博碩士論文 108522010 詳細資訊




以作者查詢圖書館館藏 以作者查詢臺灣博碩士 以作者查詢全國書目 勘誤回報 、線上人數:8 、訪客IP:44.192.26.60
姓名 林聖洋(Lin Sheng-Yang)  查詢紙本館藏   畢業系所 資訊工程學系
論文名稱 基於工廠生產資料的異常機器維修預測
(Predictive maintenance of abnormal machines based on the factory production data)
相關論文
★ 基於最大期望算法之分析陶瓷基板機器暗裂破片率★ 基於時間序列預測的機器良率預測
★ 基於OpenPose特徵的行人分心偵測★ 建構深度學習CNN模型以正確分類傳統AOI模型之偵測結果
★ 一種結合循序向後選擇法與回歸樹分析的瑕疵肇因關鍵因子擷取方法與系統-以紡織製程為例★ 融合生成對抗網路及領域知識的分層式影像擴增
★ 針織布異常偵測方法研究★ 應用方位感測器之手機使用者識別機制
★ 非侵入式多模組之手機使用者識別機制 :基於動態方法★ 多分類器組合應用於財務危機預測
★ 漸進式模型應用於財務危機預測問題★ Bus Arrival Prediction - to Ensure Users Not to Miss the Bus (Preliminary Study based on Bus Line 243 Taipei)
★ 公車路線規劃系統之資料自動收集系統實作★ 特徵挑選方法和分類器在財務危機預測問題中比較
★ OR ensemble 應用於財務危機預測★ 智慧型手機使用者操作姿勢對於非侵入式識別機制的影響分析:基於動態方法
檔案 [Endnote RIS 格式]    [Bibtex 格式]    [相關文章]   [文章引用]   [完整記錄]   [館藏目錄]   至系統瀏覽論文 ( 永不開放)
摘要(中) 在工廠生產模式中,異常的機器容易在生產時造成產品損壞,導致不良品數量上升,
嚴重者更可能停機導致整條生產線呆滯。普遍工廠都會安排定期維修保養,希望減少此情
況發生。在智慧製造 4.0 中,提出了對機器加裝感應器或是紀錄器協助生產,即便感應器
或紀錄器在生產中發出聲響提醒,此刻仍需停機維修,延誤了生產線。進而衍發出利用感
應器上的詳細資料進行剩餘維修剩餘天數預測。若機器能在異常前提早維修,便可減少不
良品以及省下不必要的停機時間,降低生產成本。然而並非每家工廠都願意額外花費安裝
感應器抑或者並非每部機器都可以嵌入安裝感應器來進行維修預測。
本研究針對上述問題,提出了一個利用工廠每週生產後提供的生產資料,建立了下
個週期的時間序列維修機器推薦列表模型。並使用了 Online Learning 的概念,隨著時間推
移更新每週的生產資料,預測下個週期的維修機器。利用實際生產維修紀錄當作比較結果,
維修預測列表中的機器在未來週期內皆會發生異常維修,可以利用來安排機器排程,避免
即將異常之機器進入產線。而考量工廠實際生產情況,使用六天(一周)的生產維修紀錄並
當為比較結果,十天的維修建議列表的平均 Recall 為 57.1%,再考量工廠每日維修能量有
限的情況下,將維修機器推薦列表進行排序,並給出前五名的名單十天的平均準確度為
58%,即建議清單中的五部機器有三部在未來六天內會發生維修。
摘要(英) In the factory production mode, the abnormal machines are easier damage the pieces during
production, leading to increase the bad pieces’ number. What is more serious is the abnormal machines
could stop that production line. All factories will arrange the regular maintenance, hoping to reduce the
situation happened. In Smart Manufacturing 4.0, proposed a method to add sensors or recorders on the
machines to assist in production. Even if the sensors or recorders makes a sound to reminder during
production. It still needs to stop the production line to repair the machine, which delays the production
line. After the concept of sensors, developing that using the detailed data on the sensors to predict the
machines remaining useful life (RUL). If the machines can be repaired before abnormal conditions, it can
reduce the bad pieces’ number and save the unnecessary stop production line time. The production cost
can be reduced. However, not every factory is willing to spend extra money to install sensors or not
every machine can embed with sensors for predictive maintenance.
