跳到主要內容

臺灣博碩士論文加值系統

(44.211.31.134) 您好!臺灣時間:2024/07/22 20:37
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:張翔淨
研究生(外文):CHANG, HSIANG-CHING
論文名稱:使用 Azure 實現預知保養系統架構-以 TFT-LCD 廠為 例
論文名稱(外文):The Implementation of a Predictive MaintenanceArchitecture Using Azure - The Case for TFT-LCDManufacturing
指導教授:楊朝棟楊朝棟引用關係
指導教授(外文):YANG, CHAO-TUNG
口試委員:許慶賢劉榮春欉振坤張志宏
口試委員(外文):HSU, CHING-HSIENLIU, JUNG-CHUNTSUNG, CHEN-KUNCHANG, CHIH-HUNG
口試日期:2022-07-05
學位類別:碩士
校院名稱:東海大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2022
畢業學年度:110
語文別:英文
論文頁數:74
中文關鍵詞:智慧製造預知保養雲端服務數據處理機器學習ETLPySpark
外文關鍵詞:Smart ManufacturingPredictive MaintenanceCloud Services,Data ProcessingMachine Learning,ETLPySpark
相關次數:
  • 被引用被引用:0
  • 點閱點閱:219
  • 評分評分:
  • 下載下載:38
  • 收藏至我的研究室書目清單書目收藏:0
以往設備維護的方式是設備壞了才修,以此降低維護成本,又或是計畫性維修,維修人員依照過往經驗,到了機器運行的一定次數或是時間來定期更換,但這樣的方式無法考量到環境及不同元件造成的差異,仍會造成設備損壞,而非預期的停機,讓不管是產能還是維修等費用都大大的損失。工業 4.0 的興起帶起全球邁向智慧製造,製造業結合物聯網、大數據及 AI 等技術,讓現在設備維護的工作可以透過收集機器的電流、溫度及其他機台參數資訊,進一步進行數據分析來做到機台的預知保養,提早進行機台保養、維修,避免非預期的停機,影響產線運行。本論文將以 TFT-LCD 面板零組件製造業作為實驗場域,實作透過 Azure 雲端服務平台來建置 TFT-LCD 機台預知保養系統,透過皮爾森相關性等分析,找到適合本實驗場域使用的參數,利用 PySpark 提高資料處理的速度,並利用分區方式優化資料表,Operator Cost、I/O Cost 和 CPU Cost 分別提升了 98.77%、98.78% 和 98.74%,且在面對不同機台數據會有差異的情況下,每一個機台建置一個隨機森林模型來進行數據的分析,模型準確率為 0.99,且將模型部屬至 Azure Kubernetes 來進行即時的評分,最後也將數據以及模型分析結果視覺化,讓工廠的維修人員能夠透過數據以及分析結果來調整製程參數、提早了解機台健康狀況,達到預知保養的工作。
The rise of Industry 4.0 has brought the world to smart manufacturing. The
manufacturing industry combines technologies such as the Internet of Things, big
data, and AI. The current equipment maintenance can further analyze the recent
equipment maintenance work by collecting machine current, temperature, and other machine parameter information. To achieve the predictive maintenance of the machine, carry out the maintenance and repair of the device in advance to avoid unexpected downtime and affect the operation of the production line. This paper work will take the TFT-LCD panel component manufacturing industry as the experimental field and implement the TFT-LCD machine predictive maintenance system through the Azure cloud service platform. Through applying the analysis of Pearson correlation, the influential parameters of the experimental domain could be found. PySpark was utilized to shorten the processing time. Also, using partition to optimize the data table, the operator cost, I/O cost and CPU cost were decreased by 98.77%, 98.78%, and 98.74% respectively. And in view of the face of differences in the data from different machines, each machine builds a random forest model to analyze the data, The experimental model selected was with a F1 score of 0.99 and deployed onto Azure Kubernetes Service for real-time factory maintenance personnel can adjust the process parameters through the data and analysis results, understand the health status of the machine in advance, and achieve predictive maintenance.

