CAJ | 학술논문

随着分布式电源(DG)和电动汽车的大量发展，对接入配电网的电动汽车与 DG 进行协同研究具有重要意义。文中以协调配电公司、DG 投资商和公共社会三者之间利益为出发点，综合考虑了配电公司的运行费用、DG 投资商的投资费用，以及 DG 的环境效益和电动汽车入网(V2G)所节省的电网投资等社会效益，建立了基于机会约束规划的含 V2G 配电网中 DG 优化规划的数学模型，采用基于混合编码的改进自适应遗传算法对该模型进行了求解。在优化计算过程中充分考虑了负荷预测值的不确定性、风电源的输出功率的随机性以及电动汽车充放电功率的不确定性，提出了电动汽车充放电对系统最大功率影响的数学模型，并采用基于半不变量法的随机潮流算法对规划模型中的约束条件进行了检验。最后，以某实际配电网系统为仿真算例，在不同置信水平约束下对该系统内 DG 分别进行了优化规划，验证了文中所建数学模型及相应求解算法的有效性。
수착분포식전원(DG)화전동기차적대량발전，대접입배전망적전동기차여 DG 진행협동연구구유중요의의。문중이협조배전공사、DG 투자상화공공사회삼자지간이익위출발점，종합고필료배전공사적운행비용、DG 투자상적투자비용，이급 DG 적배경효익화전동기차입망(V2G)소절성적전망투자등사회효익，건립료기우궤회약속규화적함 V2G 배전망중 DG 우화규화적수학모형，채용기우혼합편마적개진자괄응유전산법대해모형진행료구해。재우화계산과정중충분고필료부하예측치적불학정성、풍전원적수출공솔적수궤성이급전동기차충방전공솔적불학정성，제출료전동기차충방전대계통최대공솔영향적수학모형，병채용기우반불변량법적수궤조류산법대규화모형중적약속조건진행료검험。최후，이모실제배전망계통위방진산례，재불동치신수평약속하대해계통내 DG 분별진행료우화규화，험증료문중소건수학모형급상응구해산법적유효성。
With the great development of distributed generator (DG) and electric vehicle (EV),it is of great significance to carry out collaborative research of DG and EV accessed to distribution network.Based on the coordination of benefits between distribution companies,DG investors and social communities,this paper comprehensively takes into consideration the operating cost of distribution companies,investment cost of DG investors,and environmental benefit of DG and grid investment saved by vehicle to grid(V2G).A mathematical model for DG optimal planning in distribution network involving EV is developed based on chance constrained programming,with a hybrid coding based improved adaptive genetic algorithm to solve the model.In the calculation process,uncertainties of load forecast,output power of DG and charge or discharge power of EV are fully considered.And the mathematical model of EV influence on the grid maximum power is proposed.The constraints in the model are checked by probabilistic power flow based on semi-invariant and Gram-Charlier expansion theory.Finally,DG planning under different confidence level constraints is respectively computed within an actual distribution network to verify the effectiveness of the mathematical model and corresponding algorithm proposed.It can be concluded from the simulation results that with the reduction of confidence level in the mathematical model,the DG planning capacity in the system has significantly increased,so that the system loss is reduced and environmental benefit is increased.However,the risk of node voltage and branch power constraint violation has also increased,which leads to making higher requirements on real-time control research on the distribution network in the future. This work is supported by Shanghai Green Energy Grid Connected Technology Engineering Research Center (No.13DZ225 1 900).