基于智能电网的充电站运营模型优化研究

Optimisation Study of Charging Station Operation Model Based on Smart Grid

介绍了一种以用户为基础的Q-Learning算法模型,目的在于模拟用户充电行为,精确预测充电站网络中各站点充电需求,进而制定最优运营策略。在电动汽车充电阶段,充电站负责监控并调整电动汽车充放电量,使之与电力系统负载响应相匹配,从而辅助电力系统处理风电大量接入问题。基于此,构建了一种二维多级调度操作模式,该模式对风能发电不确定性进行了预估,提升了电力调配应急储备能力。通过管理和协调电力充电站,增强电网对风能发电吸收能力,充分发挥充电站调度各类资源优势,进而改善运营策略。

A user-based Q-Learning algorithm model is described,which aims to simulate the charging behaviour of users,accurately predict the charging demand of each station in the charging station network,and then formulate the optimal operation strategy.During the EV charging phase,the charging station is responsible for monitoring and adjusting the EV charging and discharging quantities to match the load response of the power system,thus assisting the power system to deal with the problem of large-scale access to wind power.To cope with this,a two-dimensional multilevel dispatch operation model is constructed,which anticipates the uncertainty of wind power generation and improves the emergency reserve capacity for power deployment.Through the management and coordination of power charging stations,the absorption capacity of the grid for wind power generation is enhanced,and the advantages of various types of resources in charging station dispatch are fully utilised to improve the operation strategy.