Research And Implementation Of Health State Evaluation Method Of Bending Machine Based On Digital Twin

China's manufacturing industry is gradually changing from "manufacturing" to "intelligent manufacturing".The newly proposed "fourteenth five-year plan" proposes to "consolidate and expand the real economy,and deeply implement the strategy of manufacturing power",which points out the direction for China's industrial intelligence.Numerical control bending machine is a kind of general equipment of numerical control machine tool to complete the accurate forming of sheet metal,which is widely used in mechanical manufacturing,aerospace,and other important production fields.At present,after the failure of the CNC bending machine,it is still based on alarm and shutdown,lacking an effective operation monitoring mechanism.Digital twin technology is used to build a digital model of bending machine,analyze its working process,monitor its running state,catch tiny abnormal in time,reduce serious faults and effectively extend its working life.To solve this problem,under the guidance of digital twin technology,the main work of this paper is as follows:(1)Based on the existing five-dimensional model of the digital twin,a model framework of digital twin applied to equipment health monitoring and evaluation is proposed.In this framework,the common faults of CNC bending machines are analyzed,and the digital twin model of multi domain fusion of CNC bending machine is preliminarily constructed based on the multi domain modeling software,and the model is calibrated with the actual operation data.The twin model can preliminarily reflect the running status of the CNC bending machine.(2)According to the operation monitoring data of the equipment and the t win data generated by the digital twin,a health assessment method of the hydraulic system based on the fusion of monitoring data and twin data is proposed.This paper focuses on the analysis of the healthy state of the accumulator in the hydraulic system when the equipment is running and uses wavelet transform,empirical mode decomposition,and Hilbert transforms to analyze and extract the characteristics of the accumulator in different states.A stack sparse Autoencoder(SSAE-Softmax)network model is designed to identify the different health states of the accumulator.Experimental results show that the algorithm is effective and twin fusion data can effectively improve the accuracy of recognition.(3)The structure of remote intelligent operation and maintenance system is designed,and the intelligent operation and maintenance system of the CNC bending machine is developed under the structure framework.At present,the system can remotely locate the position of the CNC bending machine,display the detailed operation and maintenance information of the CNC bending machine,and monitor and evaluate the health status of the bladder accumulator in the hydraulic system of the bending machine.