The Expert System Of Gear's Fault Diagnosis Based On Neural Network

With the development of science and technology, high performance, high efficiency, high automation and high reliability have become the tendency of mechanical equipments. Because of the advantages of the fixed transmission ratio, the lager transmission torque and the compact structure, the gear case is widely applied to adjusting velocity and transferring power. Although gear is an important component of machine equipments, it arises some faults frequently, which leads to poor performance of the whole machine in operating condition. Therefore the research on fault diagnosis of gear case brings more attention for many researchers. Also lots of results have been presented since last two decades.At present, traditional methods such as time region and frequency region analysis are mainly applied to the fault diagnosis of gear case, which can solve many practical problems effectively. In order to improve the accuracy and rapidity of fault diagnosis, there appear many advanced methods to gear's fault diagnosis, such as small wave transform method, neural network method, expert system method etc.In this thesis, on the basis of the traditional method of signal analysis, advanced analysis methods are integrated into gear's fault diagnosis. Thus a perfect and intelligent expert system of fault diagnosis based on neural method is constructed and the typical mechanism of gear's fault and its vibration characteristic are summarized. Then the methods of gear's vibration signal analysis and its fault diagnosis are described in detail. Aiming at the typical fault characteristics of gear, the method is because time region analysis can be integrated in frequency region analysis. An expert system of gear's fault diagnosis based on neural network is constructed by using of Visual C++program.The validity of the analysis method proposed in this thesis is proved through collecting and analyzing gear's fault signal. The result indicates the effectiveness of this method since that it can improve the efficiency of diagnosis and servicing for equipment and redude the cost of servicing.