The Bearing’s Anomaly Weak Signal Recognition Based On Chaos Weak Signal Detection Technology

Nowadays, Rolling bearing is one of the most used mechanical components, at scene almost all of the equipments are rotating machinery, and you can see the figure in all of the equipments. Every year for rotating machinery fault caused economic losses more than tens billions, and the bearing fault occupies as high as1/3of rotating machinery fault. Every failure is starting from the small fault, until the cause serious accident. So finding the fault at the beginning is very important.As a nonlinear branch of science--chaos theory, now has already applied in different engineering field. Applied the chaos theory to the weak signal detection is one of the main methods of studying the bearing failure. This paper based on the basis of the research at home and abroad in terms of chaos theory, and widely absorbs the profound understanding and research results of chaotic theory in various fields, and use the chaos theory to identify the fault signal of the bearing, final we can determine the location of bearing fault.This paper mainly does the following several aspects of contents:Firstly, this paper studies the conceptual framework、the development history and the main achievements of chaos theory, and give the discriminated method of judge the usual faults.This paper briefly introduced the small-signal, provided some simple processing method and reviewed the signal’s performances.Secondly, this paper established the model to detect the chaotic weak periodic signal. Use the dynamics of Duffing chaotic system to detect the weak periodic signal, add the weak periodic signal to the system when it is in chaotic state, and observe whether the phase track diagram of Duffing equation is mutated. It can reflect if exist the weak periodic signal which is bigger SNR.Lastly, make the traditional Duffing equation to further improvement, and use the improved equation to detect the pitting fault of bearing vibration signals. The bearing fault signal is a very complicated composite signal which formed by several different signal mixed together, so the bearing fault signal is chaos in a sense. Each different type of bearing and the different location of the bearing all have the different characteristic signals, and we according to the uniqueness of characteristic signal to put forward the method to detect the bearing fault which based on the chaos theory. This method is only sensitive to the characteristic signal, and has good inhibition to the noise signal. There have a lot of useless signals in the original signal what we collected, in order to reduce the amount of calculation, we should preliminary noise reduction use the way of wavelet packet de-noising before we analyse the signal with Duffing equation. The final experiment results show that this method can identify the pitting fault of the bearing outer ring.The final experiment results show that this method can identify the pitting fault of the bearing outer ring, so it can be seen that this method is an effective method to discover the fault of rolling bearing.