跳到主要內容

臺灣博碩士論文加值系統

(100.28.0.143) 您好!臺灣時間:2024/07/14 23:28
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:徐培文
研究生(外文):HSU, PEI-WEN
論文名稱:應用機器學習於射出成型之品質辨識
論文名稱(外文):Implement Machine Learning for Quality Identification of Injection Molding Process
指導教授:蘇純繒蘇純繒引用關係
指導教授(外文):SU, CHWEN-TZENG
口試委員:呂學毅黃喬次
口試委員(外文):LU, HSUEH-YIHUANG,CHIAO-TZU
口試日期:2022-06-27
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:工業工程與管理系
學門:工程學門
學類:工業工程學類
論文種類:學術論文
論文出版年:2022
畢業學年度:110
語文別:中文
論文頁數:59
中文關鍵詞:射出成型隨機森林LightGBMXGBoost
外文關鍵詞:Injection MoldingRandom ForestLightGBMXGBoost
相關次數:
  • 被引用被引用:0
  • 點閱點閱:247
  • 評分評分:
  • 下載下載:62
  • 收藏至我的研究室書目清單書目收藏:1
導致射出成型成品不良的原因包含操作人員、機台設備、原料種類、參數設定、生產環境等原因,其中機台參數設定為關鍵因素,射出參數設定包含壓力、速度、溫度與時間,過往以現場人員調整參數,常會出現誤差並導致整批成品不良的情況發生,為了有效解決上述之問題,需以非人工的方式去調整參數以維持成品良率,減少射出成型業者成本之浪費。
本研究提出機器學習框架辨識成品良率,首先使用特徵篩選,從具有較高的相互關係的機台參數中選取關鍵變數,以提升模型運算效率,接著建立三種機器學習模型,其中包含隨機森林、LightGBM與XGBoost,最終使用AUC、ACC、Recall、Precision、F-Measure做為模型績效評估指標。研究結果顯示經過特徵篩選後模型績效皆優於未特徵篩選的模型,而LightGBM在AUC、ACC、Precision、F-Measure優於隨機森林與XGBoost,而XGBoost在Recall的表現上是最好的,結果顯示特徵篩選搭配機器學習模型有效的提升模型運算效率與績效。

The reasons for the defective injection molding products include operators, machine equipment, raw material types, parameter settings, production environment, etc., among which machine parameter settings are key factors, and injection parameter settings include pressure, speed, temperature and time. In the past, it was on site. When personnel adjust parameters, errors often occur and result in the failure of the entire batch of finished products. In order to effectively solve the above problems, it is necessary to adjust the parameters in a non-manual way to maintain the yield of the finished product, reduce the cost of injection molding industry waste, and enhance industry competition.
The research proposes a machine learning framework to improve the yield of finished products., mainly using the important parameters in the machine parameters of the machine learning, with the effect of the model prototype, making the production, building Random Forest, LightGBM, XGBoost, performance evaluation indexs are AUC, ACC, Recall , Precision , F-Measure. The research results show that after filtering the features, the models have effectively completed the models better than have not completed feature selection, LightGBM outperforms Random Forest and XGBoost in AUC, ACC, Precision, and F-Measure, XGBoost has the best performance in Recall. The results show that feature selection combine machine learning can effectively improve model computing efficiency and accuracy.

