智能制造

面向不平衡数据集的汽车零部件质量预测

展开
  • 1.复旦大学软件学院, 上海,200433
    2.复旦大学上海市数据科学重点实验室, 上海,200433
李敏波,男,1970年生,副教授,博士。研究方向为工业大数据分析、物联网智能信息处理。发表论文40余篇。Email:limb@fudan.edu.cn。

收稿日期: 2020-06-12

  网络出版日期: 2022-01-25

基金资助

国家重点研发计划(2018YFB1703104);国家自然科学基金(61671157)

Quality Prediction of Automotive Parts for Imbalanced Datasets

Expand
  • 1.Software School,Fudan University,Shanghai,200433
    2.Shanghai Key Laboratory of Data Science,Fudan University,Shanghai,200433

Received date: 2020-06-12

  Online published: 2022-01-25

摘要

针对汽车零部件质检数据存在合格品与不合格品数量不平衡的问题,提出了基于密度聚类与多工序制造特征的MCDC-MF-SMOTE质检数据过采样方法。先对少数类(不合格)与多数类(合格)样本分别进行密度聚类,再对多工序制造数据和类簇样本分布进行过采样权重计算;根据设定的过采样比和类簇权重,在少数类簇中进行过采样数据生成。使用MCDC-MF-SMOTE过采样方法生成汽车零部件质检的平衡数据集,并采用随机森林排序制造特征的重要性,对分类模型LightGBM、XGBoost、SVM和MNB进行Stacking集成来预测不合格品。与随机抽检相比,该方法对不合格产品的检出率提高了约63%。

本文引用格式

李敏波, 董伟伟 . 面向不平衡数据集的汽车零部件质量预测[J]. 中国机械工程, 2022 , 33(01) : 88 -96 . DOI: 10.3969/j.issn.1004-132X.2022.01.010

Abstract

In response to the problems of imbalance between quantity of qualified and unqualified automotive parts for the product inspection datasets, a MCDC-MF-SMOTE oversampling method of quality inspection data was proposed based on density clustering and multi-process manufacturing features. Firstly, density clustering was carried out for the minority(unqualified) and the majority(qualified) samples respectively, and then oversampling weight was calculated by multi-process manufacturing data and cluster sample distribution. Data were generated in a few clusters according to oversampling ratio and cluster weights. The MCDC-MF-SMOTE method was used to generate a balanced quality dataset of automotive parts. Random Forest was used to rank feature importance to reduce feature dimension. LightGBM, XGBoost, SVM, and MNB models for Stacking were integration to predict unqualified products. After experiments, this method has higher stability and quality prediction performance. Compared with random sampling, the detection rate of unqualified products is increased by approximately 63%.

参考文献

[1]COSTA C F, NASCIMENTO M A. Ida 2016 Industrial Challenge: Using Machine Learning for Predicting Failures[C]∥International Symposium on Intelligent Data Analysis. Cham, 2016: 381-386.
[2]LUO B, WANG H, LIU H, et al. Early Fault Detection of Machine Tools Based on Deep Learning and Dynamic Identification[J]. IEEE Transactions on Industrial Electronics, 2019, 66(1): 509-518.
[3]KROLL B, SCHAFFRANEK D, SCHRIEGEL S, et al. System Modeling Based on Machine Learning for Anomaly Detection and Predictive Maintenance in Industrial Plants[C]∥19th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA). Barcelona, 2014: 1-7.
[4]KAUR H, PANNU H S, MALHI A K. A Systematic Review on Imbalanced Data Challenges in Machine Learning: Applications and Solutions[J]. ACM Computing Surveys (CSUR), 2019, 52(4): 1-36.
[5]KRAWCZYK B. Learning from Imbalanced Data: Open Challenges and Future Directions[J]. Progress in Artificial Intelligence, 2016, 5(4): 221-232.
[6]HAIXIANG G, YIJING L, SHANG J, et al. Learning from Class-imbalanced Data: Review of Methods and Applications[J]. Expert Systems with Applications, 2017, 73: 220-239.
[7]SHAHEE S A, ANANTHAKUMAR U. An Adaptive Oversampling Technique for Imbalanced Datasets[C]∥Industrial Conference on Data Mining ICDM2018. New York, 2018: 1-16.
[8]KAUR P, GOSAIN A. FF-SMOTE: a Metaheuristic Approach to Combat Class Imbalance in Binary Classification[J]. Applied Artificial Intelligence, 2019, 33(5): 420-439.
[9]YAN Y, LIU R, DING Z, et al. A Parameter-free Cleaning Method for SMOTE in Imbalanced Classification[J]. IEEE Access, 2019, 7: 23537-23548.
[10]DOUZAS G, BACAO F. Self-Organizing Map Oversampling (SOMO) for Imbalanced Data Set Learning[J]. Expert Systems with Applications, 2017, 82:40-52.
[11]朱海平,尹惠,邓宇浩,等.过程均值偏移随机的 EWMA 控制图优化设计[J]. 计算机集成制造系统,2014, 20(11): 2820-2825.
ZHU Haiping, YIN Hui, DENG Yuhao,et al. Optimization Design of EWMA Control Chart under Random Process Mean Shift[J]. Computer Integrated Manufacturing Systems, 2014, 20(11):2820-2825.
[12]吕旭泽.基于PSO-LS-SVM 的发动机多工序装配质量预测模型研究[D].合肥:合肥工业大学,2016.
LYU Xuze. Research on the Engine Crankshaft Multistep Assembly Quality Prediction Model Based on LS-SVM of Particle Swarm Optimization[D]. Hefei: Hefei University of Technology,2016. 
[13]赵双凤.基于 MES 的机加车间制造过程工序质量控制方法与系统研究[D]. 重庆:重庆大学, 2016
ZHA Shuangfeng. Research on Method and System of Process Quality Control for Machine Shop Manufacturing Process Based on MES[D]. Chongqing: Chongqing University,2016.
[14]于文靖.汽轮机模锻叶片加工质量预测及控制方法研究[D].哈尔滨:哈尔滨工业大学,2018.
YU Wenjing. Research on the Processing Quality Prediction Method of Turbing Die-forging Blade[D]. Harbin: Harbin Institute of Technology,2018.
[15]冯尔磊. 基于 SMT 大数据的产品质量预测方法研究与实现[D].西安:西安电子科技大学,2018.
Feng Erlei. Research on Analysis Method of SMT Product Quality Prediction Based on Big Data[D]. Xi'an: Xidian University, 2018.
[16]杨岚. 基于大数据的多工序制造过程产品质量控制研究 [D]. 绵阳:西南科技大学,2020.
YANG Lan. Research On Product Quality Control Of Multistage Manufacturing Process Based On Big Data[D]. Mianyang: Southwest University of Science and Technology,2020.
[17]MA L, FAN S. CURE-SMOTE Algorithm and Hybrid Algorithm for Feature Selection and Parameter Optimization Based on Random Forests[J]. BMC Bioinformatics, 2017, 18(1): 169.
文章导航

/

〈 〉