机电工程技术2025，Vol.54Issue(1)：9-16,85,9.DOI:10.3969/j.issn.1009-9492.2025.01.002
基于数字孪生的智能辊压成形专家系统构建方法研究
Research on Construction Method of Intelligent Roll Forming Expert System Based on Digital Twin
摘要
Abstract
With the rapid advancement of Industry 4.0,the manufacturing sector is undergoing a transformation toward digitalization and intelligence.However,challenges such as low data utilization,prolonged debugging cycles,and unstable product yield persist in complex processes.Addressing these issues,focuses on the roll forming process is focused on,and an expert system construction method is introduced.The proposed framework comprises six interconnected layers:physical,virtual,data,connection,application,and expert layers.A network architecture supporting multiple communication protocols is developed to enable the acquisition,processing,and analysis of heterogeneous data sources,such as sensor signals and imaging data,ensuring precise perception of the forming process.Additionally,a virtual workshop and interactive interface are created using Unity and VR technologies,while the integration of 5G communication and PLC control achieves bidirectional virtual-to-physical synchronization with a latency of 40 ms,significantly optimizing the coordination of servo motors.Furthermore,an expert knowledge base is constructed utilizing orthogonal experimental design to systematically analyze the effects of material properties and process parameters on the rebound behavior of cap-shaped parts.This analysis aims to inform and optimize the manufacturing process.This research significantly enhances the process optimization capabilities and quality control precision of roll forming,shortens debugging cycles,reduces production costs,and offers a systematic solution for the intelligent transformation of roll forming processes.It provides valuable insights and a practical pathway for the digitalization and intelligence of manufacturing.关键词
智能辊压/专家系统/数字孪生/回弹预测Key words
intelligent rolling/expert system/digital twin/springback prediction分类
矿业与冶金引用本文复制引用
吴豫席,胡鑫宇,任银旺,张正勋,郭俊朗,熊自柳,肖骏峰,孙勇..基于数字孪生的智能辊压成形专家系统构建方法研究[J].机电工程技术,2025,54(1):9-16,85,9.基金项目
深圳市科技创新委员会(JCYJ20220818103200001,ZDSYS20210813095534001,JCYJ20210324123202008,JCYJ20210324115412035) （JCYJ20220818103200001,ZDSYS20210813095534001,JCYJ20210324123202008,JCYJ20210324115412035）
广东省基础与应用基础研究基金委员会(2022A1515010363,2021A1515110880) （2022A1515010363,2021A1515110880）
深圳市龙华区数智化成形系统装备协同创新平台(J10M00792305290004) （J10M00792305290004）
中南大学国家重点实验室高性能复杂制造(Kfkt2021-05) （Kfkt2021-05）
深圳市龙华区科技创新委员会(J10M00792305290004) （J10M00792305290004）