摘要 i
Abstract ii
目錄 iii
表目錄 v
圖目錄 vi
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 2
1.3 研究架構 3
第二章 文獻探討 5
2.1 射出成型定義與特性 5
2.2 影響射出成型良率因素 6
2.3 特徵篩選 6
2.4 射出成型良率預測方法 9
2.4.1 機器學習 10
2.4.2 隨機森林 12
2.4.3 輕量化梯度提升演算法 13
2.4.4 極限梯度提升演算法 14
2.5 小結 17
第三章 研究方法 19
3.1 研究架構 19
3.2 資料介紹 20
3.3 資料前處理 21
3.3.1 資料清洗 21
3.3.2 特徵篩選 24
3.3.3 資料正則化 25
3.4 模型建立 26
3.4.1 隨機森林 26
3.4.2 輕量化梯度提升演算法 27
3.4.3 極限梯度提升演算法 29
3.5 績效評估 30
3.5.1 混淆矩陣 30
3.5.2 F-Measure 32
3.5.3 ROC曲線 32
第四章 結果探討 34
4.1 特徵篩選結果 34
4.1.1 隨機森林特徵篩選結果 34
4.1.2 輕量化梯度提升演算法特徵篩選結果 36
4.1.3 極限梯度提升演算法特徵篩選結果 37
4.1.4 特徵篩選結果比較與小節 39
4.2 模型建立 39
4.2.1 隨機森林 39
4.2.2 輕量化梯度提升演算法 40
4.2.3 極限梯度提升演算法 41
4.3 模型比較與小結 42
第五章 結論與建議 45
參考文獻 46

[1]Almeida, A., Orduña, P., Castillejo, E., López-de-Ipiña, D., & Sacristán, M. (2013). A method for automatic generation of fuzzy membership functions for mobile device’s characteristics based on Google Trends. Computers in Human Behavior, 29(2), 510-517.
[2]Arinaldi, A., Pradana, J. A., & Gurusinga, A. A. (2018). Detection and classification of vehicles for traffic video analytics. Procedia computer science, 144, 259-268.
[3]Biau, G., & Scornet, E. (2016). A random forest guided tour. Test, 25(2), 197-227.
[4]Borges, F., Pinto, A., Ribeiro, D., Barbosa, T., Pereira, D., Magalháes, R., ... & Ferreira, D. (2020). An unsupervised method based on support vector machines and higher-order statistics for mechanical faults detection. IEEE Latin America Transactions, 18(06), 1093-1101.
[5]Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
[6]Buratti, Y., Dick, J., Le Gia, Q., & Hameiri, Z. (2019, June). A machine learning approach to defect parameters extraction: Using random forests to inverse the shockley-read-hall equation. In 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) (pp. 3070-3073). IEEE.
[7]Chen, Y., Chen, F., Peng, L., He, C., & Li, C. (2020, September). Research on Rolling Force Prediction Method of High Precision Cold Rolling Based on XGBoost Algorithm. In 2020 7th International Forum on Electrical Engineering and Automation (IFEEA) (pp. 966-971). IEEE.
[8]Cheng, G., & Tong, X. (2018, July). Fuzzy clustering multiple kernel support vector machine. In 2018 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR) (pp. 7-12). IEEE.
[9]Dang, H., Ma, R., Zhao, D., Xie, R., Li, H., & Liu, Y. (2020, October). A Novel Diagnosis Method of Proton Exchange Membrane Fuel Cells Based on the PCA and XGBoost Algorithm. In IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society (pp. 3951-3956). IEEE.
[10]Dehuri, S., & Ghosh, A. (2013). Revisiting evolutionary algorithms in feature selection and nonfuzzy/fuzzy rule based classification. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 3(2), 83-108.
[11]Dembczynski, K., Waegeman, W., Cheng, W., & Hüllermeier, E. (2011). An exact algorithm for F-measure maximization. Advances in neural information processing systems, 24.
[12]Demidova, L., & Ivkina, M. (2020, June). Approach to Determining the Boundaries of the Search Range for the Number of Trees in the Random Forest Algorithm. In 2020 9th Mediterranean Conference on Embedded Computing (MECO) (pp. 1-4). IEEE.
[13]Deng, H., Yan, F., Wang, H., Fang, L., Zhou, Z., Zhang, F., ... & Jiang, H. (2021, December). Electricity Price Prediction Based on LSTM and LightGBM. In 2021 IEEE 4th International Conference on Electronics and Communication Engineering (ICECE) (pp. 286-290). IEEE.
