摘要: | 著眼於虛實整合技術(Cyber-Physical Technology)對建置智慧工廠的重要性,本研究利用物聯網與雲端運算等虛實整合技術,開發一個可應用於智慧工廠的資通平台。本資通平台包含3部份:工廠端、雲端及使用者端。在工廠端,本研究利用樹梅派(Raspberry Pi) 袖珍型電腦(簡稱RPi)設計許多資料收集器(Data Collector),可用來收集工廠內機台的各種資料。同時,本研究採用RESTful網路服務技術,設計資料收集器之間的通訊機制,使得資料收集器間可以很容易地互相通訊,進而形成工廠內的物聯網平台,得以實現工廠內機台的水平整合。本研究也利用RPi設計一個閘道器(Gateway),可用來管理工廠內的資料收集器,並且能與雲端服務通訊。在雲端,本研究採用RESTful網路服務技術開發各種製造相關的服務(例如,資料收集服務、模型建置服務、虛擬量測服務等),並部署於私有雲上,使得工廠內各種不同的物件(如:閘道器、Web GUIs等)可透過網路隨需使用雲端服務。雲端也包含網頁伺服器,用以裝載各式網頁式使用者介面(Web GUIs),也包含資料庫,用以儲存工廠內生產的各種相關資料。在使用者端,使用者可利用瀏覽器從雲端網路伺服器下載Web GUIs,透過網路來操作本資通平台。本論文已提出建立智慧工廠資通平台的系統性方法,相關成果可作為建構智慧工廠之參考。
Considering the importance of the Cyber-Physical technology on building smart factories, this thesis leverages technologies of IoT (Internet of Things) and cloud computing to build an information and communication platform (ICP) for smart factories. The ICP consists of three parts: factory side, cloud side, and user side. On the factory side, Raspberry Pi minicomputers are used to develop data collectors, each of which can be used to collect production data of machines or equipment. Also, RESTful Web service technology is used to design the communication mechanism among data collectors so that data collectors are capable of easily communicating with each other, thereby forming an IoT platform in the factory, which can achieve horizontal integration of machines. In addition, a Raspberry Pi minicomputer is used to develop a gateway in the factory, which can be used to manage data collectors and communicate with cloud services. On the cloud side, several RESTful cloud services (e.g., data collection service, model creation service, and virtual metrology service) are constructed so that different objects (e.g., gateway and Web GUIs) can easily access and consume these cloud services on demand. On the user side, the user can use a browser to download Web GUIs from the cloud to remotely operate the ICP through networks. This thesis has presented a systematic approach for creating the information and communication platform of smart factories, and related research results can be a useful reference for constructing smart factories. |