Research On Sensor Fault Identification And Monitoring System Of Active Magnetic Bearing

Active magnetic bearing is a high-performance complex electromechanical integration system with many disciplines.Because of its excellent performances such as high speed,high efficiency,low loss and no friction,it has been widely applied in industry in recent years.At the same time,the shortcomings of the magnetic suspension bearing also make it face more and more problems in its practical application.Because of high cost and poor the working environment of the magnetic bearing,if the safety and stability of the working process can not be guaranteed,the system will not only work normally,but also damage the bearing and cause irreparable damage.Therefore,it is very necessary to monitor the magnetic bearing system on line and study the method of fault identification.The main research work of this paper is around the two parts and the specific contents are as follows:Firstly,the fault source and fault manifestation in the mechanical and electrical control parts of the active magnetic bearing system are introduced and analyzed in detail.Then the failure causes and failure modes of the differential eddy current displacement sensor are emphatically analyzed.It is pointed out that the fault of short circuit and broken line can be directly judged by observing the amplitude of output signal of sensor,and the influence of environmental temperature and electromagnetic interference can be eliminated by differential form.However,the incomplete failure of the probe,such as the loosening of the probe,does not change significantly but has a great impact.Therefore,it is the focus of research and identification.Secondly,aiming at the failure of the differential displacement sensor probe loosening,the relationship between the fault frequencythe in differential output signal of the sensor and output signal of the controller is deduced in theory,and a fault detection and recognition algorithm based on Fourier transform(FFT)is proposed,and the simulation experiment and experimental results are carried out with LabVIEW.It is shown that when the sensor 1 is fault,the phase difference of the fault frequency in the two signal is 180 degrees;when the sensor 2 is malfunction,the phase difference of the fault frequency in the two signal is 0degrees,which is the same as the theoretical result.Finally,the active magnetic bearing on-line monitoring and fault diagnosis system is built based on LabVIEW.The system mainly includes the user login module,the dataacquisition and storage module,the time domain analysis module and the frequency domain analysis module.The data acquisition card is used to collect the output signal of the displacement sensor in real time.Experiments show that the system can collect,process and analyze signals in real time and accurately,and then identify fault sensors and alarm.