Classification of Loss of Coolant Accident for Advanced Boiling Water Reactor

摘要

In this study, an artificial intelligent (AI) approach is proposed to replace the model-based approach for accident diagnosis and classification to assist operators' decision making. A neutral network is constructed using the support vector classification (SVC) technique to classify LOCA events for advanced boiling water reactor (ABWR). The simulator based on the model of Taipower Lungmen nuclear power station (LNPS) is used to provide the required training and test data sets. Demonstration has shown that the present technique is a feasible approach for events classification.