As a main participator in the design for and research on engineering the control system, in the project of eliminating ash with water in Jianbi Power Plant, the author emphasized on the intelligent fault diagnosis for the complex hydraulic system by the background of that project in this paper. First, several methods of processing traditional digital signal applied in the machine fault diagnosis are discussed in the paper. Then it turns to the analysis on time and frequency, to the different sound signals of a rolling piston radial pump, under both in normal and abnormal conditions. By comparing the apparent characteristics indicating the leakage fault which are difficult to be found out, several reasons result in the analysis are stated. For the nonstationarity of the signal in field, eigenvector of fault is computed with newly advanced wavelet analysis, as the preprocessing in diagnosis procedure.The predominance of signal processing with artificial neural network (ANN) results very well in solving fault diagnosis problems of the complex mechanical equipment. For example, the simulation is given to prove ANN's noiseproof feature in this paper. The intelligent fault diagnosis based on BP ANN is designed for recognition of leakage fault of a rolling piston radial pump. It combines with the predominance of diagnosis methods based on mathematic model and artificial intelligence, and it is achieved through software on personal computer. The usability of system is proved by the sample data. In the end of the paper, things waited on improvement are pointed out as well as the development tendency of the intelligent fault diagnosis. |