Pattern recognition of control charts based on data feature enhancement and ensemble learning of classifiers for dimensional accuracy of products
Hongyan Chu Yuefeng Dong Qiang Cheng Jun Yan Yongsheng Zhao Jianqiang Cao Caixia Zhang Xuehan Chen a Institute of Advanced Manufacturing and Intelligent Technology,Beijing University of Technology,Beijing,People's Republic of Chinab Beijing Key Laboratory of Advanced Manufacturing Technology,Beijing University of Technology,Beijing,People's Republic of Chinac Jiangxi Tianyue Auto Electric Co,Jiangxi,People's Republic of ChinaHongyan Chu received her Ph.D. degree in mechanical design and theory from Beijing University of Technology,Beijing,China,in 2003. She is currently a professor of the Beijing Key Laboratory of Advanced Manufacturing Technology,Beijing University of Technology. She has published more than 40 SCI/EI papers. Her research interests include intelligent manufacturing technology,optimisation algorithm,reliability and accuracy retention of manufacturing equipment.Yuefeng Dong obtained his Master's degree in Mechanical Engineering from Beijing University of Technology in 2024. He studied at the Beijing Key Laboratory of Advanced Manufacturing Technology,which is affiliated with Beijing University of Technology. His research interests include intelligent manufacturing technology,quality control,and predictive techniques.Qiang Cheng received his Ph.D. degree from Huazhong University of Science and Technology in 2009. He worked as a professor of College of Mechanical and Energy Engineering at Beijing University of Technology from 2018 to present. He is the deputy director of Beijing Key Laboratory of Advanced Manufacturing Technology and the director of Key Laboratory of Digital Design and Testing Technology for Heavy Duty Machine Tools in Mechanical Industry. He has published more than 100 SCI papers and 30 EI papers,has been authorised more than 40 national invention patents and software copyrights,and received 7 provincial and ministerial level science and technology awards.Jun Yan received his Ph.D. degree in mechanical engineering from Beijing University of Technology,Beijing,China,in 2022. He is currently a lecturer of the Institute of Advanced Manufacturing and Intelligent Technology,Beijing University of Technology. His research interests include optimisation algorithm,fault diagnosis and their applications in machining operation. He has authored more than 10 articles on journals including International Journal of Production Research,Journal of Cleaner Production,Robotics and Computer-integrated Manufacturing and so on.Yongsheng Zhao received his Ph.D. degree in mechanical engineering from Huazhong University of Science and Technology,Wuhan,China,in 2007. He is currently a Professor of the Institute of Advanced Manufacturing and Intelligent Technology,Beijing University of Technology. His research interests include machine tool nonlinear dynamics,low stress assembly for high-end CNC machine tools. He has authored more than 50 articles on journals including Tribology International,Nonlinear Dynamics,Structural Engineering and Mechanics and so on.Jianqiang Cao received his B.S. degree in mechanical and electronic engineering from the Hebei University of Science and Technology,Shijiazhuang,China,in 2022. He is currently studying at the Beijing Key Laboratory of Advanced Manufacturing Technology at Beijing University of Technology. His research interests include intelligent manufacturing technology and green manufacturing.Caixia Zhang received her Ph.D. degree in mechanical engineering from Tsinghua University,Beijing,China,in 2015. She is currently an associate professor of the Institute of Advanced Manufacturing and Intelligent Technology,Beijing University of Technology. Her research interests include intelligent manufacturing technology,production scheduling and interface analysis. She has authored more than 20 articles on journals including IEEE Access,Applied Mathematical Modelling,Journal of Manufacturing Systems,Mechanical and Machine Theory and so on.Xuehan Chen received a diploma in Automotive Service Technology from Vancouver Community College in Canada 2017. He is currently the manager of the Jiangxi Tianyue Automobile Electric Appliance Co.,Ltd. He is overseeing both the development and sales in the company.
DOI: https://doi.org/10.1080/00207543.2024.2387095
IF: 9.018
2024-08-23
International Journal of Production Research
Abstract:Product quality in manufacturing is directly related to the reliability, durability and safety of the product. Product quality can be controlled by building a product control chart pattern recognition model. The complexity and temporal sequence of product quality data make it difficult to extract effective control chart features, and it is difficult for common classifiers to play an effective role for each control chart pattern category, thus affecting the accuracy of pattern recognition. Therefore, this paper proposes a data feature enhancement processing method that combines Smote algorithm and smooth shift processing. This method reveals data features by performing feature enhancement processing on raw data and extracts statistical and shape features for dimensionality reduction to improve recognition accuracy and efficiency. Based on this approach, this paper also proposes an ensemble learning model based on the Stacking method, which incorporates BP, ELM, PNN and SVM classifiers for control chart pattern recognition. Through extensive experimental validation, the data feature enhancement method improves the accuracy of the weak classifier normal pattern recognition by 30%, and effectively reduces the misclassification in the quality control process. The ensemble learning model is stronger than the weak classifier, and its recognition accuracy can reach 97.31%.
engineering, manufacturing, industrial,operations research & management science