Fault Diagnosis of Gas Blower Based on Genetic Fuzzy Neural Network

In order to make full use of the capability of GA´s global searching and BP network´s local searching, a genetic fuzzy neural network model is proposed. And the way of fault characteristic parameters´ fuzzy processing and optimizing the weights and thresholds of ANN by GA are studied. As a result, the convergence speed and convergence precision are greatly increased. Application to the fault diagnosis of a gas blower system shows that the new model overcomes the low learning rate and local minimum of BP algorithm and the fault diagnosis precision is effectively improved.