Research On Vibration Monitoring And Fault Diagnosis Of Rotating Machinery Based On LabVIEW

With the rapid development of modern science and technology, the level of China’s industrialization is improved constantly. To satisfy the need of industrial production, machinery equipment is developing toward the trend of automation, large-scale, high-speed rapidly.The rotating machinery, as one of the most common forms of machinery equipment, is widely used in important areas such as aviation, electric power, mining, metallurgy, petrochemical and so on. Any component in rotating machinery goes wrong may lead the equipment or the entire production line to invalid, causing serious accident.Therefore, to ensure the stable and reliable operation, carrying out the research on vibration monitoring and fault diagnosis of rotating machinery is very important for improving the security and economy of company.Firstly, the method of rotating machinery fault diagnosis is researched in the paper. Support vector machine has good results when dealing with problems on non-linear, over-learning, small sample size. Beacause of the blindness when people extracting features, the result of diagnosis is usually inaccurate.So the method based on support vector machine and neighborhood rough set is proposed in the paper. It can discard redundant information from the raw data, filter out the valuable characteristic attributes, create a support vector machine classifier model though the feature information and obtain the diagnosis result quickly and accurately.Besides, the system of rotating machinery online monitor and fault diagnosis aimed at the usual faults of rotor is researched and designed in this paper. It is based on the platform of LabVIEW software,utilizes the sensors, signal conditioning, PCI data acquisition card, industrial computer and other hardware devices and combines with Matlab system. The system is able to achieve the rotating machinery vibration online monitoring, signal analysis and processing, data storage and playback and alarm because of excessive vibration. Besides, it also uses the SVM multi-classification algorithm and Hybrid Programming, achieves the function of intelligent fault diagnosis.At last, the common fault types of rotating machinery rotor are summarized and characteristics of the turbine rotor vibration faults are analysised in the paper. And Bently-RK4 rotor vibration test bench verifies the effectiveness of the system.