基于PCA-SVM的小型自然循环铅冷快堆传感器故障诊断方法研究

Research on Sensor Fault Diagnosis Method of a Small Natural Circulation Lead-cooled Fast Reactor Based on PCA-SVM

离岸式小型自然循环铅冷快堆运行在恶劣环境下,且通常少人值守,在这种情况下,传感器故障可能对系统的安全性产生严重影响。为及时检测和诊断传感器故障,提出了一种结合主成分分析(PCA)和支持向量机(SVM)的方法。所提方法基于数据驱动,不需要系统的详细数学模型或特定状态的先验知识。采用PCA方法,能够有效降低数据维度。使用MATLAB/Simulink中构建的小型铅冷快堆动态模型,产生传感器数据,以训练和建立故障诊断模型,并进行了性能测试,验证了所提出的故障诊断方法的准确性和有效性。

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.