Research and application of pumped-storage hydroelectricity video monitoring agent based on cloud tube edge collaboration

Abstract

With the development of the new generation of artificial intelligence technology, machine learning and deep learning frameworks have become important tools for developing and promoting artificial intelligence applications. On the premise of fully considering the limitations of computing resources of end devices, a video monitoring intelligent agent for pumped storage power stations based on cloud management edge end collaboration was proposed to solve the problem of traditional reliance on cloud data centers for intensive computing. Through the hierarchical deployment of algorithm models and the cloud management side to side collaboration mechanism, the problems of the video surveillance agent in the pumped storage power plant in computing power, network, storage, etc. are solved to a great extent, the high latency and high energy consumption caused by the transmission of large amounts of data over a long distance WAN are reduced, the agent edge computing mode is optimized, and the ability to apply AI technology with high computational complexity is improved, Improve the real-time application level of video monitoring intelligent agents in pumped storage power stations.

Keywords

intelligent internet of things / edge computing / markov decision process / task reasoning / offloading decision / intelligent video surveillance