(44.192.66.171) 您好!臺灣時間:2021/05/18 01:29
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:鄧翔運
研究生(外文):Xiang-Yun Deng
論文名稱:基於機器學習之織布定型機與冰水主機節能分析與肇因診斷
指導教授:王文俊王文俊引用關係
指導教授(外文):Wen-June Wang
學位類別:碩士
校院名稱:國立中央大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:82
中文關鍵詞:織布定型機冰水主機機器學習演算法節能資料探勘
外文關鍵詞:Stenter machineChillerMachine learning algorithmEnergy savingData mining
相關次數:
  • 被引用被引用:0
  • 點閱點閱:135
  • 評分評分:
  • 下載下載:39
  • 收藏至我的研究室書目清單書目收藏:0
本論文主要針對紡織產業和空調系統中耗電量較大的織布定型機與冰水主機做資料分析與建立機器學習模型,以達到節能之目標。在紡織工廠中,熟悉機台參數的老師傅凋零,經驗無法累積,烘乾參數無法反應少量多樣接單模式。因此,本研究目標是建立織品烘乾過程的精確輸入/輸出關係模型,以常見機器學習演算法如決策樹、隨機森林、AdaBoost、KNN、XGBoost、神經網路、Voting regressor和Stacking regressor建立模型。通過Grid Search的方式尋找各種機器學習模型的最佳參數,並以整合學習方法來降低變異量、偏差,建立準確率更高的模型。也建立倒傳遞神經網路(BPNN),加入Embeddings,透過One-cycle policy的學習方法,訓練神經網路達到滿意的含水率預測。最後,比較不同機器學習模型,以最佳的模型XGBoost來預測含水率,並計算SHAP(Shapley Additive exPlanations)值解釋模型輸入和輸出的正向或負向關聯。當給予一組含水率較大的測試資料,以前述最佳模型模擬定型機的運轉,藉此來調整某些輸入特徵,以求達到滿意的織物含水率,因而達到節能的目的。
在空調系統中,冰水主機約占總耗電量之60%,以目前200噸的主機而言,使用三年後的冰機效率約只達到新機的70%,因此,目標在以K-means將不同能耗的冰機分群,分析同類型(同噸數)冰機在不同聚類下造成低效肇因的輸入特徵(參數)。本文透過建立XGBoost模型,計算SHAP值,選擇和輸出關聯性最大的四個特徵,以平均輪廓分數(Silhouette score)最高來決定耗能冰機的聚類,比較和分析聚類結果和實際冰機資料,視覺化呈現造成冰機低效的肇因。最後,建立圖形使用者介面的冰機診斷系統,進行線上(On-line)診斷,以獨立樣本t檢定來檢驗當前冰機效率是否比模型估測效率還差,當檢定為真說明冰機效率開始變差,透過改善因子分析,告訴使用者可能的輸入肇因,藉此來調整設備,讓效率提升。
Based on the machine learning algorithm and data analysis technique, this thesis studies the energy saving for the cloth setting machine in the textile industry and chiller in the air conditioning system. In the textile factories, the master who is familiar with the machine parameters is reducing year by year, hence, the useful and helpful experience could not pass on. On the other hand, the same parameters cannot reflect small-volume, large-variety production. The first purpose of this study is to build an accurate input-output model in the fabric drying process, therefore, a series of familiar machine learning algorithms, such as decision tree, AdaBoost, KNN (k-nearest neighbors), XGBoost, neural network, voting regressor, and stacking regressor are applied. The grid search is used to find the best parameters of machine learning models, and the integrated learning methods are used to reduce variation, bias and deviation such that the model with higher accuracy can be established. A back-propagation network with embeddings is trained with one-cycle policy to achieve satisfactory moisture content prediction. After comparing the performance of machine learning models, the best trained model is XGBoost. The model will predict the moisture of the fabric, and the SHAP (Shapley Additive exPlanations) value will be calculated to explain the positive or negative relationship between the input and output of model. When a set of test data with a larger moisture content is given, we use the aforementioned best model to simulate the operation of the stenter machine so that we can know how to adjust input characteristics to achieve a satisfactory fabric moisture content. Thus the purpose of energy saving is achieved.
