Fault prediction of boilers with fuzzy mathematics and RBF neural network

How to predict potential faults of a boiler in an efficient and scientific way is very important. A lot of comprehensive research has been done, and promising results have been obtained, especially regarding the application of intelligent software. Still there are a lot of problems to be studied. It combines fuzzy mathematics with. RBF neural network in an intuition and natural way. Thus a new method is proposed for the prediction of the potential faults of a coal-fired boiler. The new method traces the development trend of related operation and state variables. The new method has been tested on a simulation machine. And its predicted results were compared with those of traditional statistical results. It is found that the new method has a good performance.