[14]Elaidi, H., Elhaddar, Y., Benabbou, Z., & Abbar, H. (2018, April). An idea of a clustering algorithm using support vector machines based on binary decision tree. In 2018 International Conference on Intelligent Systems and Computer Vision (ISCV) (pp. 1-5). IEEE.
[15]Gajera, V., Gupta, R., & Jana, P. K. (2016, July). An effective multi-objective task scheduling algorithm using min-max normalization in cloud computing. In 2016 2nd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT) (pp. 812-816). IEEE.
[16]Gao, H., Zhang, Y., Zhou, X., & Li, D. (2018). Intelligent methods for the process parameter determination of plastic injection molding. Frontiers of Mechanical Engineering, 13(1), 85-95.
[17]Ge, D., Gu, J., Chang, S., & Cai, J. (2020, April). Credit card fraud detection using lightgbm model. In 2020 international conference on E-commerce and internet technology (ECIT) (pp. 232-236). IEEE.
[18]Guo, W., Deng, F., Meng, Z., Hua, L., Mao, H., & Su, J. (2020). A hybrid back-propagation neural network and intelligent algorithm combined algorithm for optimizing microcellular foaming injection molding process parameters. Journal of Manufacturing Processes, 50, 528-538.
[19]Hancock, J., & Khoshgoftaar, T. M. (2021, August). Leveraging lightgbm for categorical big data. In 2021 IEEE Seventh International Conference on Big Data Computing Service and Applications (BigDataService) (pp. 149-154). IEEE.
[20]Hanley, J. A. (1989). Receiver operating characteristic (ROC) methodology: the state of the art. Crit Rev Diagn Imaging, 29(3), 307-335.
[21]Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York: springer.
[22]Hernandez, P. A. (2012, June). Neural-fuzzy approach to optimize process parameters for injection molding machine. In 2012 Eighth International Conference on Intelligent Environments (pp. 186-189). IEEE.
[23]Jiang, J., Liu, X., Lian, M., Pan, Y., Chen, Q., Liu, H., Zheng, G., Guo, Z., Schubert, D. W., & Shen, C. (2018). Self-reinforcing and toughening isotactic polypropylene via melt sequential injection molding. Polymer Testing, 67, 183-189.
[24]Kim, S., Kim, S., & Ryu, K. R. (2018, October). Deep learning experiments with skewed data for defect prediction in plastic injection molding. In 2018 IEEE/ACS 15th International Conference on Computer Systems and Applications (AICCSA) (pp. 1-2). IEEE.
[25]Kuhn, M., & Johnson, K. (2019). Feature engineering and selection: A practical approach for predictive models. CRC Press.
[26]Lan, H., & Pan, Y. (2019, June). A crowdsourcing quality prediction model based on random forests. In 2019 IEEE/ACIS 18th International Conference on Computer and Information Science (ICIS) (pp. 315-319). IEEE.
[27]Lee, L. C., & Jemain, A. A. (2021). On overview of PCA application strategy in processing high dimensionality forensic data. Microchemical Journal, 169, 106608.
[28]Leevy, J. L., Hancock, J., Zuech, R., & Khoshgoftaar, T. M. (2020, October). Detecting cybersecurity attacks using different network features with lightgbm and xgboost learners. In 2020 IEEE Second International Conference on Cognitive Machine Intelligence (CogMI) (pp. 190-197). IEEE.
[29]Lhamon Jr, W. (1996). American Plastic: A Cultural History. The Journal of American History, 83(3), 1028.
[30]Li, J.-X., Li, Y.-H., Qin, Z.-B., & Dong, G.-Y. (2018). Ultrasound assisted synthesis of a zinc (II) coordination polymer with nano-flower morphology and the use as precursor for zinc (II) oxide nanoparticles. Polyhedron, 155, 94-101.
