High-Precision Timing Method of BeiDou-3 System Based on Reinforcement Learning

Abstract

All aspects of smart grid operation require precise timing technology to improve the timeliness and precision of the network. However, due to the complex and time-varying network environment and the low precision of device timing, the existing timing technology cannot meet the timing precision required by smart grid services. Therefore, according to the multi-armed bandit (MAB) theory and the upper confidence bound (UCB) algorithm, this paper proposes a high-precision timing method for power transmission and distribution park named adaptive timing-aware upper confidence bound (AUCB). The long-term average reward is maximized based on local information and historical decision-making experience. The simulation results show that the proposed high-precision timing method of AUCB in the power transmission and distribution park can effectively improve the device timing precision and realize self-adaptive timing for power system devices.