Research On Health State Assessment And Maintenance Strategy Of Vertical Roller Mill

With the increasing functionality of the equipment,its complexity is also increasing,the company's reliability requirements of the equipment is getting higher and higher,the health of equipment has gradually become the focus.Vertical roller mill(referred to as vertical mill)as the key equipment in the field of grinding,its health status directly affect the production efficiency and economic efficiency.As the vertical mill with complex structure,many parts and working in poor conditions,the traditional fault diagnosis is not only costly,but also difficult to accurately grasp the state of vertical mill,resulting in frequent downtime maintenance,high maintenance costs and other issues.Therefore,here has done some research on the above issues,as follows:Firstly,according to the characteristics of vertical mill structure,a health evaluation model was established.The model was divided into three evaluation levels:machine-subsystem-component evaluation level.By analyzing the fault characteristics of each component,the parameters were selected to evaluate vertical mill.And the health value and grade were determined to describe the health of vertical mill.Secondly,according to the model,the health status of vertical mill is evaluated layer by layer.In the evaluation process,the combination of weight method was used to determine the weight of each level,and introducing a variable weight function to reflect the fluctuation of the index.In the component layer,gray cloud model was used for gray clustering estimation.In order to eliminate the information conflict of clustering assessment,the improved method of evidence fusion is used to evaluate the systematically.The comparison between the evaluation and the actual situation shows the evaluation method can better evaluate the health status of the vertical mill.Thirdly,using the results of health assessment to predict the residual life of the mill.Combined with the historical monitoring data,the health status of the grinding mill was evaluated and quantified several times,and the change law of health degree of the vertical grinding system was obtained.Then by using the LS-SVM and AR(p)combined model,the multi-step prediction of the future health degree of the verticalmill is realized,and the remaining life is obtained.The experimental comparison shows that the forecasting method is effective.Finally,combined with the vertical mill maintenance way,based on the prediction of the residual life of vertical mill,the maintenance strategy of multiple preventive repair and periodic renewal is put forward.With maximum availability and minimum expense rate as the optimization target,the optimal preventive maintenance period and the maximum number of repairs were obtained.Finally,the validity of the optimization model was verified by sensitivity analysis.