[31]Lu, S., Li, Q., Yu, H., & Wang, X. (2020, November). Damage Evaluation Method of CFRP Structures Based on PCA and Random Forest Algorithm. In 2020 Chinese Automation Congress (CAC) (pp. 3804-3807). IEEE.
[32]Mei, Y., & Wu, D. L. (2009, July). Research on the quantification of defects in injection molded parts based on digital image processing. In 2009 International Conference on Machine Learning and Cybernetics (Vol. 2, pp. 1210-1213). IEEE.
[33]Mohan, L., Pant, J., Suyal, P., & Kumar, A. (2020, September). Support vector machine accuracy improvement with classification. In 2020 12th International Conference on Computational Intelligence and Communication Networks (CICN) (pp. 477-481). IEEE.
[34]Morán-Fernández, L., Bolón-Canedo, V., & Alonso-Betanzos, A. (2018). Feature selection with limited bit depth mutual information for embedded systems. Multidisciplinary Digital Publishing Institute Proceedings, 2(18), 1187.
[35]Mun, J., & Jeong, J. (2021, May). Design and Analysis of Optimal Recipe Prediction Model Based on Deep Learning for Advanced Composite Material Injection Molding. In 2021 International Conference on Computer Communication and Artificial Intelligence (CCAI) (pp. 176-179). IEEE.
[36]Oktem, H., Erzurumlu, T., & Uzman, I. (2007). Application of Taguchi optimization technique in determining plastic injection molding process parameters for a thin-shell part. Materials & design, 28(4), 1271-1278.
[37]Park, H. S., Phuong, D. X., & Kumar, S. (2019). AI based injection molding process for consistent product quality. Procedia Manufacturing, 28, 102-106.
[38]Ribeiro, B. (1999, July). Fault detection in a thermoplastic injection molding process using neural networks. In IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No. 99CH36339) (Vol. 5, pp. 3352-3355). IEEE.
[39]Rodríguez-Trejo, J. J., & Femat-Diaz, A. (2018, May). Identification of significant variables using random forest, in a process of injection moulding: A case study of yield reduction analysis in changing plastic injection moulds for auto parts products. In 2018 XIV International Engineering Congress (CONIIN) (pp. 1-6). IEEE.
[40]Sharma, S. K., Sharma, N. K., & Potter, P. P. (2020, December). Fusion approach for document classification using random forest and SVM. In 2020 9th International Conference System Modeling and Advancement in Research Trends (SMART) (pp. 231-234). IEEE.
[41]Shi, J., Niu, N., & Zhu, X. (2019, May). A fault diagnosis method for multi-condition system based on random forest. In 2019 Prognostics and System Health Management Conference (PHM-Paris) (pp. 350-355). IEEE.
[42]Simongyi, M., & Chongstitvatana, P. (2018, July). Abnormality Detection in Hard Disk Drive Assembly Process Using Support Vector Machine. In 2018 15th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON) (pp. 612-615). IEEE.
[43]Sindhu, S., Patil, S. P., Sreevalsan, A., Rahman, F., & AN, M. S. (2020, October). Phishing detection using random forest, SVM and neural network with backpropagation. In 2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE) (pp. 391-394). IEEE.
[44]Song, K., Yan, F., Ding, T., Gao, L., & Lu, S. (2020). A steel property optimization model based on the XGBoost algorithm and improved PSO. Computational Materials Science, 174, 109472.
[45]Spruyt, V. (2014). The Curse of Dimensionality in classification. Computer vision for dummies, 21(3), 35-40.
[46]Sun, L. (2020, December). Application and improvement of Xgboost algorithm based on multiple parameter optimization strategy. In 2020 5th International Conference on Mechanical, Control and Computer Engineering (ICMCCE) (pp. 1822-1825). IEEE.
[47]Tesoriero, A. J., Gronberg, J. A., Juckem, P. F., Miller, M. P., & Austin, B. P. (2017). Predicting redox‐sensitive contaminant concentrations in groundwater using random forest classification. Water Resources Research, 53(8), 7316-7331.
