考虑电动汽车灵活性与风电消纳的区域电网多时间尺度调度

Multi-time-scale Scheduling for Regional Power Grid Considering Flexibility of Electric Vehicle and Wind Power Accommodation

随着电动汽车数量逐渐增多,电动汽车接入电网会给电网运行控制带来困难。同时,以新能源为主体的新型电力系统面临功率平衡挑战。基于自适应噪声完全集成的经验模态分解(CEEMDAN)和双向长短期记忆神经网络(BILSTM),提出了一种考虑电动汽车灵活性与风电消纳的区域电网多时间尺度调度方法。首先,通过对风电和基础负荷历史数据进行CEEMDAN,得到不同频率下的本征模态分量;其次,根据极大值点个数判据对本征模态分量进行重构;再次,利用BILSTM预测重构分量,得到风电和基础负荷数据的预测值。基于上述预测的数据,采用模型预测控制方法,建立区域电网多时间尺度调度模型。最后,通过仿真结果表明,所提预测方法具有普适性;所提的多时间尺度调度方法有效且经济,该方法可以在平抑负荷波动、降低风电并网影响的同时,利用电动汽车灵活性对风电预测实时偏差进行补偿,维持系统平衡。

As the number of electric vehicles(EVs)gradually increases,the integration of EVs to power grids brings the difficulties to the operation and control of power grids.At the same time,the new power system with renewable energy as the main body faces the challenges of power balance.Based on complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)and bi-directional long short-term memory(Bi LSTM)neural network,a multi-time-scale scheduling method for regional power grid considering EV flexibility and wind power accommodation is proposed.Firstly,the intrinsic mode functions at different frequencies are obtained by CEEMMDAN of historical wind power and load data.Secondly,the intrinsic mode functions are reconstructed according to the criterion of the number of maximum value points.Thirdly,Bi LSTM is used to predict the reconstructed components to obtain the predicted data of wind power and load data.Based on the predicted data,the multi-timescale scheduling model for the regional power grid is established based on the model predictive control(MPC)method.Finally,the simulation results show that the proposed prediction method has universal applicability.The proposed multi-time-scale scheduling method is effective and economic,which can not only suppress the load fluctuation and reduce the influence of wind power gridconnection,but also use the flexibility of EVs to perform the real-time deviation compensation of wind power prediction,and to maintain the system balance.