Research On Equipment State Prediction And Evaluation Based On Multivariate Information Model

In the process of running the production equipment,a large number of state data will be generated.These data contains the state information of running equipment。 How to extract the key information from the massive feature information in order to know the running state of the equipment and predict possible failures has become the key link of equipment maintenance.Therefore,this paper presents a method of equipment state prediction and evaluation based on multivariate information model,extract the key variables from the mass data,to forecast the change trend of the data and establish a scientific and effective evaluation model,mining and evaluating the equipment status information contained in the running data,so as to establish a complete equipment condition monitoring and fault early warning system,which can guide the maintenance of the equipment.The method proposed in this paper combines the advantages of two aspects: data processing and fuzzy evaluation,it avoids the shortcomings of the traditional methods,such as too much noise in the state data,the lack of index data,the difficult to understand the results of prediction,the subjective evaluation methods and so on,with the help of the original enterprise information platform,play the advantages of informationization of modern manufacturing industry level complete,the prediction and evaluation of equipment condition can be accurate,efficient and continuous operation,ensure the equipment maintenance plan of the scientif ic development.