[48]Tian, M. J., Cui, R. Y., & Huang, Z. H. (2018, July). Automatic Extraction Method for Specific Domain Terms Based on Structural Features and Mutual Information. In 2018 5th International Conference on Information Science and Control Engineering (ICISCE) (pp. 147-150). IEEE.
[49]Tsou, H. H., Lee, K. D., Wang, Z. H., & Huang, C. C. (2020, December). The Feature Extraction Modeling of Product Analysis in Injection Molding for Intelligent Manufacturing. In 2020 International Computer Symposium (ICS) (pp. 325-329). IEEE.
[50]Van Rijsbergen, C. (1979, September). Information retrieval: theory and practice. In Proceedings of the Joint IBM/University of Newcastle upon Tyne Seminar on Data Base Systems (Vol. 79).
[51]Wang, G., Zhao, G., Zhang, L., Mu, Y., & Park, C. B. (2018). Lightweight and tough nanocellular PP/PTFE nanocomposite foams with defect-free surfaces obtained using in situ nanofibrillation and nanocellular injection molding. Chemical Engineering Journal, 350, 1-11.
[52]Wang, L. (2019, March). Research on distributed parallel dimensionality reduction algorithm based on pca algorithm. In 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) (pp. 1363-1367). IEEE.
[53]Wang, M. W., Jeng, J. H., & Fu, G. L. (2008, October). Optimal molding parameter design of PLA micro lancet needles using Taguchi method. In 2008 IEEE International Conference on Service Operations and Logistics, and Informatics (Vol. 2, pp. 2731-2735). IEEE.
[54]Xie, J., Li, Z., Zhou, Z., & Liu, S. (2020). A Novel Bearing Fault Classification Method Based on XGBoost: The Fusion of Deep Learning-Based Features and Empirical Features. IEEE Transactions on Instrumentation and Measurement, 70, 1-9.
[55]Yang, D., Danai, K., & Kazmer, D. (2001). A knowledge-based tuning method for injection molding machines. J. Manuf. Sci. Eng., 123(4), 682-691.
[56]Yang, L., Li, Y., & Di, C. (2019, July). Application of XGBoost in identification of power quality disturbance source of steady-state disturbance events. In 2019 IEEE 9th International Conference on Electronics Information and Emergency Communication (ICEIEC) (pp. 1-6). IEEE.
[57]Yin, J., Medellín-Azuara, J., Escriva-Bou, A., & Liu, Z. (2021). Bayesian machine learning ensemble approach to quantify model uncertainty in predicting groundwater storage change. Science of The Total Environment, 769, 144715.
[58]Yuan, H., Zhu, C., Cao, X., Wang, X., & Wang, H. (2012, May). A selection method of Acoustic Emission characteristic parameters based on mutual information and distance measurement. In 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery (pp. 1377-1381). IEEE.
[59]Zhang, S., & Liu, Y. (2021, December). Fault diagnosis for the planetary gearbox based on an improved LightGBM. In 2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes (SAFEPROCESS) (pp. 1-6). IEEE.
[60]Zhang, X. (2020, June). Ion channel prediction using Lightgbm Model. In 2020 International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) (pp. 349-352). IEEE.
[61]Zhou, N., Shao, Q., Zhou, J., & Changjie, H. (2021, March). Fault classification of proton exchange membrane fuel cells for vehicles based on XGBoost. In 2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) (pp. 1054-1058). IEEE.
[62]Zhou, Q., Zhou, H., & Li, T. (2016). Cost-sensitive feature selection using random forest: Selecting low-cost subsets of informative features. Knowledge-based systems, 95, 1-11.
[63]Zhou, X., Zhang, Y., Mao, T., & Zhou, H. (2017). Monitoring and dynamic control of quality stability for injection molding process. Journal of Materials Processing Technology, 249, 358-366.


QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