In the air condition system, the chiller accounts for about 60% of total power consumption. Actually, for the 200-ton chiller, the performance is only about 70% in comparison with the new one after it runs for three years. Therefore, the second purpose of this study is to cluster different efficient chillers and look for the input characteristics (parameters) of inefficient causes. An XGBoost model is chosen and SHAP value is calculated to select the four features which have the highest correlation with the output. The highest average silhouette score is used to determine the clustering of energy-consuming chillers. The clustering results and actual chiller data are compared and analyzed to visualize the causes of the inefficiency of the chiller. Finally, a chiller diagnostic system with a graphical user interface is established to perform on-line diagnosis. An independent sample t test is used to test whether the current chiller efficiency is worse than the model estimated efficiency. When the test is true, it indicates the chiller efficiency starts to deteriorate. Then through improved factor analysis, we can find the possible input characteristics (parameters) of inefficient causes.
摘要 i
Abstract ii
致謝 iv
目錄 v
圖目錄 vii
表目錄 x
第一章 緒論 1
1.1 研究背景與動機 1
1.2 文獻回顧 1
1.3 論文目標 3
1.4 論文架構 3
第二章 研究方法與演算法 4
2.1 決策樹 (Decision tree)[14] 4
2.2 整合學習 (Ensemble learning) 6
2.2.1 Bagging演算法[35] 7
2.2.2 隨機森林[15] 8
2.2.3 AdaBoost[16] 9
2.2.4 XGBoost[17] 10
2.2.5 Voting regressor 15
2.2.6 Stacking regressor[19] 15
2.3 KNN (k-nearest neighbor) 17
2.4 倒傳遞神經網路 17
2.5 PCA演算法 19
2.6 K-means 20
2.7 模型解釋性[25] 21
2.8 軟體工具 23
第三章 紡織工廠-織布定型機品質建模 25
3.1 資料預處理 25
3.2 機器學習參數組合 28
3.3 神經網路訓練 32
3.4 結合不同機器學習模型 35
3.5 視覺化資料 36
3.6 織物含水率估測 38
第四章 冰水主機-能耗異常診斷模組 41
4.1 資料預處理 43
4.2 建立機器學習模型 44
4.3聚類結果分析 45
4.4 冰機診斷系統 55
4.4.1 獨立樣本t檢定 56
4.4.2線上(On-line)診斷 58
第五章 結論與未來展望 64
5.1 結論 64
5.2 未來展望 65
參考文獻 66
[1] 台灣電力公司, 台電系統歷年(39-108)發購電量分七類, 2020, Available: https://reurl.cc/ar52bD.
[2] G. Galoppi, L. Ferrari, G. Rerrara, and E. A. Carnevale, “Experimental investigation on industrial drying process of cotton yarn bobbins: energy consumption and drying time,” Energy Procedia, vol. 126, pp. 361-368, 2017.
[3] U. Akyol, A. C. Han, and K. Kahvec, “Experimental and theoretical investigation of drying kinetics of cotton bobbins,” 7th International Conference - TEXSCI , 2010, pp. 6-8.
[4] 黃政勳, 黃慶堂, 林于棟, 王立人, "織物烘乾製程能源管理系統建構, "Journal of the Hwa Gang Textile, vol. 23, no. 5, pp. 327-331, 2016.
[5] P. Yildirim, D. Birant, and T. Alpyildiz. "Data mining and machine learning in textile industry," Data Mining and Knowledge Discovery, vol. 8, no.1, 2018.
[6] J. Boubaker and F. Khedher, "Evaluation of sewed thread consumption of jean trousers using neural network and regression methods," Fibres & Textiles in Eastern Europe, vol. 23, no. 3, pp. 91-96, 2015.
[7] I. G. Mariolis and E. S. Dermatas, "Automated assessment of textile seam quality based on surface roughness estimation," The Journal of The Textile Institute, vol. 101, no. 7, pp. 653-659, 2010.
[8] J. H. Kim, N. C. Seong, and W. Choi, “Modeling and optimizing a chiller system using a machine learning algorithm,” Energies, vol. 12, no. 15, pp. 2860, 2019.
[9] 黃仲翊, “外部輸入非線性自動迴歸模型應用於冰水主機耗能分析,” 碩士論文, 冷凍空調工程系所, 國立台北科技大學, 台北市, 2017.
