| One of the important functions of the equipment health management system,Prognostic and Health Management(PHM),can monitor the working data of the equipment in real time,find out the cause and location of the fault in time,and predict the future operation status of the equipment,so as to help enterprises to check the equipment failure in advance and reduce the loss.Taking high temperature vacuum press as an example,this paper puts forward a general PHM system suitable for small and medium-sized equipment,which can meet the needs of remote monitoring,real-time understanding of equipment work,rapid detection of problems and solving problems,improve work efficiency,and bring greater economic benefits to enterprises and society.In this paper,case-based Reasoning(CBR)is selected to study,and the corresponding fault prediction model is designed.The CBR algorithm is improved,and the accuracy of the improved algorithm is checked,so as to effectively improve the stability and reliability of product assembly system.This paper takes the high temperature vacuum press produced by Xianyang Weidi Electrical and Mechanical Technology Co.,Ltd.as a case study,uses sensor technology to collect data and information generated by equipment work,verifies the proposed model,and achieves fault prediction and health management.Real-time data will be generated during the operation of equipment,which can quantitatively indicate the quality of equipment operation.Firstly,the characteristic attribute parameters of the sensor acquisition equipment are stored in the background.If the system wants to calculate and access the background data in a more convenient and regular way,it must store them in a special organizational form,and finally form a case base.The systembuilds a B-tree structure of the secondary case base,and stores historical cases in the file system for fault prediction and health management.At the same time,before the fault diagnosis and fault prediction,the system needs to determine the characteristic attribute weights of the three data “temperature”,“vacuum degree” and “pressure”.In order to improve the efficiency of PHM technology,this paper studies the search algorithm of CBR prediction method,Nearest Neighbor algorithm(KNN),and proposes a clipping and dividing method,which reduces the number of searches by clipping the search area to improve the efficiency of the algorithm.The simulation results show that the efficiency of the algorithm can be improved by 25.08%,14.48% and 0.58% respectively when the target case is in the middle of the case attribute space,deviating 70% and 90%.When the target case is located in an area beyond 10% of the edge of the case attribute space,the efficiency of clipping and dividing is close to that of cyclic traversal.Compared with the traditional cyclic traversal method,the cut-and-conquer algorithm has advantages in efficiency.Finally,the PHM system of high temperature vacuum press is realized. |
