高比例新能源接入下的历史策略库辅助源网荷储协同实时电压控制研究

Research on real-time voltage control of source network load storage collaboration assisted by historical strategy library under high ratio new energy access

随着“双碳”目标的推进,储能、充电桩等用户侧新型负荷设备数量增多。在此背景下,由于量测条件不满足造成的不可观测区域将直接导致传统的电压控制方法难以完成对分布式电源的精准调控。为解决上述问题,提出一种基于蜣螂优化算法(DBO)的极限学习机(ELM)构建历史策略库,用以辅助源网荷储协同实时电压控制的方法,可实现对配电网电压实时精确控制。首先介绍了基于近似灵敏度计算的电压控制方法,然后介绍了DBO改进的极限学习机和历史策略库的概念及结合应用方法,构建了以基于近似灵敏度计算的电网节点有功及无功功率为输入,母线期望电压为输出的ELM模型。模型输出的母线电压作为控制依据,进一步转换为下发的用户侧可调设备调节指令。仿真算例的结果验证了所提方法的有效性和优越性。

With the promotion of the"double carbon"goal,the number of new load equipment on the user side such as energy storage and charging piles has increased.In this context,the unobservable region caused by unsatisfied measurement conditions will directly make it difficult for traditional voltage control methods to achieve accurate regulation of distributed power supply.In order to solve the above problems,a method of constructing historical strategy library based on the improved Extreme Learning Machine(ELM)based on the Dung Beetle Optimizer(DBO)was proposed to assist the real-time voltage control of source network load storage coordination,which can realize the real-time and accurate voltage control of the distribution network.Firstly,a voltage control method based on approximate sensitivity calculation was introduced,then the concept and combined application method of DBO improved Extreme Learning Machine and historical strategy library was introduced,and an ELM model based on approximate sensitivity calculation of active and reactive power of grid nodes as input and expected voltage of bus as output was built.The bus voltage output of the model was used as the control basis and converted into the user-side adjustable device adjustment instruction issued.The results of simulation examples verified the effectiveness and superiority of the proposed method.