請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/92210
標題: | 應用於智慧製造之自配置物聯網系統之設計與驗證 Design and Validation of a Self-Configuring IoT System for Smart Manufacturing |
作者: | 林奕安 Ian Lin |
指導教授: | 蔡曜陽 Yao-Yang Tsai |
關鍵字: | 工業物聯網,分佈式計算,霧計算,服務放置問題,多目標優化, Industrial Internet of Things,distributed computing,Fog Computing,Service Placement Problem,Multi-objective Optimization, |
出版年 : | 2024 |
學位: | 碩士 |
摘要: | 製造業正處於智慧化轉型與集成管理需求急劇增長的關鍵時期,然而建立適合製造場域的網路架構並不容易。歷年來,將分布式計算應用於工業物聯網,特別是在協同和標準化的網路架構方面,已經得到了充分的研究與發展。然而,這一領域仍面臨諸多挑戰,例如目前的分佈式計算環境優化目標與製造場域的需求不夠契合;或者,模型構建時忽略生產設備的異質性以及生產任務對通訊行為的影響。
本研究旨在開發一套專門應對智慧製造領域服務放置問題(Service Placement Problem, SPP)的系統,該系統會針對目標場域建構一個最佳化模型,以提供網路架構配置。系統核心是以「生產設備與智能化設備連線」特徵為基礎建立的啟發式網路架構最佳化模型,模型的優化目標涵蓋了場域的網路拓樸及拓樸節點的軟體部署,著重於通訊服務品質的最大化和場域失效風險的最小化。 實驗結果顯示,此系統不僅能夠提供有效的網路架構方案,其效能也顯著優於非結構化的隨機生成方案。 The manufacturing industry is currently undergoing a critical period of intelligent transformation and an increasing demand for integrated management. However, establishing a network architecture suitable for manufacturing environments is challenging. Over the years, the application of distributed computing in the Industrial Internet of Things (IIoT ), especially regarding collaborative and standardized network architectures, has been extensively researched and developed. Yet, this field still faces numerous challenges, such as the current distributed computing environment''s optimization objectives not fully aligned with the needs of manufacturing sites, or the overlook of production equipment heterogeneity and the impact of production tasks on communication behavior during model construction. This study aims to develop a system specifically designed to address the Service Placement Problem (SPP) in the smart manufacturing domain. The system constructs an optimization model for the target domain to provide network architecture configuration. At its core, the system features a heuristic network architecture optimization model based on the characteristics of "connections between production and intelligent devices." The optimization objectives of the model include the network topology of the site and the software deployment on topology nodes, focusing on maximizing communication service quality and minimizing the risk of domain failure. Experimental results show that this system not only provides effective network architecture solutions but also significantly outperforms unstructured random generation schemes. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/92210 |
DOI: | 10.6342/NTU202400739 |
全文授權: | 同意授權(限校園內公開) |
顯示於系所單位: | 機械工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-112-1.pdf 授權僅限NTU校內IP使用(校園外請利用VPN校外連線服務) | 3.73 MB | Adobe PDF | 檢視/開啟 |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。