多风电场短期输出功率的联合概率密度预测方法

发布时间：2018-05-28 10:03

本文选题：短期风电功率预测 + 联合概率密度预测　；参考：《电力系统自动化》2014年19期

【摘要】：提出一种多风电场短期输出功率的联合概率密度预测方法。首先利用支持向量机对每座风电场的输出功率进行单点值预测,对预测误差建立稀疏贝叶斯学习模型进行误差的概率密度预测,得到单一风电场输出功率的边际概率密度函数预测结果;对多风场输出功率预测误差特性进行统计分析,发现同一区域内,风电场输出功率预测误差之间存在线性时空关联特性,进而运用动态条件相关回归模型求得相关系数矩阵,定量描述多风电场短期输出功率预测误差之间的动态时空相关关系;最后,综合单一风电场输出功率边际概率密度预测结果和相关系数矩阵得到多风电场输出功率的联合概率密度函数,并借助多元随机变量抽样技术形成包含动态时空关联特性的多维场景。通过实例分析,表明了所提出方法的有效性。
[Abstract]:A joint probability density prediction method for short-term output power of multi-wind farms is proposed. Firstly, support vector machine is used to predict the output power of each wind farm, and the probability density of the error is predicted by establishing a sparse Bayesian learning model. The prediction results of the marginal probability density function of the output power of a single wind farm are obtained, and the prediction error characteristics of the output power of the multi-wind field are statistically analyzed, and it is found that in the same region, The linear time-space correlation exists between the prediction errors of wind farm output power, and then the correlation coefficient matrix is obtained by using the dynamic conditional correlation regression model, and the dynamic space-time correlation relationship between the prediction errors of the short-term output power of multi-wind farm is quantitatively described. Finally, combining the prediction results of the marginal probability density of the output power of a single wind farm and the correlation coefficient matrix, the joint probability density function of the output power of the multi-wind farm is obtained. The multi-dimensional scene with dynamic time-space connection is formed by multi-variable sampling technique. The effectiveness of the proposed method is demonstrated by an example.
【作者单位】：电网智能化调度与控制教育部重点实验室(山东大学);国网山东省电力公司电力科学研究院;
【基金】：国家重点基础研究发展计划(973计划)资助项目(2013CB228205) 国家自然科学基金资助项目(51007047,51077087) 山东省自然科学基金资助项目(ZR2010EQ035)~~
【分类号】：TM614

【相似文献】