| In the producing industry, in order to make good use of the equipment, people always hope to predict the equipment fault, to nip in the bud, ensure that the equipment runs under the condition of safety, stabilization, long-period and full-load, and to void "over-maintain" or "ill-maintain" . So the discipline of equipment fault diagnosis comes into being and attracts large numbers of scientific researchers to research on it, so as to keep it on being developed and perfected.Along with equipment's development towards high-speed, high-power, high-stabilization and large-scale, it's difficult for the traditional method to diagnose the complex equipment fault. In view of this, Local Wave method and Artificial Intelligence are introduced into the field of fault diagnosis -to solve the problem of complex equipment fault diagnosis.Firstly, local-wave (LW) method is imported especially for solving questions about nonlinear and non-stationary of complex equipment surface vibration signals. The main innovations embodied in this method are the introduction of the intrinsic mode functions based on local properties of signals, which make the instantaneous frequency meaningful. Local-wave method overcomes validly disadvantages caused by using traditional methods, which eliminate the need for spurious harmonics to represent nonlinear and non-stationary signals. In addition, complicated tested data are decomposed into several intrinsic mode functions by this way, which low analyzing error, and predigest processing.Then, the common process of human being's intelligent action from the angel of information theory is analyzed, and the features of complex equipment fault diagnosis are summarized, and the generalized frameworks for intelligent fault diagnosis system is constructed. The following work is studied, put forward and realized on the basis of local-wave method.1. A approach to solve the end effect of local-wave decomposition based on Chebyshev numerical approximation is proposed. The method possesses the features of model simplicity, small observed sample size needed, easy real-time calculation and high predicting accuracy, and helps to make quick and accurate local-wave decomposition.2. Feature extraction is the bottle-neck of fault diagnosis now, which plays an important role in the accurate fault diagnosis and credible prediction. The concept of local energy in time-frequency plane, based on local wave theory, is defined.Then a new feature extraction method based on the local energy in time-frequency plane is given. The definition of local energy that expresses the time-varying of signal's frequency is an effective mean to analyze and solve the non-stationary signal accurately. Moreover, a new feature extraction method based on wavelet packet decomposition is presented on the basis of local energy theory.3. A new fault diagnosis system model based on rough sets theory-local wave-ANN is proposed. The diagnosis model firstly makes feature extraction with local wave method, and then dose attribute reduction and value reduction using rough sets theory. The method can solve ANN architecture, sample size, and sample quality, decrease the computation time, and increase the diagnosis correctness.4. A new method of multi-NN based on the evidence theory of D-S is proposed to solve the problem of fault diagnosis after analyzing the advantages and disadvantages of the evidence theory of D-S and.NN. The method not only solve the problem that the NN structure becomes bigger and bigger with fault parameters increasing, but also gives the basic probability assignment of the theory of D-S. Taking the gearbox fault diagnosis for example, the results indicate that the method is highly accurate and credible.At last, taking G79004 gearbox made in China 1st mobile Gearbox factory as background, the author develops 'Automobile Gearbox Performance Examining System' which is a integrated examining system including examining, on-line monitoring, signal analyzing and diagnosis. In the syst...
|
PDF Full Text Request