The Study Of Gear Box Fault Diagnosis Based On The Theory Of The Extension-Fuzzy

As the most common mechanical equipment,gear box is used widely in the industry, agriculture, transportation and other fields,gear box is the essential mechanism which is used for the connection and transmission in mechanical equipment,and its working condition directly relates to the running state of the machine. Therefore, the state evaluation and fault diagnosis of the gear box has a great significance. The vibration signals of the gear box contains all the information of its running state. When the working condition of the gear box is changed, its vibration signals will change. Studying the characteristics of the vibration of the gear box, extracting the effective signals and analysing them accurately, can undertake the state evaluation and fault diagnosis of gear box. Taking effective measures to eliminate the fault,it can improve the safety and reliability of the gear box and reduce maintenance costs and reduce unnecessary loss by taking effective measures to eliminate fault.Extension theory, which develops rapidly in recent years, is a method used to solve contradiction problems. Its use the extension matter-element model of qualitative to research characteristic parameters, and the theory of extension to determine the membership degree of the characteristic parameters, so it provides a formal and modeling, method which combines qualitative and quantitative to solve the determine of fault problem. The combination of extension theory and fuzzy theory can solve the problem of fuzziness, incompatibility and contradiction in the fault of gear box. Finally, the feasibility and effectiveness of the method are verified in the experiments. The main work of this thesis is as follows:First, this thesis briefly introduces the structure of the gear box and its main fault modes briefly, and analyzes the common fault types, fault vibration signal characteristics and failure mechanism of gear and rolling bearing.Secondly, on the basis of the knowledge of vibration mechanism and state evaluation of the gear box. the failure experiment platform of gear box is set up to simulate the normal state, the wear state and the broken teeth state of the gear,and the related characteristic parameters of time domain and frequency domain are measured.Finally, the extension fuzzy fault diagnosis model is established, and the extension fuzzy model is optimized by using the principal component analysis to select the feature parameters. The entropy method is used for determining each characteristic index weight objectively, and determining each characteristic parameter weights quantitatively. According to the results of the experiment, it shows that the method is feasible for the fault diagnosis of gear box, and it provides a new tool andmethod for mechanical fault diagnosis.