Study On Fault Diagnosis Of Rotating Machinery Based On Fractal Theory

In modern society, it is an important problem to state control and fault diagnosis of mechanical equipment and has being paid more and more attention to, for it has great economy and project value. In recent years, many scholars have made a lot of job on it, and put up some new methods of fault diagnosis. Time-field analysis, frequency-field analysis and correlation function analysis in traditional, have been improved in application of fault diagnosis. And gray system, obscure subjection degree, expert system, wavelet analysis and neural network, etc. have been applied in running state of qualitative analysis system, however, it is the issue having been concerned all along in academic circle that the fault state of system is analyzed quantitatively. It is the new trend in academic circle of the world that fractal theory being applied in the field of machinery fault diagnosis. Used with fractal theory, the running state of the system can be analyzed not qualitatively but also quantitatively, and then the faults of complex machinery system can be diagnosed.In this paper, it has been studied based on the fractal theory, multi-fractal theory, calculation and analysis of each fractal dimension and General dimension being set as fault feature that fault diagnosis and identification of rotating model and steam-electric generating set.Based on the theories described above, calculation formula of General dimension of least square measure and multi-fractal fault diagnosis and analysis system have been established with Borland C++ Builder 5.0, and checked by pure sinusoidal and cosine signal and sinusoidal signal with noise, and verified by the examples. It was chased down that General dimension series value changed with the variation of the sampling time and noise.General dimension and its chart were calculated and painted based onGeneral dimension. Box dimension, Information dimension and Correlation dimension can be extracted from General dimension series. And in the series of General dimension, the method of General dimension furthest distance was corrected, General dimension curve utmost approach was brought forward and fault sensitive dimension of system state was confirmed with it.Vibration signal was gained through the vibrate testing of rotating model and steam-electric generating set. It is founded that the model space samples storeroom that set General dimension and chart and fault sensitive dimension as feature group. It is brought forward that all kinds of faults can be quantitative distinguished from with sensitive dimension method, and the results of diagnosis were fine. With General dimension series single value approach and General dimension most correlation coefficient method brought forward in this paper, fault classification and identification and diagnosis of rotating model system were checked, and favorable results were obtained.Used with General dimension furthest correlation coefficient method and General dimension series single value approach, it is not only the single fault of system state, but also the coupling fault of complicated system state can be diagnosed, classified and identified. It is the new discovery of the multi-fractal theory applied in mechanical equipment diagnosis engineering, and also the new application of series single value approach applied in the field of fault diagnosis in this paper.