Fractal-wavelet Based Wear And Friction Properties Study Of Low-speed Bearing

The low-speed rotating machinery are widely applied in many fields, such as wind power generation, drug manufacture, sewage treatment, petroleum chemical industry, metallurgy and papermaking and so on. Along with the trend of development of large-scale, complication, automation and continuousness, the loss will be getting bigger for the entire production and the society produced by the low-speed machinery's faults and breakdown. Especially the faults of low-speed bearing will seriously influence the normal operation because of the long-term attrition and create the huge economic loss and the significant social impact.Because the low-speed bearings are generally forced on the continual heavy load, and the tradition metering equipment are unable to examine bearing's operating frequency, the feature are not easily discovered after the low-speed bearing begin to wear. In this paper, a new measuring technique and the signal processing method are applied, and the stress situation of the wearing low-speed bearing is analyzed for coping with the difficult problems above of the extraction low-speed bearing wearing failure characteristics. The physical property and the geometry characteristic of metal abrasive are researched. Through this kind method of the comprehensive diagnosis and the recognition from macroscopic to microscopic, we can discover the early-time wearing failure accurately and effectively existing in the equipment to prevent the breakdown evolution and to raise the equipment efficiency.In this paper, the high frequency stress wave signals are used as the characteristic parameter. The breakdown signals of low-speed rotating machinery are collected to solve limited question of vibration and acoustic emission in low-speed situation. It is discovered that because the main stress waves produced by rubbing, are Rely wave which has nothing to do with the rotational speed. Because wearing fault has the different mechanism, the propagation mode and the physical property of the stress wave are different which can provide the foundation for analyzing the stress condition of wearing bearing and the extracting breakdown characteristics.The solid model of low-speed rolling bearing has been established, and the finite element method has been applied to compute distinctly the stress and strain of the complete rolling bearing and the wearing rolling bearing. The computed result showed that the stress and strain value will been change on the surface of the low-speed wearing bearing's outside roller. Moreover, it is smaller for the change value when the distance of attrition place is farther outside roller conveyer surface. The effect of surface stress and the strain distribution rule for the inner loop breakdown roller is smallest, the roller breakdown next, and the outer circle breakdown is biggest. The sensor's installment position should cause the receiving direction to aim at roller's load direction, and reduce between the contact surface of the bearing and sensor as far as possible.The breakdown stress wave signals have been analyzed using the Fourier transformation, but the breakdown feature frequency cannot be obtained after undergoing the Fourier transformation in the spectrograph. Thus it is proved that Fourier's method applying on signal processing for fault stress waves is invalid. Then, according to the characters of the fault stress waves, through the comparison of several mother wavelet, and Db10 was selected to analyze the signals, because of Db10 wavelet function have the best similarity with the fault stress waves. Multi-scaled decomposition was carried out in the analysis, and then, the D3 and D4 was reconstructed based on the comparison of the scale energy. At last, the feature frequency of fault stress waves was extracted successfully. It is showed that the wavelet-stress wave method is suitable to the fault diagnosis of low-speed rolling bearings.The analysis of low-speed bearing's lubrication condition is very important. On the basis of fractal theory, a number of tests are carried on the bearings including the dynamic and static load test, the superficial appearance test, vibration test and so on. Massive calculation is computed according to oil-retaining bearing surface appearance and the vibration performance's fractal characteristic, the fractal feature and the fractal parameters. The results indicate that the correlation dimension of vibration signals will increases gradually along with the operating time in the different friction stage. Image recognition method is proposed using fractal and chaos. The trajectory diagram demonstrates the vibration the time series, but the fractal dimension falling into the space of points attributes the probability distribution after the low-speed bearings breakdown. The chaos graphs show the final act in lower subspace of a complex dynamic system. A three-dimensional recognition and diagnosis method is formed through three aspects of the abrasive distribution, the stress variation and the extent of damage.