Transient identification for nuclear power plants: Effect of the length of measured data to identification result

The measured data are the basis for all data-driven processing methods. Determining the length of measured data used to prepare a data set for subsequent processing is very important. The length of measured data is critical, especially for the transient identification for nuclear power plants, because it potentially determines both the identification delay and the identification rate. The transients must be identified in time (i.e., to make the identification delay as short as possible) and correctly (i.e., to make the identification rate as high as possible) during the nuclear power plant operation to enhance safety and achieve more economic benefits. Thus, this study investigates the effect of the length of measured data to the identification result and proposes an algorithm to obtain the optimal data length. The proposed algorithm was verified to be effective in determining the optimal data length, which resulted in the shortest identification delay and the highest identification rate. The investigation results showed the trend/law of the effect of the signal-to-noise rate to the optimal data length. Verifications were performed using our previously proposed identification method based on linear representation and the data from the simulator of the Chinese High Temperature Gascooled Reactor Pebble-bed Module project.