含随机负荷的微电网储能优化运行策略

Optimized Operation Strategy of Energy Storage in Microgrid with Random Load

供需两侧的随机性会对微电网的稳定运行产生影响,因此提出了一种调度策略以提高微电网的稳定性,降低运行成本。首先,利用蒙特卡罗方法建立电动汽车无序充电负荷模型,通过电量电价弹性矩阵确定电价与电动汽车充电负荷之间的函数关系。然后,建立微电网的运行成本、维护成本和污染治理成本最低的优化目标函数,结合可再生能源和负荷情况,分别使用双Q学习算法和Q学习算法设计调度策略,得到优化后的实时电价和调度方案。仿真结果表明,相比于Q学习算法,双Q学习算法在工作日和节假日情景下都得到了更加经济稳定的调度策略,且降低了微电网的经济成本。

The randomness on both sides of supply and demand affects the stable operation of the microgrid.Therefore,a dispatching strategy is proposed to improve the stability of microgrid and reduce the operation cost.Firstly,the disordered charging load model of electric vehicles is established by Monte Carlo method,and the functional relationship between electricity price and charging load of electric vehicles is determined by price elasticity matrix of demand.Then,the optimization objective function is established to minimize the operation cost,maintenance cost and pollution control cost of the microgrid.Combined with the renewable energy and load situation,the double Q-learning algorithm and Q-learning algorithm are used to design the dispatching strategy respectively,and the optimized real-time electricity price and dispatching schemes are obtained.By compared with the Q-learning algorithm,the simulation results show that the double Q-learning algorithm obtains a more economical and stable dispatching strategy on both working days and holidays,and reduces the economic cost of microgrid.