Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2021, Vol. 55 Issue (3): 441-447    DOI: 10.3785/j.issn.1008-973X.2021.03.003
机械工程     
机加零件质量预测与工艺参数优化方法
于勇1,2(),薛静远1,戴晟1,鲍强伟1,赵罡1,3
1. 北京航空航天大学 机械工程及自动化学院，北京 100191
2. 北京市高效绿色数控加工工艺及装备工程技术研究中心，北京 100191
3. 北京航空航天大学 航空高端装备智能制造工信部重点实验室，北京 100191
Quality prediction and process parameter optimization method for machining parts
Yong YU1,2(),Jing-yuan XUE1,Sheng DAI1,Qiang-wei BAO1,Gang ZHAO1,3
1. School of Mechanics and Automation, Beihang University, Beijing 100191, China
2. Beijing Engineering Technological Research Center of High-Efficient and Green CNC Machining Process, Beijing 100191, China
3. Key Laboratory of Aeronautics Smart Manufacturing, Beihang University, Beijing 100191, China
 全文: PDF(1052 KB)   HTML
摘要：

为了有效利用机加零件工艺信息和检测信息，提出基于机器学习算法的质量预测与工艺参数优化方法. 以集成工艺信息和检测信息的基于模型定义（MBD）模型为输入，通过对三维建模软件的二次开发实现参数提取，并建立结构化数据集. 利用多种机器学习分类器构建基于工艺参数与质量分类标签的质量预测模型. 结合信息增益算法对所有工艺参数进行优先级排序，筛选出对质量影响最大的工艺参数；开发质量预测与工艺参数优化工具集，利用梯度提升树模型优化对质量影响最大的工艺参数. 以某航空企业提供的铣削实验数据验证所提出方法的有效性和可靠性. 验证结果表明，该方法能够较好地实现机加零件的质量预测和工艺参数优化.
关键词： 基于模型定义（MBD）;  机加零件;  工艺信息;  检测信息;  质量预测;  工艺参数优化;  机器学习    
Abstract:

A novel method based on machine learning algorithms was proposed to realize the quality prediction and the process parameter optimization, in order to reuse the process information and the inspection information of machining parts effectively. A model-based definition (MBD) model which was integrated with process information and inspection information was treated as input. Process and inspection parameter extraction based on the MBD model was developed and the corresponding structured data set was established through the secondary development of three-dimensional modeling software. Several classifiers in machine learning were used to construct the quality prediction model based on process parameters and quality classification labels. Combining the information gain algorithm, after sorting all process parameters, the process parameter that had the greatest impact on quality was selected. Quality prediction and process parameter optimization tool set was developed to realize the optimization of the selected parameter by using the gradient boost decision tree algorithm. The validity and the reliability of the proposed method were verified by the milling experiment data provided by an aviation company. Results show that the proposed method can realize the quality prediction and process parameter optimization of machining parts effectively.
Key words: model-based definition (MBD)    machining parts    process information    inspection information    quality prediction    process parameter optimization    machine learning
收稿日期: 2019-10-31 出版日期: 2021-04-25
CLC:  TP 391  
基金资助: 工信部相关专项科研技术研究资助项目（**-2017-*-**）
作者简介: 于勇（1977—），女，副教授，从事复杂产品数字化协同研制、数字孪生模型定义与应用、产品构型管理研究.orcid.org/0000-0001-9524-7230. E-mail： yuyong@buaa.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
于勇
薛静远
戴晟
鲍强伟
赵罡

引用本文:

于勇,薛静远,戴晟,鲍强伟,赵罡. 机加零件质量预测与工艺参数优化方法[J]. 浙江大学学报(工学版), 2021, 55(3): 441-447.

Yong YU,Jing-yuan XUE,Sheng DAI,Qiang-wei BAO,Gang ZHAO. Quality prediction and process parameter optimization method for machining parts. Journal of ZheJiang University (Engineering Science), 2021, 55(3): 441-447.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2021.03.003    或    http://www.zjujournals.com/eng/CN/Y2021/V55/I3/441

