Fault diagnosis for steam turbine based on flow graphs and naïve Bayesian classifier

Steam turbine is key equipment in the power generating plants. Because of the complexity of its running environment, the fault diagnosis of steam turbine is a difficult problem. In order to improve the intuition and correctness of fault diagnosis, a novel fault diagnosis method for the steam turbine based on flow graphs and Naïve Bayesian classifier is proposed in this paper. Firstly, the initial FG is constructed according to the typical fault data of steam turbine. Then, the algorithms of layer and node reduction are used to eliminate the redundant and irrelevant attribute layers and attribute nodes to obtain the minimum FG. Finally, the case data is inputted into naïve Bayesian classifier to obtain classification results. To verify the proposed method, an experiment is carried out to apply this method to a case. The results show that it can intuitively and correctly diagnose the case fault.