基于全概率公式的含风电配电系统概率潮流计算

发布时间：2018-05-06 11:35

本文选题：分布式风电 + 配电系统　；参考：《中国电机工程学报》2015年02期

【摘要】：针对风电出力不满足正态分布的特点,对风电出力分段离散化,并将多个风电场的离散化结果进行组合得到系统风电出力状态的所有组合,将问题转换为求解风电注入功率确定的多个条件概率分布问题。利用全概率公式整合多个条件事件的原点矩以计算系统状态变量的期望和方差。基于所有条件概率分布满足正态分布的特点,将所得的多个Gauss函数按全概率公式累加的方法得到同时考虑风电随机性和负荷随机性的概率密度函数,摒弃了传统的Gram-Charlier级数拟合概率密度函数的方法。以IEEE 33节点配电网络接入多个风电场为例对所提方法进行验证,并与多个抽样规模的蒙特卡罗模拟法进行对比,结果证明所提方法具有与千万次抽样的蒙特卡罗模拟法等同的精度。
[Abstract]:In view of the fact that the wind power output does not satisfy the normal distribution, the wind power output is discretized by stages, and the discrete results of multiple wind farms are combined to obtain all the combinations of the wind power output states of the system. The problem is transformed into the problem of multiple conditional probability distributions for determining the power of wind power injection. The full probability formula is used to integrate the origin moments of multiple conditional events to calculate the expectation and variance of system state variables. Based on the fact that all conditional probability distributions satisfy the normal distribution, the probability density function which considers both the randomness of wind power and the randomness of load is obtained by summing up multiple Gauss functions according to the full probability formula. The traditional method of Gram-Charlier series fitting probability density function is abandoned. Taking the IEEE 33 node distribution network access to multiple wind farms as an example, the proposed method is verified and compared with the Monte Carlo simulation method with multiple sampling scales. The results show that the proposed method has the same accuracy as the Monte Carlo simulation method.
【作者单位】：南昌大学电气与自动化系;
【基金】：国家自然科学基金资助项目(51167012) 江西省研究生创新专项资金资助项目(YC2013-S056)~~
【分类号】：TM744

【参考文献】