i 工艺参数1 工艺参数2 $\cdots$ 工艺参数 $n$ 零件质量标签
0 $x_1^{(0)}$ $x_2^{(0)}$ $\cdots $ $x_n^{(0)}$ ${L^{(0)}}$
1 $x_1^{(1)}$ $x_2^{(1)}$ $\cdots $ $x_n^{(1)}$ ${L^{(1)}}$
$\vdots $ $\vdots$ $\vdots$ $ $ $\vdots$ $\vdots$
m $x_1^{(m)}$ $x_2^{(m)}$ $\cdots $ $x_n^{(m)}$ ${L^{(m)}}$
表 1  零件质量预测的结构化数据格式
图 1  零件质量预测流程
i 工艺参数1 $\cdots $ 工艺参数 $n$ 零件质量标签 工艺参数 $o$
0 $x_1^{(0)}$ $\cdots $ $x_n^{(0)}$ ${L^{(0)}}$ $x_o^{(0)}$
1 $x_1^{(1)}$ $\cdots $ $x_n^{(1)}$ ${L^{(1)}}$ $x_o^{(1)}$
$\vdots $ $\vdots $ $\vdots$ $\vdots$ $\vdots$ $\vdots$
m $x_1^{(m)}$ $\cdots $ $x_n^{(m)}$ ${L^{(m)}}$ $x_o^{(m)}$
表 2  工艺参数优化的结构化数据格式
图 2  机加零件质量预测与工艺参数优化系统框架
工艺参数 信息增益
径向切深 0.022 702 294
轴向切深 0.141 402 794
主轴转速 0.240 937 382
表 3  信息增益计算结果
图 3  检测信息结构化格式
图 4  工艺信息结构化格式
图 5  集成工艺信息和检测信息的MBD模型
图 6  工艺信息与检测信息提取流程
图 7  零件质量预测与工艺参数优化工具集用户界面
模型名称 十折交叉验证平均准确率
LR 0.658 0
SVM 0.813 2
XGBoost 0.828 8
GBDT+LR 0.863 1
表 4  四类分类器的准确率
数据
编号 		径向
切深 		轴向
切深 		主轴
转速 		表面粗
糙度 		原始质量
标签 		预测零件
质量
0 4.2 2.2 13500 3.62 0 0
1 4.2 2.2 14000 3.23 0 0
2 4.2 2.2 14500 3.14 1 1
3 4.2 2.2 15000 2.95 1 1
4 4.2 2.2 15500 3.01 1 1
表 5  零件质量预测结果
图 8  工艺参数优化结果
1 QUINTANA V, RIVEST L, PELLERIN R, et al Will model-based definition replace engineering drawings throughout the product lifecycle? A global perspective from aerospace industry[J]. Computers in Industry, 2010, 61 (5): 497- 508
doi: 10.1016/j.compind.2010.01.005
2 ALEMANNI M, DESTEFANIS F, VEZZETTI E Model-based definition design in the product lifecycle management scenario[J]. The International Journal of Advanced Manufacturing Technology, 2011, 52 (1–4): 1- 14
doi: 10.1007/s00170-010-2699-y
3 HEDBERG T, LUBELL J, FISCHER L, et al Testing the digital thread in support of model-based manufacturing and inspection[J]. Journal of Computing and Information Science in Engineering, 2016, 16 (2): 1- 10
4 于勇, 范胜廷, 彭关伟, 等 数字孪生模型在产品构型管理中应用探讨[J]. 航空制造技术, 2017, 526 (7): 41- 45
YU Yong, FAN Sheng-ting, PENG Guan-wei, et al Study on application of digital twin model in product configuration management[J]. Aeronautical Manufacturing Technology, 2017, 526 (7): 41- 45
5 于勇, 胡德雨, 戴晟, 等 数字孪生在工艺设计中的应用探讨[J]. 航空制造技术, 2018, 61 (18): 26- 33
YU Yong, HU De-yu, DAI Sheng, et al Study on application of digital twin in process planning[J]. Aeronautical Manufacturing Technology, 2018, 61 (18): 26- 33
6 于勇, 周阳, 曹鹏, 等 基于MBD模型的工序模型构建方法[J]. 浙江大学学报: 工学版, 2018, 52 (6): 1025- 1034
YU Yong, ZHOU Yang, CAO Peng, et al In-process model construction method based on model-based definition model[J]. Journal of Zhejiang University: Engineering Science, 2018, 52 (6): 1025- 1034
7 田富君, 田锡天, 耿俊浩, 等 基于模型定义的工艺信息建模及应用[J]. 计算机集成制造系统, 2012, 18 (5): 913- 919
TIAN Fu-jun, TIAN Xi-tian, GENG Jun-hao, et al Model-based definition process information modeling and application[J]. Computer Integrated Manufacturing Systems, 2012, 18 (5): 913- 919
8 胡祥涛, 程五四, 陈兴玉, 等 基于MBD的产品信息全三维标注方法[J]. 华中科技大学学报: 自然科学版, 2012, 40 (Suppl.2): 60- 63
HU Xiang-tao, CHENG Wu-si, CHEN Xing-yu, et al Full 3D annotation of product information based on MBD[J]. Journal of Huazhong University of Science and Technology: Natural Science Edition, 2012, 40 (Suppl.2): 60- 63
9 叶柏超. 基于MBD的检测工艺模型定义与表达[D]. 沈阳: 沈阳航空航天大学, 2016: 10-11.
YE Bai-chao. Expression and definition of inspection process model based on MBD [D]. Shenyang: Shenyang Aerospace University, 2016: 10-11.