摘要
致謝
Abstract
致謝詞
Table of Contents
List of Tables
List of Figures
Chapter 1 Introduction
Chapter 2 Background Review and Related
Works
Chapter 3 System Design and
Implementation
Chapter 4 Experimental Results
Chapter 5 Conclusions and Future Works
References
Appendix A ETL
Appendix B Model Deploy
Appendix C Create New Job
[1]Serkan Ayvaz and Koray Alpay. Predictive maintenance system for production
lines in manufacturing: A machine learning approach using iot data in realtime. Expert Systems with Applications, 173:114598, 2021.

[2] Ari Saptawijaya Siti Aminah Denny, I. Putu Medagia Atmaja. Implementation of change data capture in etl process for data warehouse using hdfs
and apache spark. In International Workshop on Big Data and Information
Security, pages 49–55. IEEE, 2017.

[3] Priyanshi Singh Julian Soh. Data Science Solutions on Azure. Apress Berkeley, CA, 2020.

[4] Kaijisse Waaijer Christoph Korner. Mastering Azure Machine Learning.
Packt, 2020.

[5] Ravi Helon M. S. Ferreira, Lucas O. de Figueiredo, Rafael B. C. Lima, Luiz
Antonio Pereira Silva, and Pericles R. Barros. Iiot solution for predictive monitoring based on vibration data from motors using microsoft azure machine
learning studio and power bi. In 2021 14th IEEE International Conference
on Industry Applications (INDUSCON), pages 145–152, 2021.

[6] Ned Bellavance Steve Buchanan, Janaka Rangama. Introducing Azure Kubernetes Service. Apress Berkeley, CA, 2020.

[7] Shimon Ifrah. Store and Manage Docker Container Images on Azure Container Registry (ACR), pages 27–55. Apress, Berkeley, CA, 2020.

[8]Ajit Pratap Kundan Mainak Chakraborty. Monitoring Cloud-Native Applications. Apress Berkeley, CA, 2021.

[9] Adam Freeman. Pro ASP.NET MVC 5 Platform. Apress Berkeley, CA, 2014.

[10] Eng Chai Ang and Shahrel Azmin Suandi. Smart manufacturing with an artificial neural network to predict manufacturing healthiness. In 2019 IEEE 15thInternational Colloquium on Signal Processing and Its Applications (CSPA),
pages 120–123, 2019.

[11] Serkan Ayvaz and Koray Alpay. Predictive maintenance system for production
lines in manufacturing: A machine learning approach using iot data in realtime. Expert Systems with Applications, 173:114598, 01 2021.

[12] Thyago P. Carvalho, Fabrizzio A. A. M. N. Soares, Roberto Vita, Roberto
da P. Francisco, Joao P. Basto, and Symone G. S. Alcala. A systematic literature review of machine learning methods applied to predictive maintenance.
Computers and Industrial Engineering, 137:106024, 2019.

[13] Matteo Calabrese, Martin Cimmino, Francesca Fiume, Martina Manfrin,
Luca Romeo, Silvia Ceccacci, Marina Paolanti, Giuseppe Toscano, Giovanni
Ciandrini, Alberto Carrotta, Maura Mengoni, Emanuele Frontoni, and Dimos
Kapetis. Sophia: An event-based iot and machine learning architecture for
predictive maintenance in industry 4.0. Information, 11(4), 2020.

[14] Chin-Wei Tien, Tse-Yung Huang, Ping Chun Chen, and Jenq-Haur Wang.
Automatic device identification and anomaly detection with machine learning
techniques in smart factories. In 2020 IEEE International Conference on Big
Data (Big Data), pages 3539–3544, 2020.

