Study on Fault Diagnosis in Nuclear Power Plant Based on Rough Sets and Support Vector Machine

The faults of Nuclear Power Plant （NPP） are featured with complication and uncertainty. A NPP fault diagnosis method based on Rough Sets （RS） and Support Vector Machine （SVM） is proposed. Firstly, the uncertain data is reduced based on RS theory. According to the chosen reduction a SVM multi-classifier is designed for fault diagnosis. Finally this method is used to diagnose four typical failures, i.e., steam generator tube rupture accident, cold leg rupture accident, vapour phase rupture accident and loss of heat sink accident. The result shows that this method can diagnose the faults of the NPP rapidly and accurately.