| With rapid development of modern industry,the equipment is complex,with many components and difficult maintenance at this stage.It is necessary to identify the equipment health state in time and ensure the safety and reliability in the working process.It has become one of the important goals of modern enterprise management.The Internet of things,sensors and other technologies have enriched the means of equipment state measurement.Many enterprises have accumulated a large number of equipment state data.It is possible to study equipment health state analysis based on data-driven.According to the above situation,health state recognition and state data processing are combined in this paper,the main work are as follows.(1)The equipment health state is represented by multiple equipment health state parameters.And the multi parameter data of running state provided by spot inspection system was taken as the sample.The principal component analysis method was used to analyze the data,and the equipment state data was fused in the process.The important health state information was selected,and the health state assessment index system was established.(2)In order to identify equipment health state,the relationship between equipment health state and equipment health state level should be built.The health state assessment model was built by fusing history,real-time equipment multi-state parameters using deep learning.And the model input,model parameters and model output were designed.Hidden layer number,learning rate,hidden layer node number were researched on model verification and optimization.(3)In view of the above theories and methods,a corresponding software system was developed on the basis of experimental verification and enterprise application.According to the design objectives of the system,the functional modules of the system were designed,and the key technologies,software configuration and hardware configuration of the system development were studied. |
PDF Full Text Request