In response to the above problems, this research proposes a method. Using the production data
provided by the factory every week to build a time series model of the recommended machine list. It
also uses the concept of Online Learning to update weekly production data over time and predict the
machines need to maintain in next period. Using the actual production maintenance record as the
comparison result, all machines in recommended machine list will be abnormally repaired in the future
period, the result can be used to arrange the machine schedule to prevent the abnormal machine will
influence the production line. Considering the actual production situation of the factory, using the next
six days (one week) maintenance as the comparison result, the average recall of ten days’ prediction of
recommended machine list is 57.1%, and then considering the factory’s limited daily maintain energy,
we decide a rule to rank the recommended machine list and only give the top 5 machine for
recommended machine list. The ten days’ prediction of top 5 recommended machine list average
precision is 58%, it means that there are three machines will be repaired in the recommended five
machines.
關鍵字(中) ★ 維修預測
★ 推薦系統
★ 生產資料
★ 時間序列
關鍵字(英) ★ Predictive Maintenance
★ Recommendation system
★ production data
★ time series
論文目次 目錄
中文摘要........ i
Abstract..........ii
目錄..iii
圖目錄...........vi
表目錄..........vii
第一章 緒論.. 1
1-1 研究背景 ..... 1
1-2 研究動機 ..... 2
1-3 研究貢獻 ..... 2
1-4 論文架構 ..... 3
第二章 相關研究....... 4
2-1 預測性維修 . 4
2-2 預測模型 ..... 5
2-2-1 Random Forest ........... 5
2-2-2 Support Vector Machine-rbf.... 5
2-2-2 Long Short-Term Memory ....... 7
2-3 預測風險問題 .......... 8
2-4 EM 演算法估計過去的機器良率...... 9
第三章 問題定義與研究...... 11
3-1 問題定義 ... 11
3-2 目標式定義 ............ 12
第四章 系統架構..... 14
4-1 系統背景說明 ........ 14
4-1-1 工廠運作模式.......... 14
4-1-2 資料型態說明.......... 15
4-1-3 系統實驗摘要.......... 16
4-2 系統架構說明 ........ 17
第五章 實驗與討論. 19
5-1 資料集 ....... 19
5-2-1 資料前處理......... 20
5-2-2 資料集切割......... 21
5-3 實驗架構 ... 22
5-4 實驗一: 尋找與維修紀錄有高關聯性的機器 .......... 23
5-4-1 實驗目的..... 23
5-4-2-1 實驗方法: 資料統計.......... 23
5-4-2-2 實驗結果: 資料統計.......... 25
5-4-3-1 實驗方法: Association Rule ............ 25
5-4-3-2 實驗結果: Association Rule ............ 27
5-5 實驗二: 模型比較與參數設定 ........ 28
5-5-1 實驗目的..... 28
5-5-2 實驗評估標準.......... 28
5-5-3-1 實驗方法: 單週期資料與累計資料之比較 29
5-5-3-2 實驗結果.. 30
5-5-4-1 實驗方法: 預測模型參考天數之比較........ 31
5-5-4-2 實驗結果.. 31
5-5-5-1 實驗方法: 特徵因子之比較........... 32
5-5-5-2 實驗結果.. 33
5-5-6-1 實驗方法: Random Forest、 SVM、 LSTM 演算法比較...... 33
5-5-6-2 實驗結果.. 35
5-5-7 實驗總結..... 35
5-6 實驗三: 異常維修預測模型在工廠的實際表現 ...... 36
5-6-1-1 實驗方法: 以資料集的標籤作為評斷標準 36
5-6-1-2 實驗結果.. 37
5-6-2-1 實驗方法: 以六天內的維修機器作為評斷標準..... 37
5-6-2-2 實驗結果.. 38
5-6-3-1 實驗方法: 推薦前幾名機器維修列表........ 39
5-6-3-2 實驗結果.. 40
第六章 結論與未來展望...... 42
6-1 結論........... 42
6-1 未來展望... 42
參考文獻..... 44
參考文獻 [1] 黃培凱, "Analyze the micro-crak rate of PCB based on Expectation-Maximization
algotithm" National Center University, Taoyuan, 2019.
