Application of data mining on fault detection and prediction in Boiler of power plant using artificial neural network

This paper tries to present a new applied method on detection and prediction of faults for the boiler´s burner system of power plant with using data mining and artificial neural network. Boiler/Steam turbine is important equipments in the industry, especially in the electric power industry. Because of the complexity of burner management systems and particularity of its running environment, the fault rate of boiler´s burner system is high. So the fault prediction is a difficult problem. The proposed approach includes data mining, data preprocessing i.e. data reduction, data clustering; learning and prediction by artificial neural networks. Boiler/turbine units constitute a critical component of a co-generation system. The operative parameters in boiler´s burner system are measured and are characterized to obtain a set of descriptors. These sets are analyzed by data mining approach. Next, these preprocessed data are used as input data of two neural networks which detect and predict the faults in a boiler of power plant. Multiplayer back propagation neural network with four hidden layers, as one of the steps in data mining process is studied. The knowledge extracted by this data mining algorithm is an important component of an intelligent alarm system. Furthermore, using this method is more valuable for the further study.