基于特征聚类的结构件数控加工工时预测方法

doi:10.3901/JME.2023.15.232

机械工程学报 ›› 2023, Vol. 59 ›› Issue (15): 232-246.doi: 10.3901/JME.2023.15.232

扫码分享

基于特征聚类的结构件数控加工工时预测方法

刘娟¹, 刘检华^1,2, 庄存波^1,2, 徐磊³, 翟思宽^1,2, 高庆霖¹

1. 北京理工大学机械与车辆学院北京 100081;
2. 北京理工大学长三角研究院(嘉兴) 嘉兴 314000;
3. 北京卫星制造厂有限公司北京 100094

收稿日期:2022-08-09 修回日期:2023-02-01 出版日期:2023-08-05 发布日期:2023-09-27
通讯作者: 庄存波(通信作者),男,1991年出生,副研究员,硕士研究导师。主要研究方向为装配MES技术、数字孪生技术。E-mail:zhuangdavid@bit.edu.cn
作者简介:刘娟,女,1996年出生。主要研究方向为数字孪生技术及其实现。E-mail:1024396117@qq.com;徐磊,男,1984年出生,副主任/高级工程师。主要研究方向为航天器智能制造技术E-mail:evanxl@163.com,;翟思宽,男,1997出生,硕士研究生。主要研究方向为装配工艺规划、知识图谱。E-mail:1164806656@qq.com
基金资助:
国家自然科学基金(52005042)、国防基础科研(JCKY2020203B016)和装备预先研究领域基金(80923010101)资助项目

Working-hours Prediction Method of CNC Machining of Structural Parts Based on Feature Clustering

LIU Juan¹, LIU Jianhua^1,2, ZHUANG Cunbo^1,2, XU Lei³, ZHAI Sikuan^1,2, GAO Qinglin¹

1. School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081;
2. Yangtze River Delta Research Institute of Beijing Institute of Technology, Jiaxing 314000;
3. Beijing Satellite Manufacturing Factory, Beijing 100094

Received:2022-08-09 Revised:2023-02-01 Online:2023-08-05 Published:2023-09-27

摘要/Abstract

摘要： 针对航天结构件等复杂产品工时定额结果不准确的问题,提出了基于特征聚类的结构件数控加工工时预测方法。基于航天结构件的产品特点、材料特点和加工特点,分析了航天结构件数控加工工时影响因素,并提出了基于BERT模型和K-Means算法的工时影响因素特征向量分析方法。基于K-Means聚类算法对BERT模型提取的工艺特征向量进行分组,在此基础上,基于分组结果建立了不同的遗传算法优化的BP神经网络工时预测模型,进而从工时影响因素特征分析和网络结构优化两方面,提高工时定额的准确性。最后,基于历史工艺数据完成了模型训练和预测,验证了所提方法的有效性。

关键词: 工时预测, 特征聚类, BERT模型, GA_BP算法

Abstract: Aiming at the problem of inaccurate results of working-hours quotas for complex products such as aerospace structural parts, this paper proposes a method for predicting the working-hours of CNC machining of structural parts based on feature clustering. Based on the product characteristics, material characteristics and processing characteristics of aerospace structural parts, the factors influencing the CNC machining working-hours of aerospace structural parts are analyzed, and the feature vector analysis method of working-hours influencing factors based on BERT model and K-Means algorithm is proposed. Based on K-Means clustering algorithm, the process feature vectors extracted from BERT model are grouped, and based on this grouping result, different BP neural network working-hours prediction models optimized by genetic algorithm are established, and then the accuracy of working-hours quotas is improved from both working-hours influencing factor feature analysis and network structure optimization. Finally, the model training and prediction are completed based on the historical process data, and the effectiveness of the proposed method is verified.

Key words: working-hours forecast, feature clustering, BERT model, GA_BP algorithm

中图分类号:

TP391

刘娟, 刘检华, 庄存波, 徐磊, 翟思宽, 高庆霖. 基于特征聚类的结构件数控加工工时预测方法[J]. 机械工程学报, 2023, 59(15): 232-246.

LIU Juan, LIU Jianhua, ZHUANG Cunbo, XU Lei, ZHAI Sikuan, GAO Qinglin. Working-hours Prediction Method of CNC Machining of Structural Parts Based on Feature Clustering[J]. Journal of Mechanical Engineering, 2023, 59(15): 232-246.