[2] O.Motaghare, A. S.Pillai, andK. I.Ramachandran, “Predictive Maintenance Architecture,”
2018 IEEE Int. Conf. Comput. Intell. Comput. Res. ICCIC 2018, pp. 13–16, 2018, doi:
10.1109/ICCIC.2018.8782406.
[3] G. A.Susto, A.Schirru, S.Pampuri, S.McLoone, andA.Beghi, “Machine learning for
predictive maintenance: A multiple classifier approach,” IEEE Trans. Ind. Informatics, vol.
11, no. 3, pp. 812–820, 2015, doi: 10.1109/TII.2014.2349359.
[4] D.Bruneo andF.DeVita, “On the use of LSTM networks for predictive maintenance in
smart industries,” Proc. - 2019 IEEE Int. Conf. Smart Comput. SMARTCOMP 2019, pp. 241–
248, 2019, doi: 10.1109/SMARTCOMP.2019.00059.
[5] A. T.Prihatno, H.Nurcahyanto, andY. M.Jang, “Predictive Maintenance of Relative
Humidity Using Random Forest Method,” 3rd Int. Conf. Artif. Intell. Inf. Commun. ICAIIC
2021, pp. 497–499, 2021, doi: 10.1109/ICAIIC51459.2021.9415213.
[6] C.Chen, N.Lu, B.Jiang, andC.Wang, “A Risk-Averse Remaining Useful Life Estimation for
Predictive Maintenance,” IEEE/CAA J. Autom. Sin., vol. 8, no. 2, pp. 412–422, 2021, doi:
10.1109/JAS.2021.1003835.
[7] C.Ding, “Risk assessment of physical and chemical analysis laboratory based on the
laboratory risk assessment model,” Proc. - 2021 2nd Int. Conf. Urban Eng. Manag. Sci.
ICUEMS 2021, pp. 293–296, 2021, doi: 10.1109/ICUEMS52408.2021.00070.
[8] J.Peng, “Safety risk assessment technology of large caisson launching for gravity wharf,”
ICTIS 2019 - 5th Int. Conf. Transp. Inf. Saf., pp. 773–777, 2019, doi:
10.1109/ICTIS.2019.8883785.
[9] W.Wang andS.Shen, “Risk assessment model and application for the urban buried gas
pipelines,” Proc. 2009 8th Int. Conf. Reliab. Maintainab. Safety, ICRMS 2009, pp. 488–492,
2009, doi: 10.1109/ICRMS.2009.5270145.
[10] B.Li, S.Wan, H.Xia, andF.Qian, “The Research for Recommendation System Based on
Improved KNN Algorithm,” Proc. 2020 IEEE Int. Conf. Adv. Electr. Eng. Comput. Appl.
AEECA 2020, pp. 796–798, 2020, doi: 10.1109/AEECA49918.2020.9213566
指導教授 梁德容(Deron Liang) 審核日期 2021-10-26
推文 facebook   plurk   twitter   funp   google   live   udn   HD   myshare   reddit   netvibes   friend   youpush   delicious   baidu   
網路書籤 Google bookmarks   del.icio.us   hemidemi   myshare   

若有論文相關問題,請聯絡國立中央大學圖書館推廣服務組 TEL:(03)422-7151轉57407,或E-mail聯絡  - 隱私權政策聲明