Research On Tension Measurement And Vibration Prediction For The Separator System Of Vertical Roller Mill

The ball mill has been replaced by Vertical Roller Mill(VRM)and High Pressure Rolling.Because of its low-power consumption and high grinding efficiency,VRM has become an important choice for the cement production grinding system.VRM is a kind of large mechanical equipment in the field of cement industry,and it has a complex mechanical and electrical structure.The running environment and improper operation will affect the safe running of VRM.If these situations cannot be monitored and predicted in time,then it will lead to shutdown and even more serious consequences.In this paper,a new Fiber Bragg Grating(FBG)bolt tension sensor has been designed for the separator system of VRM.Through the related signal processing technology,a new feature extraction method for the vibration signal has been proposed.By introducing the machine learning method,the prediction model of the upper shell vibration signal of the separator system of VRM is achieved.Finally,the tension measurement and vibration prediction of the separator system of VRM is realized.The main work is as follows:(1)The shell connecting bolts of the separator system of VRM is regarded as the research object.Combined with site installation requirements,the special FBG bolt tension sensor has been designed based on FBG sensing technology.The SolidWorks and ANSYS have been used to the simulation analysis of this bolt tension sensor,and its performance has been detected by experiments.The results show that this FBG bolt tension sensor can response the connecting bolt tension in time.(2)The Wavelet Noise Elimination and the Local Mean Decomposition have been studied.A method used to extract the features of vibration signal has been put forward.In order to solve the end effect problem of Local Mean Decomposition,an extension method based on the data expansion has been built.At last,the measured data are extracted by MATLAB,and the useful features are extracted.(3)The Least Square Support Vector Machine regression algorithm has been adopted to predict the upper shell vibration signal of the separator system of VRM by using the Particle Swarms Optimization algorithm and the features which is extracted in the last step.MATLAB simulation software has been used to simulate and analyze the measured data,and then compared the prediction results with the measured data.The results show that the prediction model proposed in this paper has a good effect on the prediction of the upper shell vibration signal of the separator system of VRM.