| 摘要: | 本研究計畫參考雲製造的概念，將利用自建的私有雲平台、最新的虛擬化技術(如Docker、Docker Compose、Docker Swarm與Kubernetes等)、開源Web技術(如HTML5、Vue.js、Angular、Node.js和RESTful Web Services等)、虛擬量測技術、大數據處理技術(Hadoop HDFS、Spark)與深度學習技術來研發一個「具備持續整合交付與巨量資料處理能力之新一代虛擬量測服務平台」。本計畫將以3年完成16個研究主題。在第1年，我們將研製一個充分利用容器技術優點且具備持續整合與交付能力之新一代虛擬量測雲端服務平台。在第2年，我們利用Docker、Kubernetes、開源Web技術與小型嵌入式電腦，研製一個「具備容器化應用程式自動化部署與高可用性能力之工廠端虛擬量測維運平台」。在第3年，我們將研發一個「具備巨量資料處理與深度學習能力之虛擬量測平台」，可以有效地解決現行AVM系統無法進行EDM即時虛擬量測的問題。本計畫所研發之基於容器虛擬化技術之「具備持續整合交付與巨量資料處理能力之新一代虛擬量測服務平台」，將可以有效解決目前基於傳統虛擬化技術之雲端虛擬量測服務平台之各項缺失以及其在實際開發與應用部署所遭遇的問題，可大幅提升現行虛擬量測系統之境界。本計畫之研究主題具有挑戰性，研究成果具備技術的創新性、前瞻性與產業利用性，將可以應用於促進製造產業之智能化生產，也可為智慧製造產業訓練相關的技術人才。
By referring to the concept of Cloud Manufacturing, this project will develop a new-generation virtual metrology service platform with the capabilities of CI/CD and big-data processing based on private cloud, the newest virtualization technologies (e.g., Docker, Docker Compose, Docker Swarm, and Kubernetes), open-source Web technologies (e.g., HTML5, Vue.js, Angular, Node.js, and RESTful Web Services), virtual metrology technology, big-data processing technologies (e.g., Hadoop HDFS and Spark), and deep learning. This project will complete 16 research topics in three years. In the first year, we will develop a new-generation virtual metrology cloud service platform which can fully leverage the merits of container technology and possesses CI/CD capabilities. In the second year, we will utilize Docker, Kubernetes, and some tiny embedded computers to develop a factory-side virtual metrology operating platform which has the capabilities of automatic deployment and high availability of containerized applications. In the third year, we will develop a virtual metrology platform which can process big data and leverage deep learning to conduct on-line virtual metrology for EDM. The new-generation virtual metrology service platform developed by this project can effectively resolve the shortcomings of the existing virtual metrology service platform built by tradition virtualization technologies in the aspects of software development, deployment, and application, which can greatly enhance the realm of the existing virtual measurement systems. The research topics of this project are challenging. The research results are expected to have technology novelty, perspectiveness, and industrial usefulness, which can facilitate the manufacturing industry to achieve intelligent production. The project can also train related technical personnel for the intelligent manufacturing industry. |
