Study On The Faults Diagnosis Expert System For Gear Transmission

Gear transmission is most popular in the machinery and equipment, and the gear fault is an important factor to inducing failure of machinery, so it is very critical to diagnose gear faults accurately and quickly. Nowadays, intelligent gear faults diagnosis system has been regarded to be a powerful measure to ensure the equipment working efficiently and safely. Therefore, the research on the intelligent gear faults diagnosis is significant in theory and valuable in application. This paper is based on research project supported by National High-tech R&D Program (863 Scheme) Remote and Intelligent On-line Diagnosis System for Large Rotational Machinery (Authorize number: 2001AA423200). In this paper, by investigating the vibration mechanism and characteristics of fault gear, the author generalizes gearbox faults diagnosis knowledge, and establishes 9 gear faults diagnosis diagrams. Based on the diagrams, the author integrates the intelligent gearbox faults diagnosis system including 3 core parts: knowledge express, combined inference and data engine. The major works of this thesis are listed as below: (1) Vibration mechanism of the failure gearbox is analyzed. By establishing and solving the coupled meshing vibration model of the gear transmission, the author discusses the factors impacting the vibration: meshing rigidity and error. Based on the above analysis, the vibration mechanisms of 9 typical gear faults are investigated. (2) Vibration characteristics of fault gearbox are investigated. By analyzing the vibration signal, the author extracts characteristics of 9 typical gear faults in both time domain and frequency domain. Based on these characteristics, the sensitive parameters are chosen carefully and the gear faults diagnosis knowledge is generalized. (3) Knowledge express, inference rules and web-based transmission are deeply investigated. Using graphic-network topological knowledge structure, the author establishes 9 faults diagnosis diagrams, and integrates the intelligent gearbox faults diagnosis system. (4) Two experiments are carried out: one is the contrast test of normal, worn and fracture gears for characteristics in time domain and frequency domain; the other is diagnosing two typical gear faults using intelligent gearbox faults diagnosis system.