Feature Extraction Method in Fault Diagnosis Based on Wavelet Fuzzy Network for Power System Rotating Machinery

A new combined fault diagnosis approach for turbo-generator set based on wavelet fuzzy network is proposed. The wavelet transform is used to extract fault characteristics and neural network is used to diagnose the faults. To improve the performance of applying traditional fault diagnosis method to the vibrant faults, a novel method based on the statistic rule is brought forward to determine the threshold of each order of wavelet space and the decomposition level adaptively, increasing the signal-noise-ratio (SNR). The fault modes are classified by fuzzy diagnosis equation based on correlation matrix which shows good ability of self-adaption and self-learning. The improved least squares algorithm (LSA) is used to fulfill the network structure and the robustness of fault diagnosis equation is discussed. By means of choosing enough samples to train the fault diagnosis equation and the information representing the faults is input into the trained diagnosis equation,and according to the output result the type of fault can be determined. Actual applications show that the proposed method can effectively diagnose multi-concurrent fault for stator temperature fluctuation and rotor vibration and the diagnosis result is correct,increasing the accuracy of the fault diagnosis for rotating machinery.