【作者】 张丽丽；

【导师】 王克文；

【作者基本信息】 郑州大学 ， 电气工程（专业学位）， 2019， 硕士

【摘要】 近年来,能源短缺和环境污染问题日益严重,风能等清洁能源受到了人们的关注,风电并网规模不断增大。由于风电存在随机性与不确定性的特点,风电并网直接影响电网运行的安全性和稳定性。准确的风电功率预测能够为电网运行提供数据参考,有利于电网的分配与调度。本文以风电功率预测为研究目标,提出一种基于K-means聚类—层次分析的风电场短期功率预测方法。首先,研究风力发电的发电模型、风资源特性和风电场出力概率分布。其次,建立基于K-means聚类-层次分析的风电场短期风速预测模型。采用K-means聚类算法,对研究区域内的风电场进行分群,选取地理位置信息作为分类指标,以研究区域的中心作为原点,建立直角坐标系,得到风电场的地理位置坐标进行聚类分析。分析不同风电场之间的风速相关性,考虑海拔、地形、温度、湿度等因素对风速的影响,采用层次分析法分析各个影响因素的权重,建立相应的层次分析模型。再次,以某一风电场的风速信息和气象、地理信息作为基础,采用K-means聚类-层次分析的风电场短期风速预测模型预测该区域内其他相关风电场的风速,并与风电场的实际风速情况进行对比验证。依据所预测的风速结果结合风电机组的装机容量预测风电场的功率。最后,通过算例分析预测结果的误差,验证算法的有效性。更多还原

【Abstract】 In recent years,due to the increasingly serious energy shortage and environmental pollution,people are paying more attention to clean energies such as wind and the scale of wind power integration grows a lot.But because of the characteristics of wind power like randomness and uncertainty,wind power integration directly influences the safety and stability of power grid operation.Accurate prediction of wind power can provide a reference for the data of grid operation and is beneficial to the allocation and dispatch of power grid.Therefore,this thesis,targeted at wind power prediction,proposes a wind farm short-term power prediction method based on K-means clustering-hierarchy analysis.First,study the wind power generator model,wind resource characteristics and wind farm output probability distribution.Then,establish the wind farm short-term power prediction model based on K-means clustering-hierarchy analysis.Implement clustering for the wind farm in the research area by means of K-means clustering algorithm and establish rectangular coordinate system with geographic position information as classification indicator and the center of the research area as origin,thereby obtaining the geographic position coordinate of the wind farm and conducting clustering analysis.Analyze the wind speed relevance among different wind farms and take the influences of factors such as altitude,topography,temperature,humidity,etc.on wind speed into consideration.Adopt analytic hierarchy process to analyze the weights of all influence factors and establish correspondent analytic hierarchy model.Moreover,on the basis of the wind speed and weather and geographic information of a certain wind farm,use wind farm short-term speed prediction model based on K-means clustering-hierarchy analysis to forecast the wind speed of other related wind farms in this area and validate and compare it with the actual wind speed of the wind farm.According to the predicted wind speed,forecast the power of the wind farm in combination with the installed capacity of the wind turbine generators.At last,analyze the error of the prediction results through examples and verify the effectiveness of the algorithm.更多还原