A High Concurrency and Low-Delay Data Scheduling Method for Massive Power Devices

Abstract

With the massive processing devices connected to the power system, the demand for data processing increases gradually. However, in the high concurrency scenario, the simultaneous processing of massive data will lead to the depletion of computing resources of the power system and the suspension of power services. To address this problem, this paper first proposes the classification method of power system data based on a neural network. Secondly, this paper proposes the data priority model based on the entropy method to determine the processing order of massive data of power devices. Then, a problem of high concurrency and low-delay data scheduling is constructed. To address this problem, we proposed an intelligent data scheduling algorithm based on upper confidence bound (UCB). Finally, the feasibility and effectiveness of the method are proved by simulation.