Accurate time synchronization of power reference station based on BD3 system

Abstract

A Beidou 3 (BD3) system-based power reference station can provide high-precision time synchronization for power distribution systems by sending synchronization data packets to devices in a multi-hop routing fashion. However, optimizing route selection to reduce both time synchronization error and delay is a challenging problem. In this paper, we establish a software-defined network-enabled power reference station time synchronization framework based on BD3. Then, we formulate the joint problem to minimize cumulative synchronization error and delay through multi-hop route selection optimization. A back propagation (BP) neural network-improved intelligent time synchronization route selection algorithm named BP-RS is proposed to learn the optimal route selection, which uses a BP neural network to dynamically adjust the exploration factor to achieve rapid convergence. Simulation results show the superior performance of BP-RS in synchronization delay, synchronization error, and adaptability with changing routing topologies.