[10] F. W. Yu, W. T. Ho, K. T. Chan, and R. K. Y. Sit, “Critique of operating variables importance on chiller energy performance using random forest,” Energy and Buildings, vol. 139, pp. 653-663, 2017.
[11] O. Renaud and M. P. Victoria-Feser, “A robust coefficient of determination for regression,” Journal of Statistical Planning and Inference, vol. 140, no. 7, pp. 1852-1862, 2010.
[12] C. F. Chien et al., "AI and big data analytics for wafer fab energy saving and chiller optimization to empower intelligent manufacturing," 2018 e-Manufacturing & Design Collaboration Symposium (eMDC), Hsinchu, 2018, pp. 1-4.
[13] X. Yu, S. Ergan, and G. Dedemen, “A data-driven approach to extract operational signatures of HVAC systems and analyze impact on electricity consumption,” Applied Energy, vol. 253, pp. 113497, 2019.
[14] W. Y. Loh, “Classification and regression trees,” Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol. 1, no. 1, pp. 14-23, 2011.
[15] L. Breiman, “Random forests,” Machine learning, vol. 45, no. 1, pp. 5-32, 2001.
[16] Y. Freund, R. Schapire, and N. Abe, “A short introduction to boosting,” Journal-Japanese Society For Artificial Intelligence, vol. 14, no. 5, pp. 771-780, 1999.
[17] T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785-794.
[18] J. H. Friedman, “Greedy function approximation: a gradient boosting machine,” Annals of statistics, vol. 29, no. 5, pp. 1189-1232, 2001.
[19] L. Breiman, “Stacked Regressions,” Machine learning, vol. 24, no. 1, pp. 49-64, 1996.
[20] A. de Brébisson, É. Simon, A. Auvolat, P. Vincent, and Y. Bengio, “Artificial Neural Networks Applied to Taxi Destination Prediction,” arXiv preprint arXiv:1508.00021, 2015.
[21] C. Guo and F. Berkhahn, “Entity embeddings of categorical variables”, arXiv preprint arXiv:1604.06737, 2016.
[22] L. N. Smith, “No more pesky learning rate guessing games,” Computer Vision and Pattern Recognition, abs/1506.01186, 2017.
[23] L. N. Smith, “A disciplined approach to neural network hyper-parameters: Part 1--learning rate, batch size, momentum, and weight decay,” arXiv preprint arXiv:1803.09820, 2018.
[24] P. J. Rousseeuw, “Silhouettes: a graphical aid to the interpretation and validation of cluster analysis,” Journal of computational and applied mathematics, vol. 20, pp. 53-65, 1987.
[25] S. M. Lundberg and S. I. Lee, "A unified approach to interpreting model predictions," Advances in neural information processing systems, pp. 4765-4774, 2017.
[26] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research , vol. 12, pp. 2825-2830, 2011.
[27] S. Raschka, “MLxtend: Providing machine learning and data science utilities and extensions to Python's scientific computing stack,” Journal of Open Source Software, vol. 3, no. 24, 2018.
[28] J. Howard and S. Gugger, “fastai: A Layered API for Deep Learning,” Information, vol. 11, no. 2, pp. 108, 2020.
[29] 定型機廢氣餘熱回收及其資源化利用, 2018,
Available: https://kknews.cc/zh-tw/home/o982y95.html.
[30] 陳輝俊, ”空調節能技術與能源管理”, 台電空調技術運用研討會, 2016.
[31] T. Hartman, “All-Variable Speed Centrifugal Chiller Plants,” ASHRAE journal, vol. 43, no. 9, pp. 45-53, 2001.
[32] M. J. Azur, E. A. Stuart, C. Frangakis, and P. J. Leaf, “Multiple imputation by chained equations: What is it and how does it work?,” International Journal of Methods in Psychiatric Research, vol. 20, no. 1, pp. 40-49, 2011.
[33] 黃老師的數位教材, 基礎統計, 第九章假設檢定, 2019, Available: http://demo1.nkuht.edu.tw/~tient/bs/chap09.pdf.
[34] 工研院, 品質深度學習模組學界分包, 人工智慧製成節能分析技術-定型機品質預測模型, 2019.
[35] L. Breiman, “Bagging predictors,” Machine learning, vol. 24, no. 2, pp. 123-140, 1996.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top