Research On On-Line Monitoring System And Fault Early Warning Technology Of Motor Intelligent Bearing

As an important driving equipment in modern factories,motors are widely used in various fields of national economy.The motor is prone to failure in the process of use,with bearing failure accounting for about one-third.Bearing failure will cause the motor to stop or replace equipment.At home and abroad,real-time monitoring systems for large-scale motor bearings have been studied,but there are widespread problems such as disassembling the motor during installation,destroying the bearing and discrete installation,high cost of the device,redundancy of various measuring devices,and low intelligence.Monitoring of small and medium-sized motor bearings is often ignored.In this thesis,the domestic and international motor bearing monitoring technology and research status are studied in detail,and the bearing failure reason and fault characterization parameters are analyzed.Then the bearing fault diagnosis and early warning characteristic parameters are determined,and an intelligent motor bearing fault warning system which can monitor and analyze the running state parameters of the bearing with big data on-line is proposed and designed.The EMD hard threshold method is used to denoise the signal,and the signal-to-noise ratio is improved.Then the “new” signal is decomposed by EEMD to extract the singular value of the envelope matrix of each order signal,and the more distinct vibration signal fault eigenvalue is extracted.The vibration characteristic value is processed by PCA dimensionality reduction,which solves the dimensional explosion problem of ANFIS model.On this basis,combined with the structure and defect location of the bearing,a fault warning model of vibration signal bearing based on EEMD-PCA-ANFIS under the defect state was established,and the fault diagnosis ofbearing defect state was realized,and the correct rate is 93.3%.Aiming at the diagnosis of bearing oil shortage or less oil condition,a second warning model of bearing temperature rise is proposed.Combined with the above-mentioned vibration signal bearing fault early warning model,bearing fault diagnosis and early warning based on joint analysis of vibration and temperature rise trend is established.model.The experimental results show that the bearing fault diagnosis and early warning model based on the combined analysis of vibration and temperature rise can accurately determine the bearing lack of oil or less oil.The intelligent motor bearing on-line monitoring system and the bearing fault diagnosis and early warning model based on the joint analysis of vibration and temperature rise trend proposed in this thesis are not only applicable to common motor bearings,but also applicable to other rotating machinery equipment.It not only effectively solves the false alarm and missed alarm problems of the traditional early warning model,improves the accuracy of bearing operation monitoring and fault warning,but also lays a foundation for realizing the intelligent algorithm of multi-information fusion fault warning,and provides a guarantee for the safe,continuous and stable operation of rotating equipment.