The technology of real-time fault diagnosis for nuclear power plants (NPP) has great significance to improve the safety and economy of reactor. Nuclear power plants are complex system, which collect and monitor the vast parameters. A parameter reduction method based on fuzzy rough sets was proposed. According to the characteristics the parameters were fuzzed, and they were reducted using the algorithm of forward greedy search. The decision tree was applied to learn from training samples which were the typical faults of nuclear power plant, i.e., loss of coolant accident (LOCA), feed water pipe rupture, steam generator tube rupture (SGTR), main steam pipe rupture, and diagnose by using the acquired knowledge. The result shows that this method can diagnose the faults of the NPP rapidly and accurately.

The Stability of SG water level plays an important role in the safety of nuclear power plants, but tuned the parameter of water level PID controller is hard. Proposed a novel algorithm, KIPSO, which tuning PID controller parameters. Determine the cluster centre through K-means value cluster algorithm, and take the cluster territory as the characteristic value of vaccine set, enhance the vaccine multiplicity. Updated vaccine extraction by self-adaptive method, improved the convergence and adaptability. Analyzed the algorithm robustness in detail, and gave the rule which the immunity selection parameter. The simulation results shows: compares with the PID controller whose parameters are tuned by ZN method, KIPSO have a smaller overshoot, a better stability, and a shorter adjustment time. The simulation results show that the proposed method is effective for tuning PID parameters.

No abstract