风电功率模拟生成模型研究

摘要

Wind energy expansion is one of the prominent ongoing evolutions of power systems in recent years. Normally power system operators carry out long-term planning studies for wind power with the previous statistic data. The study presented in this paper focuses on the artificially built wind power curves simulating the actual behavior of wind farms. The core of the studies is how to sample wind power curves via stochastic process. Firstly,the initial sampling model for power curves is built and then gradually revised considering seasonal and diurnal changes. Secondly,a stochastic differential equation fitting the model is built,and the boundary conditions are set up in view of distribution functions as well as auto correlation functions. Thirdly,the form of stochastic differential equation is simplified by means of designing algorithm simulating Brownian movement component included in the SDE. Finally, both China’s and Spain’s data are modeled to verify the feasibility and rationality of the model mainly through simulating monthly and yearly power level expectation,standard deviation as well as monthly and yearly probability density distribution.%近年来，电力系统进行中的显著变革之一是其对风能的扩大利用。系统运营者在以往的统计数据基础上对风电功率进行长期的规划研究。该研究的主要内容是人工建立风电曲线来模拟风电厂的实际运转状态。文中研究的核心问题是如何通过随机过程模拟生成风电功率曲线，首先建立风电功率模拟生成模型，再考虑到季节性和一天内变动等实际问题对模型进行修正。其次，建立适合模型的随机微分方程，并结合分布函数和自相关函数设置边界条件。再次，通过设计算法模拟随机微分方程中包含的布朗运动分量来简化方程形式，并利用Matlab软件进行求解。最后以中国和西班牙两个国家的数据做了算例分析，主要从各月和全年风电功率平均值和方差，以及各月和全年概率密度分布等方面来验证模型的可行性和解法的合理性。