基于量子遗传算法的大规模风光并网发电系统储能优化方法  

Energy storage optimization method for large-scale wind and solar grid connected power generation system based on quantum genetic algorithm

作　　者：张艳 宋新甫 李香平 辛超山 张艳来 ZHANG Yan;SONG Xinfu;LI Xiangping;XIN Chaoshan;ZHANG Yanlai(State Grid Xinjiang Electric Power Co.,Ltd.,Urumqi 830063,China;State Grid Xinjiang Electric Power Co.,Ltd.,Economic and Technical Research Institute,Urumqi 830002,China;Tianjin Tianda Qiushi Electric Power New Technology Co.,Ltd.,Tianjin 300392,China)

机构地区：[1]国网新疆电力有限公司,乌鲁木齐830063 [2]国网新疆电力有限公司经济技术研究院,乌鲁木齐830002 [3]天津天大求实电力新技术股份有限公司,天津300392

出　　处：《自动化与仪器仪表》2024年第8期207-211,218,共6页Automation & Instrumentation

基　　金：新疆维吾尔自治区创新科学技术厅项目,源网荷储场景中风、光、储、荷容量最优配比关键技术研发(2022B01016-9)。

摘　　要：大规模风光并网发电具有明显的间歇性、随机性,并且风能光能本身的发电特征又存在较大差异,配置储能是解决上述问题的有效方法。将不确定的风光出力场景转化为确定的、具有其典型特征的小规模场景集,基于该场景集为风光联合发电系统配置储能,平抑新能源出力波动,提出一种基于量子遗传算法的大规模风光并网发电系统储能优化方法。在双向DC/DC变换器将储能系统与风光互补发电系统对接下,以蓄电池与对接系统后最充放电次数为目标函数生成稳定场景;以并网发电与储能的输出功率不允许超出最大功率为约束条件;采用量子遗传算法采集蓄电池荷电状态的反馈,通过量子位概率幅值对染色体编码,借助非门变异增加优化参数数量,利用旋转门法更新量子的相位,结合梯度函数、一阶差分找出稳定条件下的蓄电池输入功率,,完成稳定的储能容量调节。实验结果表明:所提方法的功率偏差值小,且能量缺失率也较低,故储能优化效果良好。Large scale wind and solar grid connected power generation has obvious intermittency and randomness,and there are significant differences in the power generation characteristics of wind and solar energy itself.Configuring energy storage is an effective method to solve the above problems.Transforming uncertain wind and solar output scenarios into a small-scale scenario set with typical characteristics,and based on this scenario set,configuring energy storage for wind and solar combined power generation systems to suppress fluctuations in new energy output,a quantum genetic algorithm based energy storage optimization method for large-scale wind and solar grid connected power generation systems is proposed.Under the docking of the energy storage system with the wind solar complementary power generation system using a bidirectional DC/DC converter,a stable scenario is generated with the objective function of the maximum number of charges and discharges after the battery is connected to the docking system;The output power of grid connected power generation and energy storage shall not exceed the maximum power as a constraint condition;Using quantum genetic algorithm to collect feedback on the state of charge of the battery,encoding chromosomes through the probability amplitude of quantum bits,increasing the number of optimization parameters through non gate mutation,updating the quantum phase using the rotation gate method,and combining gradient function and first-order difference to find the stable input power of the battery under stable conditions,achieving stable energy storage capacity regulation.The experimental results show that the power deviation value of the proposed method is small,and the energy loss rate is also low,so the energy storage optimization effect is good.

关 键 词：风光并网 功率 储能容量 蓄电池充放电 染色体 量子遗传算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]