Intelligent collaborative control method for instrument and control system of nuclear power plant based on multi-source data fusion

Abstract

The application relates to the field of nuclear power plant instrument control and automation technology and discloses a nuclear power plant instrument control system intelligent collaborative control method based on multi-source data fusion, which comprises the following steps: collecting original state data and generating an original observation vector; performing outlier elimination and multi-source data fusion to obtain pretreated optimal data and a fusion confidence coefficient; extracting the pretreated optimal data, running a multivariable model predictive control algorithm and a fuzzy adaptive PID control algorithm in parallel, establishing a dynamic weight distribution model based on the fusion confidence coefficient, and generating a collaborative control instruction; synchronously extracting core safety parameters to generate a safety protection action instruction, and extracting features to generate an active pre-switching instruction through hyperplane division; and finally adjusting operation parameters according to the instruction, executing power-off protection or shifting a hardware transmission path. Through the application, end invalid overshoot caused by slight fluctuations of front-end data can be effectively inhibited, and the control precision and stability of the system under complex working conditions can be improved.