考虑电网调峰需求的工业园区主动配电系统调度学习优化

本文针对含光伏(Photovoltaic, PV)、全钒液流电池(Vanadium redox battery, VRB)储能装置与多类型柔性负荷的工业园区主动配电系统, 研究在考虑源荷随机性情况下该系统的动态经济调度问题. 首先, 将PV出力、多类型负荷需求和电网调峰需求的随机动态变化近似描述为连续马尔科夫过程, 并根据系统内VRB的充放电特性对储能系统进行建模; 然后, 以各决策时刻下PV出力、负荷需求、调峰需求以及储能荷电状态(State of charge, SOC)的离散等级为状态, 以储能充放电及多类型柔性负荷调整方案为行动, 在系统功率平衡等相关约束下, 以应对电网调峰需求和提高系统经济运行水平为目标, 将工业园区主动配电网系统动态经济调度优化问题建立成随机动态规划模型; 最后, 引入强化学习方法进行策略求解. 算例仿真结果表明所得策略可有效提高系统经济运行效益, 并在一定程度上满足电网调峰需求.

Learning-based Optimization of Active Distribution System Dispatch in Industrial Park Considering the Peak Operation Demand of Power Grid

The dynamic economic dispatch problem of the active distribution system combined of photovoltaic (PV), vanadium redox battery (VRB) energy storage device and multiple types of flexible load in industrial parks with uncertain renewable sources and demands is focused in this paper. First, the random dynamic variations of photovoltaic, multiple loads demand and peak operation demand are described as continuous Markov processes, and the VRB energy storage system is modeled considering its charge-discharge characteristics. Then, decision epoch, outputs level of photovoltaic, multiple load demands level, peak operation demands level and state of charge (SOC) level of VRB are defined as states of the system, the adjustment level of VRB and multiple types of flexible load are set as the actions. Based on relevant restrictions including the power balance constraint, the dynamic optimal dispatch problem for the system was described as a stochastic dynamic programming model, which aims to meet the peak operation demand of power grid and realize economic operation of the system. Finally, a reinforcement learning method is adopted to obtain the optimal policy. Simulation results show that the operational efficiency is significantly enhanced and the peak operation demand of power grid is partly satisfied by the optimal policy.