Resource Orchestration of Cloud-Edge–based Smart Grid Fault Detection

• DOI: 10.1145/3529509
• Publication Date: 2022-04-04T11:13:10Z
• Authors: Jie Li; Yuxing Deng; Wei Sun; Weitao Li; Ruidong Li; Qiyue Li; Zhi Liu
• Keywords: Edge device

Real-time smart grid monitoring is critical to enhancing resiliency and operational efficiency of power equipment. Cloud-based edge-based fault detection systems integrating deep learning have been proposed recently monitor the in real time. However, state-of-the-art cloud-based may require uploading a large amount data suffer from long network delay, while schemes do not adequately consider requirement thus cannot provide flexible optimal performance. To solve these problems, we study cloud-edge based hybrid system. Embedded devices are placed at edge monitored equipment with several lightweight neural networks for detection. Considering limited communication resources, relatively low computation capabilities devices, different accuracies supported by networks, design an computational resource allocation method this Our can maximize processing throughput system improve utilization satisfying transmission latency requirements. Extensive simulations conducted results show superiority scheme over comparison schemes. We also prototyped verified its feasibility performance real-world scenarios.