Research On Vibration Fault Diagnosis Of Steam Turbine Based On Time Domain Analysis And Wavelet Analysis

With the development of science and technology, steam turbine power will be gradually increased, steam turbine fault hazards greatly, and the failure rate is very high, if by some measures, timely and accurately diagnosis in turbine fault, will greatly reduce the steam turbine fault caused by loss. Simulated in this paper steam turbine rotor misalignment, rotor imbalance, rubbing and pedestal looseness, shaft crack five kinds of common vibration faults, and analyzes the spectrum analysis, statistical index in time domain analysis and wavelet gray moment experimental data for the five kinds of fault, so as to get the characteristics of various faults of steam turbine. The main research contents of this paper are as follows: 1, in the turbine rotor fault simulation test bench to simulate the steam turbine rotor misalignment, rotor imbalance, rub, bearing loose and shaft crack five kinds of vibration faults, to obtain a large number of failure test data. 2, the experimental data of the five kinds of fault condition are analyzed, the spectrum characteristics of different fault states are obtained, and the validity of the experimental data is proved. 3, calculation of five kinds of experimental data of fault waveform index, peak value, pulse index, the margin of index, kurtosis index, comparing the characteristics of these indicators under different fault conditions, and draw a conclusion: waveform can be used as an index for diagnosis of steam turbine unbalanced fault and pedestal looseness. Peak value, pulse index and margin index can be used to distinguish the steam turbine of unbalance imbalance fault, no fault, touch rubbing fault and pedestal looseness fault and axial crack fault and kurtosis index can be used to distinguish the turbine fault and pedestal looseness fault and shaft crack fault. And how to use these indexes to distinguish different fault states of steam turbine is described in detail. 4, on five kinds of fault state of the experimental data were wavelet gray moment analysis, calculate the values for the first-order wavelet gray moment at different speeds. The results show that under different fault conditions are first-order wavelet gray moment and rotational speed and summarizes its characteristics. In the rotor unbalance fault, the first order wavelet gray moment is in the range of 15-25, and the average value is about 20. In the rotor misalignment fault in most of the first order wavelet gray moment values are in the range of 8-14 changes, the average value is about 12. The first order wavelet gray moment value is changed in the range of 7-20, and the average value is about 12. Although overlaps with pedestal looseness in the first-order wavelet gray moment value range and misalignment, but between the two faults are very different, pedestal looseness, the first-order wavelet gray moment value not in individual speed suddenly increase a lot. In the low speed stage of the shaft crack fault, the first order wavelet gray moment is about 10, but the value of first order wavelet gray moment will be increased suddenly.