Research on Sensor Fault Diagnosis Method of a Small Natural Circulation Lead-cooled Fast Reactor Based on PCA-SVM
The small natural circulation lead-cooled fast reactor operates in harsh environments and is typically left unattended.In such circumstances,sensor failures can have a significant impact on the system safety.To promptly detect and diagnose sensor faults,a method combining the Principal Component Analysis(PCA)and the Support Vector Machine(SVM)is proposed.This data-driven approach eliminates the need for detailed mathematical models of the system or prior knowledge of specific states.By integrating the PCA,it efficiently reduces data dimensions.Using the dynamic model of the small lead-cooled fast reactor built in the MATLAB/Simulink,the sensor data is generated to train and establish the fault diagnosis model,and the test is carried out to verify the accuracy and effectiveness of the proposed fault diagnosis method.
Lead-cooled fast reactorSensor failureFault diagnosisPrincipal components analysisSupport vector machine
NETL
NSTL
万方数据