参考文献

[1] 孙冰妍,于功志. 基于耗材用量计算船体分段建造工时定额的一种创新方法[J]. 船舶工程,2022,44(S1):571-574. SUN Bingyan,YU Gongzhi. An innovative method for calculating man-hour quota of hull block construction based on consumable consumption[J]. Ship Engineering,2022,44(S1):571-574.
[2] 赵文浩,邢香园,王治,等. 基于知识挖掘的航天产品装配工时定额和管理系统[J]. 计算机集成制造系统,2021,27(6):1594-1604. ZHAO Wenhao,XING Xiangyuan,WANG Zhi,et al. Man-hour quota and management system of aerospace product assembly process based on knowledge mining[J]. Computer Integrated Manufacturing Systems,2021,27(6):1594-1604.
[3] 杨青海,祁国宁,黄哲人,等. 基于案例推理和事物特性表的零件工时估算方法[J]. 机械工程学报,2007,43(5):99-105. YANG Qinghai,QI Guoning,HUANG Zheren,et al. Estimation method of parts working hours based on case-based reasoning and tabular layouts of thing characteristics[J]. Journal of Mechanical Engineering,2007,43(5):99-105.
[4] 肖莹姣,陈友玲,周玉杰. 面向订单的模块化时间定额的确定方法[J]. 计算机应用研究,2011,28(4):1356-1359. XIAO Yingjiao,CHEN Youling,ZHOU Yujie. Method of determining modular time quota for order-oriented enterprise[J]. Application Research of Computers,2011,28(4):1356-1359.
[5] 刘长青,李迎光,王伟,等. 基于特征的飞机结构件数控加工工时预测模型[J]. 计算机集成制造系统,2011,17(10):2156-2162. LIU Changqing,LI Yingguang,WANG Wei,et al. Feature based man-hour forecasting model for aircraft structure parts NC machining[J]. Computer Integrated Manufacturing Systems,2011,17(10):2156-2162.
[6] 谈宏志,金礼伟,杨家荣,等. 基于BP人工神经网络的离散型车间生产调度指标预测模型的研究[J]. 科技视界,2016(3):16-18. TAN Hongzhi,JIN Liwei,YANG Jiarong,et al. Research on discrete shop scheduling index prediction model based on BP artificial neural network[J]. Science and Technology Vision,2016(3):16-18.
[7] 李亚杰,何卫平,董蓉,等. 基于制造执行系统数据采集的工时预测与进化[J]. 计算机集成制造系统,2013,19(11):2810-2818. LI Yajie,HE Weiping,DONG Rong,et al. Man-hour forecasting and evolution based on MES data collection[J]. Computer Integrated Manufacturing Systems,2013,19(11):2810-2818.
[8] 戴健伟,吉华,杨岗,等. 基于GA_BP算法的化工设备设计人工时预测[J]. 计算机集成制造系统,2013,19(7):1665-1675. DAI Jianwei,JI Hua,YANG Gang,et al. Based on GA_artificial hour prediction of chemical equipment design based on BP algorithm[J]. Computer Integrated Manufacturing Systems,2013,19(7):1665-1675.
[9] 于成龙,侯俊杰,蒲洪波,等. 基于数据挖掘的机加工生产进度预测方法[J]. 现代制造工程,2020(2):34-41. YU Chenglong,HOU Junjie,PU Hongbo,et al. Forecasting method of machine work schedule based on data mining[J]. Modern Manufacturing Engineering,2020(2):34-41.
[10] 王孟卓,尚鋆,白翱. 面向数控加工工时预测的智能车间制造大数据分析技术体系研究[J]. 现代制造技术与装备,2020(4):33-37. WANG Mengzhuo,SHANG Yun,BAI Ao. Manufacturing big data analysis architecture for working hours prediction of NC machining process in intelligent shop floor[J]. Modern Manufacturing Technology and Equipmen,2020(4):33-37.
[11] 王耀兵,马海全. 航天器结构发展趋势及其对材料的需求[J]. 军民两用技术与产品,2012(7):8,15-18. WANG Yaobing,MA Haiquan. Development trend of spacecraft structure and material requirement[J]. Dual Use Technologies & Products,2012(7):8,15-18.
[12] 董彦芝,刘芃,王国栋,等. 航天器结构用材料应用现状与未来需求[J]. 航天器环境工程,2010,27(1):41-44. DONG Yanzhi,LIU Fan,WANG Guodong,et al. Application status and future demand of spacecraft structural materials[J]. Spacecraft Environment Engineering,2010,27(1):41-44.
[13] 孙刚,易旺民,代卫兵,等. 航天器总装工艺流程优化的分析与思考[J]. 航天器环境工程,2008,25(4):381-383. SUN Gang,YI Wangmin,DAI Weibing,et al. Analysis and thinking on optimization of spacecraft assembly process[J]. Spacecraft Environment Engineering,2008,25(4),381-383.
[14] 王惠芬,杨碧琦,刘刚. 航天器结构材料的应用现状与未来展望[J]. 材料导报,2018,32(z1):395-399. WANG Huifen,YANG Biqi,LIU Gang. Application status and future prospect of materials for spacecraft structures[J]. Materials Reports,2018,32(z1):395-399.
[15] 赵振宁,王辉,虎琳. 航空航天先进复合材料研究现状及发展趋势[J]. 炭素,2021(2):24-29. ZHAO Zhenning,WANG Hui,HU Lin. The research status and development trend of aerospace advanced composites[J]. Carbon,2021(2):24-29.
[16] 田辉,王俊斌. 航空零件数控加工的特点[J]. 航空制造技术,2010(19):36-41. TIAN Hui,WANG Junbin. NC machining characteristics of aviation parts[J]. Aeronautical Manufacturing Technology,2010(19):36-41.
[17] DEVLIN J,CHANG M,LEE K,et al. BERT:Pre-training of deep bidirectional transformers for language understanding[EB/OL].2018[2019-22-10].
[18] ACHEAMPONG F A,NUNOO-MENSAH H,CHEN W. Transformer models for text-based emotion detection:A review of BERT-based approaches[J]. Artificial Intelligence Review,2021:1-41.

基于特征聚类的结构件数控加工工时预测方法

Working-hours Prediction Method of CNC Machining of Structural Parts Based on Feature Clustering

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	刘伟, 曹国忠, 檀润华, 于菲. 多生物效应技术实现方法研究[J]. 机械工程学报, 2016, 52(9): 129-141.
[2]	梁治国;徐科;徐金梧;宋强. 结构光三维测量中的亚像素级特征提取与边缘检测[J]. , 2004, 40(12): 96-99.