计及样本容量合理性的风电功率预测考核算法

发布时间：2018-09-01 08:36

【摘要】：在风功率预测误差服从正态分布的基础上,首先,提出了在给定置信水平、最大允许误差下最小样本容量的计算方法,以及在给定误差范围、样本容量下估计值置信水平的计算方法。其次,针对典型显著性水平、误差范围进行分析,论证了当样本容量较小时,样本统计指标置信水平低,无法代表总体指标。最后,提出了计及样本容量合理性的日前风电功率预测准确率考核算法,实际算例表明,所提考核算法更科学合理,避免了不合理的考核罚款。
[Abstract]:On the basis of the wind power prediction error obeying normal distribution, the calculation method of minimum sample size under given confidence level and maximum allowable error and the calculation method of confidence level under given error range and sample size are proposed. Secondly, the typical significance level and error range are analyzed and proved. When the sample size is small, the confidence level of the statistical indicators is low, which can not represent the overall indicators. Finally, an assessment algorithm for the accuracy of day-ahead wind power forecasting considering the rationality of the sample size is proposed. The practical example shows that the proposed assessment algorithm is more scientific and reasonable, and the unreasonable assessment penalty is avoided.
【作者单位】：南京南瑞继保电气有限公司;
【分类号】：TM614

【相似文献】