10 毛贝, 刘志存, 梅中义 基于MBD的工艺模型集成化建模与表达[J]. 现代制造工程, 2017, (5): 102- 109
MAO Bei, LIU Zhi-cun, MEI Zhong-yi Integrated modeling and expression of the process model definition based on MBD[J]. Modern Manufacturing Engineering, 2017, (5): 102- 109
11 TIRKEL I. Cycle time prediction in wafer fabrication line by applying data mining methods [C]// 2011 IEEE/SEMI Advanced Semiconductor Manufacturing Conference. Piscataway: IEEE, 2011: 1-5.
12 朱雪初, 乔非 基于工业大数据的晶圆制造系统加工周期预测方法[J]. 计算机集成制造系统, 2017, 23 (10): 2172- 2179
ZHU Xue-chu, QIAO Fei Cycle time prediction method of wafer fabrication system based on industrial big data[J]. Computer Integrated Manufacturing Systems, 2017, 23 (10): 2172- 2179
13 汪俊亮, 秦威, 张洁 基于数据挖掘的晶圆制造交货期预测方法[J]. 中国机械工程, 2016, 27 (1): 105- 108
WANG Jun-liang, QIN Wei, ZHANG Jie Data mining for orders' LT forecasting in wafer fabrication[J]. China Mechanical Engineering, 2016, 27 (1): 105- 108
doi: 10.3969/j.issn.1004-132X.2016.01.017
14 ABAJO N, DIEZ A B, LOBATO V, et al. ANN quality diagnostic models for packaging manufacturing: an industrial data mining case study [C]// Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2004: 799-804.
15 SHANNON C E A mathematical theory of communication[J]. Bell System Technical Journal, 1948, 27 (3): 379- 423
doi: 10.1002/j.1538-7305.1948.tb01338.x
16 BOSER B E, GUYON I M, VAPNIK V N. A training algorithm for optimal margin classifiers [C]// Proceedings of the 5th Annual Workshop on Computational Learning Theory. New York: ACM, 1992: 144-152.
17 CHEN T, GUESTRIN C. XGBoost: a scalable tree boosting system [C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2016: 785-794.
18 FRIEDMAN J H Greedy function approximation: a gradient boosting machine[J]. Annals of Statistics, 2001, 29 (5): 1189- 1232
19 HE X, PAN J, JIN O, et al. Practical lessons from predicting clicks on ads at facebook [C]// Proceedings of the 8th International Workshop on Data Mining for Online Advertising. New York: ACM, 2014: 1-9.
20 牛晓太 基于KNN 算法和10折交叉验证法的支持向量选取算法[J]. 华中师范大学学报: 自然科学版, 2014, 48 (3): 335- 338
NIU Xiao-tai Support vector extracted algorithm based on KNN and 10 fold cross-validation method[J]. Journal of Huazhong Normal University: Natural Sciences, 2014, 48 (3): 335- 338
[1] 战友,李强,马啸天,王郴平,邱延峻. 基于宏微观纹理特征融合的路面摩擦性能预测[J]. 浙江大学学报(工学版), 2021, 55(4): 684-694.
[2] 陈巧红,陈翊,李文书,贾宇波. 多尺度SE-Xception服装图像分类[J]. 浙江大学学报(工学版), 2020, 54(9): 1727-1735.
[3] 王慧芳,张晨宇. 采用极限梯度提升算法的电力系统电压稳定裕度预测[J]. 浙江大学学报(工学版), 2020, 54(3): 606-613.
[4] 谢乐,衡熙丹,刘洋,蒋启龙,刘东. 基于线性判别分析和分步机器学习的变压器故障诊断[J]. 浙江大学学报(工学版), 2020, 54(11): 2266-2272.
[5] 万志远,陶嘉恒,梁家坤,才振功,苌程,乔林,周巧妮. Stack Overflow上机器学习相关问题的大规模实证研究[J]. 浙江大学学报(工学版), 2019, 53(5): 819-828.
[6] 柯懂湘,潘丽敏,罗森林,张寒青. 基于随机森林算法的Android恶意行为识别与分类方法[J]. 浙江大学学报(工学版), 2019, 53(10): 2013-2023.
[7] 忽丽莎, 王素贞, 陈益强, 高晨龙, 胡春雨, 蒋鑫龙, 陈振宇, 高兴宇. 基于可穿戴设备的跌倒检测算法综述[J]. 浙江大学学报(工学版), 2018, 52(9): 1717-1728.
[8] 于勇, 周阳, 曹鹏, 赵罡. 基于MBD模型的工序模型构建方法[J]. 浙江大学学报(工学版), 2018, 52(6): 1025-1034.
[9] 王洪凯, 陈中华, 周纵苇, 李迎辞, 陆佩欧, 王文志, 刘宛予, 于丽娟. 机器学习算法诊断PET/CT纵膈淋巴结性能评估[J]. 浙江大学学报(工学版), 2018, 52(4): 788-797.
[10] 王希晨,周学军,周媛媛. 基于检测信息融合的缆系海底观测网络节点部署策略[J]. 浙江大学学报(工学版), 2015, 49(9): 1665-1671.
[11] 吴鹏洲,于慧敏,曾雄. 基于正则化风险最小化的目标计数[J]. 浙江大学学报(工学版), 2014, 48(7): 1226-1233.
[12] 宓云軿 王晓萍 金鑫. 基于机器学习的水质COD预测方法[J]. J4, 2008, 42(5): 790-793.
[13] 黄启春 刘仰光 何钦铭. 基于支持向量机的增量式算法[J]. J4, 2008, 42(12): 2121-2126.