[15] Xiaonan Shi, Takenori Baba, Daisuke Osagawa, Mitsushiro Fujishima, and
Teruaki Ito. Maturity assessment: A case study toward sustainable smart
manufacturing implementation. In 2019 IEEE International Conference on
Smart Manufacturing, Industrial and Logistics Engineering (SMILE), pages
155–158, 2019

[16]Patrick Straus, Markus Schmitz, Rene Wostmann, and Jochen Deuse. Enabling of predictive maintenance in the brownfield through low-cost sensors,
an iiot-architecture and machine learning. In 2018 IEEE International Conference on Big Data (Big Data), pages 1474–1483, 2018.

[17] Chen TY. Kristiani E. et al Yang, CT. The implementation of data storage
and analytics platform for big data lake of electricity usage with spark. The
Journal of Supercomputing, 2021.

[18] Chao-Tung Yang, Cai-Jin Chen, Yu-Tse Tsan, Po-Yu Liu, Yu-Wei Chan, and
Wei-Chen Chan. An implementation of real-time air quality and influenza-like
illness data storage and processing platform. Computers in Human Behavior,
100:266–274, 2019.

[19] Chao-Tung Yang, Shuo-Tsung Chen, Walter Den, Yun-Ting Wang, and Endah Kristiani. Implementation of an intelligent indoor environmental monitoring and management system in cloud. Future Generation Computer Systems, 96:731–749, 2019.

[20] Chen ST. Liu JC. et al Yang, CT. On construction of an energy monitoring
service using big data technology for the smart campus. Cluster Computing,
2020.

[21] Chih-Hung Chang, Fuu-Cheng Jiang, Chao-Tung Yang, and Sheng-Cang
Chou. On construction of a big data warehouse accessing platform for campus power usages. Journal of Parallel and Distributed Computing, 133:40–50,
2019.

[22] Yu-Chuan Lin, Min-Hsiung Hung, Hsien-Cheng Huang, Chao-Chun Chen,
Haw-Ching Yang, Yao-Sheng Hsieh, and Fan-Tien Cheng. Development of
advanced manufacturing cloud of things (amcot)—a smart manufacturing
platform. IEEE Robotics and Automation Letters, (3):1809–1816, 2017.

[23] Marina Paolanti, Luca Romeo, Andrea Felicetti, Adriano Mancini, Emanuele
Frontoni, and Jelena Loncarski. Machine learning approach for predictivmaintenance in industry 4.0. In 2018 14th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA),
pages 1–6, 2018.

[24] Chnar Mustafa Mohammed and Subhi R.M Zeebaree. Sufficient Comparison Among Cloud Computing Services: IaaS, PaaS, and SaaS: A Review.
International Journal of Science and Business, 5(2):17–30, 2021.

[25] Liu JC. Chen ST. et al Yang, CT. Implementation of a big data accessing and
processing platform for medical records in cloud. Journal of Medical Systems
volume, 2017.

[26] Castellano J.G. Moral-Garcia S. et al Mantas, C.J. A comparison of random
forest based algorithms: random credal random forest versus oblique random
forest. pages 10739–10754, 2019.

[27] Kaijun Wu and Yonghua Zhang. Study on the construction of a structured
scenario system for conventional emergencies based on random forest. In
2020 International Conference on Robots and Intelligent System (ICRIS),
pages 365–369, 2020.

[28] Suresh G.V. and Srinivasa Reddy E.V. Uncertain data analysis with regularized xgboost. Webology, 2022.

[29] Thorsten Joachims. Making large scale svm learning practical. Advances in
Kernel Methods: Upport Vector Machines, 10 1999.

[30] R. Rithesh. Svm-knn: A novel approach to classification based on svm and
knn. International Research Journal of Computer Science, 4, 08 2017.

[31] Yesuel Kim and Youngchul Kim. Explainable heat-related mortality with
random forest and shapley additive explanations (shap) models. Sustainable
Cities and Society, 79:103677, 2022.

[32] Debzani Deb and Russell M. Smith. Application of random forest and shap
tree explainer in exploring spatial (in)justice to aid urban planning. ISPRS
International Journal of Geo-Information, 10, 2021.

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