基于孪生数据的产品装配过程质量预测模型  

Quality Prediction Model of Product Assembly Process Based on Twin Data

作　　者：孟冠军[1] 张磊[1] 马存徽 MENG Guan-jun;ZHANG Lei;MA Cun-hui(School of Mechanical Engineering,Hefei University of Technology,Hefei 230009,China)

机构地区：[1]合肥工业大学机械工程学院,合肥230009

出　　处：《组合机床与自动化加工技术》2022年第3期126-129,共4页Modular Machine Tool & Automatic Manufacturing Technique

摘　　要：针对制造企业在产品装配过程中普遍存在的质量问题难以预测等问题,在构建了基于量子粒子群(quantum-behaved particle swarm optimization,QPSO)参数优化的支持向量回归(support vector regression,SVR)的质量预测模型基础上,结合数字孪生技术,提出了一种基于孪生数据的QPSO-SVR质量预测模型,该模型能够预测质量数据的未来状态,为产品装配环节提供事前控制的支持。以引用发动机多工序装配环节中的曲轴装配质量参数为例,将量子粒子群参数优化随机森林回归(random forest regression,RFR)的质量预测模型与QPSO-SVR作对比,结果表明QPSO-SVR质量预测模型泛化能力更强、预测精度更高以及收敛速度更快,验证了QPSO-SVR质量预测模型具备更好的预测性能与适用性。In view of the unpredictable quality problems that are common in the product assembly process of manufacturing enterprises,the quality of support vector regression(SVR)based on quantum-behaved particle swarm optimization(QPSO)parameter optimization has been constructed.Based on the prediction model and combined with the digital twin technology,a QPSO-SVR quality prediction model based on twin data is proposed.This model can predict the future state of the quality data and provide advance control support for the product assembly process.Taking crankshaft assembly quality parameters in multi-process assembly link of engine as an example,the quality prediction model of quantum particle swarm parameter optimization random forest regression(RFR)was compared with QPSO-SVR.The results show that the QPSO-SVR quality prediction model has stronger generalization ability,higher prediction accuracy and faster convergence speed,which verifies that the QPSO-SVR quality prediction model has better prediction performance and applicability.

关 键 词：装配过程 质量预测模型 支持向量回归 量子粒子群算法 数字孪生 

分 类 号：TH16[机械工程—机械制造及自动化] TG66[金属学及工艺—金属切削加工及机床]