基于最优概率分布模型的风能资源评估  

Wind Energy Resource Assessment Based on Optimal Probability Distribution Models

作　　者：徐凯 熊国江 徐波 XU Kai;XIONG Guojiang;XU Bo(School of Electrical Engineering,Guizhou University,Guiyang 550025,Guizhou Province,China)

机构地区：[1]贵州大学电气工程学院,贵州省贵阳市550025

出　　处：《全球能源互联网》2025年第1期87-97,共11页Journal of Global Energy Interconnection

基　　金：国家自然科学基金(52367006);贵州省科技计划项目(黔科合基础-ZK[2022]一般121)。

摘　　要：风能资源的分布受地形和气候的影响,不同地域的风能资源差异很大,如何建立准确的风能资源评估模型甚为关键。为了克服单个拟合模型和参数估计方法适应性有限的问题,采用3种概率分布模型(威布尔分布、对数正态分布和布尔分布)对风速分布数据进行拟合。在此基础上,应用3种数值法(最大似然法、矩估计法和最小二乘法)和一种智能优化算法(鲸鱼优化算法)分别对3种概率分布模型的参数进行估计。通过对比分析确定不同地域的最优概率分布模型及其参数估计方法。最后基于最优概率分布模型及其参数计算了风能资源年可用时间、平均风功率密度和平均风速。以中国内蒙古、西藏、湖北、上海和新疆的5个站点为例,对50 m和70 m高度处的风速分布进行了仿真研究。结果表明,不仅不同地域的最优概率分布模型有所差异,而且相同站点不同高度处的最优分布模型也不同。在4个参数估计方法中,鲸鱼优化算法的参数辨识结果最为准确。多种模型和方法的组合有效地提高了风能资源评估的准确性。The distribution of wind energy resources is affected by topography and climate,and wind energy resources vary greatly in different regions,so it is crucial to establish an accurate wind energy resource assessment model.In order to overcome the problem of limited adaptability of individual fitting models and parameter estimation methods,three probability distribution models(Weibull distribution,lognormal distribution and Burr distribution)are used to fit the wind speed distribution data;on this basis,three numerical methods(maximum likelihood method,method of moments estimation,and method of least squares)and an intelligent optimization algorithm(whale optimization algorithm)are applied to estimate the parameters of the three probability distribution models respectively.The optimal probability distribution model and its parameter estimation method for different regions are determined through comparative analysis.Finally,based on the optimal model and its parameters,the annual availability time of wind energy resources,average wind power density,and average wind speed are calculated.The wind speed distributions at the heights of 50 and 70 metres were simulated at five stations in Inner Mongolia,Tibet,Hubei,Shanghai and Xinjiang,China,for example.The results show that not only the optimal probability distribution models differ in different geographical areas,but also the optimal distribution models at different heights of the same site are different.Among the four parameter estimation methods,the whale optimization algorithm has the most accurate parameter identification results.The combination of multiple models and methods effectively improves the accuracy of wind energy resource assessment.

关 键 词：风能资源评估 参数估计 数值法 智能优化算法 

分 类 号：TK81[动力工程及工程热物理—流体机械及工程]