設計雲邊微服務部署之再生能源監控平台

Design a Cloud-Edge Microservice Deployment for Renewable Energy Monitoring Platform

徐瑛汝(Ying-Ru Hsu)

指導教授 ： 陳弘明

國立臺中科技大學/資訊與流通學院/資訊工程系碩士班/碩士(2022年)

摘要

再生能源透過物聯網技術收集發電廠設備端大量的電力感測數據，傳統以單體式應用服務結合雲端監控之架構，會將原始電力資料傳送至雲端的資料中心進行運算後，並將電力監控指標之運算結果回傳至再生能源監控平台，藉此監控再生能源發電廠之供電狀況。然而，當大量設備產生之數據傳輸至雲端，容易消耗網路的頻寬使用，對於網際網路與雲端基礎架構之運作更造成龐大的運算資源負擔。因此，本研究設計一套基於KubeEdge雲邊協同微服務部署之再生能源監控平台，以雲邊微服務部署架構解決跨網域發電廠對於再生能源平台之監控問題。KubeEdge以WebSocket串聯雲端與邊緣端，提供在不同區域發電廠下，雲端與邊緣端跨網域之溝通機制，並設計微服務之雲邊容器化部署，同時達到資料預處理與數據共享之價值。而為了確保本研究設計之KubeEdge雲邊微服務部署監控架構的正常運作，於實驗的部分將比較雲端和邊緣端之間，面對多電廠大量電力資料處理之服務延遲回應差異。同時，驗證邊緣端發電廠在面對離線之自我修復與節點重啟時，以及版本升級之滾動更新時，能夠在短時間內讓再生能源監控平台維持穩定的電力監控運作，確保高可用的再生能源監控架構。

關鍵字

微服務 ； 雲邊協同 ； KubeEdge ； 再生能源 ； 能源管理

並列摘要

Renewable energy collects a large amount of power monitoring data on the equipment side of the power plant through the Internet of Things (IoT) technology. Traditionally, a monolithic application combined with the cloud monitoring structure will transmit the raw data to the cloud data center for calculation, and then monitor the power. The calculation results of the indicators are sent back to the renewable energy monitoring platform to monitor the power supply status of the renewable energy power plants. However, when the data generated by a large number of devices is transmitted to the cloud, the bandwidth usage of the network is easily consumed, and the operation of the Internet and cloud infrastructure is even more burdened on computing resources. Hence, this study will design a renewable energy monitoring platform based on KubeEdge cloud-edge collaborative microservice deployment, and will use the cloud-edge microservice deployment architecture to solve the monitoring problem of cross-domain power plants for renewable energy platforms. KubeEdge uses WebSocket to connect the cloud and the edge to provide a cross-domain communication mechanism between the cloud and the edge under different regional power plants, and to design the cloud-edge containerized deployment of microservices, while achieving the value of data preprocessing and data sharing. In order to ensure the normal operation of the KubeEdge cloud-edge microservice deployment monitoring architecture designed in this study, the experimental part will compare the difference in service delay response between the cloud and the edge in the face of a large amount of power data processing in multiple power plants. At the same time, it is verified that the edge power plant can maintain stable power monitoring operation of the renewable energy monitoring platform in a short period of time in the face of offline self-healing and node restart, and rolling update of version upgrades. This ensures that a highly available renewable energy monitoring architecture is proposed.

並列關鍵字

Microservices ； Cloud-Edge Collaboration ； KubeEdge ； Renewable Energy ； Energy Management

參考文獻

[1]張忠良(2020)。智慧電網發展與再生能源監控實務【專題】。MAR 2020工程，93(1)，39-49。

[2]S. Ansari et al., "A Review of Monitoring Technologies for Solar PV Systems Using Data Processing Modules and Transmission Protocols: Progress, Challenges and Prospects", Sustainability, vol. 13, no. 15, 2021.

[3]A. Fakhar et al., "Smart grid mechanism for green energy management: a comprehensive review", International Journal of Green Energy, pp. 1-25, 2022.

[4]A. V. Balzamov et al., "Development of a methodology for migration of monolithic systems to microservice architecture using cloud technologies", Journal of Physics: Conference Series, vol. 2091, no. 1, 2021.

[5]S. R. Goniwada, "Cloud Native Architecture and Design Patterns", Cloud Native Architecture and Design, pp. 127-187, 2022.