Adjustable Potential Evaluation of Massive Flexible Load Resources from the Perspective of CPS
Jiaxiang Sun, Lei Gan, Xingying Chen, Peng Chen, Haochen Hua · 2023
With the demand-side flexible resources attracting widespread attention, the potential evaluation of these resources becomes the primary issue in their use. However, the efficient use of massive resources is not only determined by its physical adjustment capabilities but also affected by information systems. Therefore, it is necessary to comprehensively consider the potential evaluation of these resources from the perspective of cyber-physical systems. An adjustable potential evaluation method of massive flexible load resources is proposed here under the framework of the cloud-edge collaboration cyber system (CECS). Firstly, the CECS framework of regulating massive demand-side resources is constructed, followed by the modeling of communication latency and reliability. Then, the maximum adjustable potential evaluation model is established based on a fixed time-delay search. Finally, a deep reinforcement learning method combined with the k shortest path algorithm is proposed to solve the optimization model based on proximal policy optimization. The numerical result shows the feasibility and rationality of the proposed method.