Fault Diagnosis for the Feedwater Heater System of a 300MW Coal-Fired Power Generating Unit Based on RBF Neural Network
In this paper, a new style radial basis function (RBF) neural network is used for fault diagnosis of the high-pressure feed-water heater system of a coal-fired power generating unit. The structure of the RBF network and its training algorithm are given. Another important factor to realize neural network based fault diagnosis, fault symptom fuzzy calculating methods for two different fault symptoms and their integrated calculation, are discussed in detail. The high-pressure feed-water heater system of a 300MW coal-fired power generating unit is taken as a fault diagnosis example. The fault knowledge library of the system is summarized. The fault diagnosis is further realized based on the above RBF neural network. It is shown that good diagnostic results can be acquired with RBF neural network method by using the fault fuzzy knowledge library of the high-pressure heater system.
KeywordsRadial Basis Function Fault Diagnosis Radial Basis Function Neural Network Radial Basis Function Network Orthogonal Little Square
Unable to display preview. Download preview PDF.
- 1.Leonard, J.A., Kramer, M.A.: Radial basis function networks for classifying process fault. IEEE control systems, 31–37 (April 1991)Google Scholar
- 4.Liangyu, M., Jianqiang, G., Yongguang, M.: Fault intelligent diagnosis for high-pressure feed-water heater system of a 300MW coal-fired power unit based on improved BP neural network. In: 2002 IEEE/PES international conference on power system technology proceedings, Kunming, vol. 3, pp. 1535–1539 (2002)Google Scholar
- 5.Liangyu, M., Bingshu, W., Jianqiang, G.: A new intelligent approach to diagnose the slight and incipient fault during the production process. Proceedings of the CSEE 22(6), 115–118 (2002)Google Scholar