Research On Fault Diagnosis Of Gear System Based On Blind Signal Separation

The gear system, which is considered as the necessary mechanical transmission equipment in modern industry, has many advantages such as large driving force, exact transferring motion, smooth transmission and so on. Gear system is becoming more and more important in transferring power and motion along with the development of large-scale, high efficiency and high strength equipments. However, the gears and gear boxes would be damaged and become failure easily because of its complex structure and bad running conditions. Accordingly, it is significant to research on condition monitoring and fault diagnosis for gear system by advanced technologies, which could not only change the current postmortem servicing and regular examining to repairing according to the specific state, but also bring more economical and social benefit.This thesis is focused on the typical fault mechanism and vibration characteristic of gear system. It is mainly including two fault feature extracting techniques, one of which is the improved algorithm of the singularity value decomposition (SVD) about track matrix of attractor reconstructed by time series, and the other is the new method of using both the SVD algorithm and the blind signal separation (BSS) algorithm. Both the methods mentioned above are used in extracting the fault features, which could provide a new thought for faults diagnosis of the gear system.The main research contents and the key conclusions are shown as follows:(1)The existing singularity value decomposition (SVD) algorithm is improved in the second chapter. After studying the fundamental of the SVD about track matrix of attractor reconstructed by time series, the autocorrelation analysis is introduced, which improved the current algorithm and made it more logical. The analysis of the simulation signals and the measured signals from gear box shows that it is successful for the improved SVD algorithm in extracting modulated signals mixed by strong noises, which is significant to the gear system fault diagnosis.(2)The third chapter is mainly on the fundamental and application of the BSS algorithm. Firstly, the batch-processing algorithm such as JADE, the self-adaptive algorithm such as Infomax and the fixed-point algorithm FastICA are researched in detail. And their characteristics are also researched. The simulation shows that both the JADE algorithm and the FastICA algorithm are effective in separating source signals from multi-signals, which provides a new way for the gear system fault diagnosis.(3)The experiment for gear box faults diagnosis is carrying out in the fourth chapter. According to the laboratorial condition, some works are done, including designing the experimentation, improving the former gear box faults diagnosis platform, and carrying out the experiment with different diagnosis styles.(4)The fifth chapter is focused on a new diagnosis technology based on both SVD and BSS. The signal-to-noise-ratio (SNR) of the measured signal is enhanced by the SVD algorithm at first. Then the signals after processing are separated by the BSS algorithm. The separation of typical diagnosis signals is quite successful by using the new method. The characteristic of the separated signals is quite accordant with the setting faults. At the same time, the separated results show that both the SVD algorithm and the BSS algorithm can do well in processing measured signals.At last, the main results of this dissertation are summarized in the last chapter, and some possible valuable research directions are also pointed out.