Research On Modeling And Fault Diagnosis Method Of Control Rod Drive Mechanism Power Supply System In Nuclear Power Plant

Nuclear power has the characteristics of clean,efficient,low carbon,economic and so on.Because it conforms to the development trend of energy,it is favored by all countries and has a very broad prospect.The power supply system of control rod drive mechanism is one of the important equipments of nuclear power reactor.Therefore,whether it can operate safely and stably will directly affect the safety of nuclear power plant units.The drawer skip trip events of the ac power supply system of the conventional island of nuclear power plant may lead to the loss of power of 380 V bus and transient state of the units.The thesis takes it as the background.Using MATLAB/ Simulink software for the motor power supplys,motors,generators,outlet circuit breakers and public bus,generator excitation systems and other components of the bar power supply unit part modeling,and simulation of the model,gets the normal operation of the bar power supply system electrical volume.According to the multiple faults of the control rod drive mechanism power supply system of nuclear power plant and the simulation model,the multiple faults are simulated,and the variation of the electrical quantity under various fault states is obtained.Next,the application of the mode maximum method based on wavelet transform and the phase difference method based on complex wavelet transform in fault detection is tested and analyzed,and applied to the fault detection of control rod drive mechanism power supply system.According to the defects of the test results.Based on these two methods,a new fault detection method is proposed to make it more suitable for control rod drive mechanism power system fault detections,and the errors of this method is tested,which can meet the requirements of engineering practice.According to the characteristics of Virtex series FPGA,the realization circuit of the improved algorithm is designed and simulated.The simulation result shows that the algorithm can achieve the expected function well.The error and its cause are analyzed according to the characteristics of the algorithm and circuit.Finally,according to the faults of the control rod drive mechanism power supply system,the traditional artificial immune system is used to diagnose the faults,and according to the results of fault diagnosis,the faults are classified into the first type,the second type and the third type.Applying the improved fault detection method,two steps of fault detection and fault extraction are added to the traditional artificial immune system.Fault detection can shorten the diagnosis time of the first type of faults and reduce the probability of fault diagnosis.Fault extraction can effectively reduce the sample size required for fault diagnosis.The addition of these two steps greatly improves the efficiency of artificial immune system in control rod drive mechanism power system fault diagnosis.Based on the idea of multi-source information fusion,the traditional artificial immune system is added with secondary immune system,and a new input of secondary immune system is proposed according to the characteristics of the electricity quantity obtained from the simulation of the control rod drive mechanism power system,which can greatly improve the accuracy of fault diagnosis by introducing secondary immune system for the first and second types of faults.Based on rough set theory,the matching classification algorithm of artificial immune system is improved,and the accuracy of fault diagnosis is improved again on the basis of secondary immunity.