Fast Calculation Method of Probabilistic Optimal Power Flow for Renewable Dominated Power Grid Based on Improved Convex Relaxation
崔伟 1柴龙越 2王聪 1王炜 1汪莹 1杨仑2
扫码查看
点击上方二维码区域，可以放大扫码查看
作者信息
1. 国网电网有限公司西北分部,陕西西安 710000
2. 西安交通大学自动化科学与工程学院,陕西西安 710049
折叠
摘要
现有概率最优潮流计算侧重于概率计算方法的设计和改进,难以从本质上提高概率最优潮流的计算效率.为此,以交直流新能源电网为研究对象,考虑风电、光伏发电的不确定性,建立交直流互联新能源电网概率最优潮流模型.首先,提出一种改进凸松弛技术处理非线性非凸潮流方法,将其转化为凸规划形式下的概率最优潮流模型;其次,利用Nataf变换处理非正态分布随机变量间的相关性,进而采用结合拉丁超立方采样技术的蒙特卡罗模拟法(monte carlo simulation,MCS)进行求解以降低MCS的计算量;最后,通过改进的IEEE 39 节点、118节点以及 500节点系统验证所提方法的有效性.
Abstract
The existing probabilistic optimal power flow(POPF)studies mainly focus on the design and improvement on probabilistic calculation methods,which may be difficult to improve the computational efficiency of POPF as POPF is a nonconvex and nonlinear programming problem under uncertainty.Therefore,this paper centers on renewable-dominated AC-DC power grid and proposes a POPF model considering uncertainties associated with wind and solar power.To efficiently solve the POPF model,an improved convex relaxation is proposed to address the nonconvex and nonlinear power flow equations and reformulate the nonlinear POPF model as convex one.Furthermore,the Nataf transformation is adopted to address the correlations of non-normal distribution and then a Monte Carlo Simulation based Latin Hypercube sampling technique is developed to solve the convex POPF model.Finally,the effectiveness of the proposed improved convex relaxation based POPF method is demonstrated by a set of case results tested on the modified IEEE 39-bus,118-bus,and 500-bus systems.
关键词
概率最优潮流/凸松弛/不确定性/拉丁超立方
Key words
probabilistic optimal power flow/convex relaxation/uncertainty/Latin hypercube Sampling
维普
万方数据