CAJ | 학술논문

大规模的电动汽车(plug-in electric vehicle,PEV)和风力、太阳能等可再生能源(renewable energy sources,RES)发电并网使未来智能配电网规划需考虑更多不确定因素。在考虑PEV充电随机性和RES出力间歇性的基础上,利用机会约束规划法建立了计及环境成本、DG总费用和有功损耗的多目标分布式电源优化配置模型,并提出一种考虑随机变量相关性的拉丁超立方采样蒙特卡洛模拟嵌入纵横交叉算法(crisscross optimization algorithm- correlation Latin hypercube sampling Monte Carlo simulation,CSO-CLMCS)的方法对优化模型进行求解。该方法首先根据PEV和RES的概率模型及随机变量间的相关性,利用CLMCS概率潮流计算方法计算配电网概率潮流,并根据概率潮流结果检验约束条件及计算目标函数值,最后由CSO算法进行全局寻优得到最优配置方案。采用实际算例进行仿真,结果验证了所提模型和方法的可行性和有效性。
대규모적전동기차(plug-in electric vehicle,PEV)화풍력、태양능등가재생능원(renewable energy sources,RES)발전병망사미래지능배전망규화수고필경다불학정인소。재고필PEV충전수궤성화RES출력간헐성적기출상,이용궤회약속규화법건립료계급배경성본、DG총비용화유공손모적다목표분포식전원우화배치모형,병제출일충고필수궤변량상관성적랍정초립방채양몽특잡락모의감입종횡교차산법(crisscross optimization algorithm- correlation Latin hypercube sampling Monte Carlo simulation,CSO-CLMCS)적방법대우화모형진행구해。해방법수선근거PEV화RES적개솔모형급수궤변량간적상관성,이용CLMCS개솔조류계산방법계산배전망개솔조류,병근거개솔조류결과검험약속조건급계산목표함수치,최후유CSO산법진행전국심우득도최우배치방안。채용실제산례진행방진,결과험증료소제모형화방법적가행성화유효성。
As large-scale of plug-in electric vehicles (PEV) and renewable energy sources (RES) such as wind and solar energy integrated into power grid, the smart distribution network planning inevitably needs to consider more uncertainties. Given this background, under the chance constrained programming framework, a multi-objective optimal distributed generation planning model was established considering the randomness and correlation of RES generation and PEV, in which the environmental cost, DG’s total cost and network loss were taken as the objectives. A correlation Latin hypercube sampling Monte Carlo simulation (CLMCS) embedded crisscross optimization algorithm (CSO)-based approach (CSO-CLMCS) was proposed to solve the optimization model. The method firstly used CLMCS to calculate the probabilistic load flow (PLF) according to the probability model of random variables. Then the constraint conditions were checked and the objective function value was obtained using the result of PLF. Finally, the optimal planning scheme was obtained by a new smart search algorithm, i.e. CSO. The simulation was conducted on an actual distribution network. The results validate the feasibility and effectiveness of the proposed model and method.