Transient identification for nuclear power plants based on the similarity of matrices

A transient is defined as a process where a system transforms into an abnormal state from a normal state. In order to enhance safety and achieve greater economic benefits, it is very important to detect and identify transients in a timely manner during the operation of nuclear power plants. Thus, according to the ideas employed in image processing and photo retouching, we propose a novel method for transient identification based on the similarity of matrices. We theoretically formalize how to identify transients based on data measurements. Next, to improve the feasibility and reduce the identification delay, the method is extended using the “moving window” approach. The proposed method was verified based on sampling data obtained from the simulator of the high temperature gas-cooled reactor pebble-bed module, which is a new demonstration nuclear power plant with gas-cooled reactors that is under construction in China. The results demonstrated that the proposed method could detect the transients correctly with a short identification delay. We also showed that the computation load of the proposed method is low. Thus, this method is suitable for real-time applications.