• CSCD核心库收录期刊
  • 中文核心期刊
  • 中国科技核心期刊

电力建设 ›› 2020, Vol. 41 ›› Issue (7): 117-128.doi: 10.12204/j.issn.1000-7229.2020.07.015

• 新能源发电 • 上一篇    

计及风-光-荷相关性的概率-区间约束潮流模型与算法

郇嘉嘉1,洪海峰1,隋宇1,余梦泽1,潘险险1,汪超群2   

  1. 1. 广东电网有限责任公司电网规划研究中心,广州市 510080;2. 浙江大学电气工程学院,杭州市 310007
  • 出版日期:2020-07-01
  • 作者简介:郇嘉嘉(1983),女,博士,高级工程师,研究方向为综合能源系统规划; 洪海峰(1992),男,硕士,高级工程师,研究方向为综合能源系统规划; 隋宇(1983),男,高级工程师,研究方向为电力系统通信; 余梦泽(1980),男,博士,高级工程师,研究方向为电力系统规划; 潘险险(1989),女,硕士,高级工程师,研究方向为综合能源系统规划; 汪超群(1990),男,博士,高级工程师,研究方向为电力系统最优运行与规划。
  • 基金资助:
    中国南方电网公司科技项目(GDKJXM20184328)

Model and Algorithm of Probabilistic Interval Constrained Power Flow Considering the Correlation Among Wind, Solar and Load

HUAN Jiajia1, HONG Haifeng1, SUI Yu1, YU Mengze1,PAN Xianxian1,WANG Chaoqun2   

  1. 1. Grid Planning & Research Center, Guangdong Power Grid Co., Ltd., Guangzhou 510080, China; 2. College of Electrical Engineering, Zhejiang University, Hangzhou 310007, China
  • Online:2020-07-01
  • Supported by:
    This work is supported by China Southern Power Grid Corporation Research Program(No.GDKJXM20184328).

摘要: 提出一种考虑风-光-荷随机性和相关性的概率-区间混合不确定约束潮流模型,并采用基于仿射空间变换的蒙特卡罗法加以求解。首先,基于不确定变量历史统计数据是否充裕2种情形,分别采用概率变量和区间变量描述其随机变化特征;然后,利用Box-Cox正态变换和区间变换对服从不同分布的随机变量进行标准化处理,同时将样本相关系数推广至含概率和区间2种随机变量的不确定分析当中,建立多元(概率-区间)随机变量的样本相关系数矩阵;基于该矩阵,通过Cholesky分解建立仿射空间坐标系,将具有相关性的概率和区间变量映射为仿射随机空间内线性无关的独立变量;最后,由蒙特卡罗模拟法获得系统待求变量的概率和区间分布。IEEE-57和118节点系统的计算结果表明,概率-区间约束潮流模型能定量分析不确定性和相关性对电力系统潮流的影响,有效实现了概率潮流和区间潮流的联合统一,对于指导系统规划运行具有一定的参考意义。

关键词: 概率区间潮流, 相关性, Box-Cox变换, 仿射变换, 蒙特卡罗法

Abstract: A probabilistic interval power flow (PIPF) model considering the randomness and correlation of wind-solar-load is proposed, and it is solved by Monte Carlo (MC) method based on affine space transformation. Firstly, on the basis of two cases of whether the historical statistical data of uncertain variables are abundant, probabilistic variables and interval variables are used to describe their random change characteristics. Then, Box-Cox normal transformation and interval transformation are applied to normalize random variables subject to different distributions. Moreover, the sample correlation coefficient is extended to the uncertainty analysis that includes probability and interval random variables, and a correlation coefficient matrix of multivariate random variables is established. On the basis of the matrix, an affine space coordinate system is established by Cholesky decomposition, and the probabilistic and interval variables with correlation are mapped to independent random variables in the affine space. Finally, the interval and probability distribution of the variables to be calculated in PIPF are obtained by MC method. The calculation results of the IEEE 57-node and 118-node systems show that PIPF can quantitatively analyze the influence of uncertainty and correlation on power system, effectively realize the unity of probabilistic power flow and interval power flow, and have certain reference significance for guiding system planning and operation.

Key words: probabilistic interval power flow, correlation, Box-Cox transformation, affine transformation, Monte Carlo method

中图分类号: