Research on the Construction of Diagnosis Knowledge Base of Multi-Base RPN Intermediate Range ACCG4 Board Based on Expert System

The nuclear power plant's external nuclear measurement system (RPN) plays a vital role in reactor power monitoring. The ACCG4 board, as the core component of the system, is responsible for converting the current signal output by the intermediate range detector into a voltage signal and a switch signal. The frequent failure of the ACCG4 board has become a common problem faced by nuclear power plants, affecting the reliability of the system and the safe operation of nuclear power plants. Traditional fault diagnosis methods are mostly based on threshold settings, but when faced with complex fault modes, the diagnostic ability and accuracy of these methods are obviously insufficient. Therefore, this paper presents a method of building a KDB based on an expert system, which can improve the precision and efficiency of the ACCG4 board. To validate the validity of the proposed algorithm, a simulation environment was built, and its advantages were proved by the test of each typical failure situation. Compared with the traditional method, the new method can increase the precision of fault recognition by 30% and reduce the diagnostic time by 40%. The research provides a powerful technical support for the safe running of RPN system in nuclear power plant, and provides a new method to diagnose faults in relevant areas.