampling rate can be achieved in the algorithm by interpolation-filter-decimation, and then dynamic principal component analysis is implemented. The proposed method not only makes full use of the samples in a large number of incomplete data but also reduces the multi-rate sampling bias which caused by the data, the offline modeling and online monitoring strategies are also proposed. Finally, the Tennessee Eastman process is used to test the effectiveness of the proposed algorithm, and the simulation results show that the proposed method has a better performance in multi-rate sampling systems with serial correlation data than other fault detection methods based on principal component analysis.