RFID数据驱动的柔性制造车间产品加工周期预测  

Research on RFID Data-Driven Product Cycle Time Prediction for Flexible Manufacturing Workshop

作　　者：吴立辉 任俊飞[2] 李元生 张中伟 WU Lihui;REN Junfei;LI Yuansheng;ZHANG Zhongwei(School of Mechanical Engineering Shanghai Institute of Technology,Shanghai 201418,China;BOE Technology Group Co.,Beijing 100176,China;School of Mechanical and Electrical Engineering He’nan University of Technology,He’nan Zhengzhou 450001,China)

机构地区：[1]上海应用技术大学机械工程学院,上海201418 [2]京东方科技集团股份有限公司,北京100176 [3]河南工业大学机电工程学院,河南郑州450001

出　　处：《机械设计与制造》2025年第1期265-271,共7页Machinery Design & Manufacture

基　　金：国家自然科学基金项目(U1704156);河南省科技攻关计划项目(212102210357)。

摘　　要：基于RFID数据的产品加工周期预测对提升柔性制造车间生产效率和管理水平具有重要意义。为对柔性制造车间产品加工周期进行准确预测,首先,在分析描述RFID原始数据获取方式的基础上,对RFID原始数据进行了预处理并与车间生产事件进行了集成处理;其次,基于RFID数据并围绕车间运行状态、生产有效时间和设备工况水平等因素构建了柔性制造车间关键特征集;最后,基于关键特征集设计了基于条件互信息最大算法和BP神经网络集成的预测方法。实验研究表明,这里提出的基于RFID数据的柔性制造车间产品加工周期预测方法是有效的。Product processing cycle prediction based on RFID data is of great significance to improve the production efficiency and management level of the flexible manufacturing workshop.To accurately predict the product processing cycle in the flexible manufacturing environment,the acquisition mode of RFID raw data is described and the RFID raw data is preprocessed and inte⁃grated with the workshop production events firstly.Then,the key feature collection of the flexible manufacturing workshop is con⁃structed by considering the factors such as workshop operating states,effective production times,and equipment working condi⁃tions,which are extracted based on the preprocessed RFID raw data.Finally,with the key feature set,a conditional mutual in⁃formation maximum and BP neural network-based prediction method is proposed to predict the product processing cycle.The ex⁃perimental results show that the product processing cycle prediction method with RFID data is effective.

关 键 词：柔性制造车间 加工周期 预测 RFID数据 关键特征 神经网络 

分 类 号：TH16[机械工程—机械制造及自动化] TH166