考虑风电功率条件相关性的广义椭球不确定集合建模

发布时间：2018-05-27 13:32

本文选题：风电功率 + normal　；参考：《中国电机工程学报》2017年09期

【摘要】：近年来,鲁棒优化在电力系统优化调度领域取得了优良的效果。鲁棒优化的一个关键问题是确定风电功率不确定集合,而现有方法难以依据最新风电功率预测信息更新不确定集合。针对这一问题,该文基于条件copula copula模型,构建考虑风电功率条件相关性的广义椭球不确定集合。该方法综合考虑了预测误差的时间相关性,非正态的边缘分布特性,以及与风电功率预测值的条件相关性。此外,提出了条件normal copula模型的解析表达式以及相应采样方法。该方法实现了依据最新预测信息更新不确定集合,提高了模型的准确性,降低了集合的保守度。对某风场历史数据进行仿真分析,结果验证了所提算法的有效性,以及预测误差统计特性对不确定集合拟合效果的影响。
[Abstract]:In recent years, robust optimization has achieved good results in the field of power system optimization. One of the key problems of robust optimization is to determine the uncertain set of wind power, but it is difficult for existing methods to update the uncertain set based on the latest wind power prediction information. In order to solve this problem, a generalized ellipsoidal uncertain set considering the dependence of wind power condition is constructed based on conditional copula model. This method takes into account the temporal correlation of prediction error, the edge distribution characteristics of non-normal distribution and the conditional correlation with the predicted wind power. In addition, the analytical expression of conditional normal copula model and the corresponding sampling method are proposed. This method can update the uncertain set according to the latest prediction information, improve the accuracy of the model and reduce the conservatism of the set. The simulation analysis of the historical data of a certain wind field shows that the proposed algorithm is effective and the statistical characteristics of prediction errors affect the fitting effect of uncertain sets.
【作者单位】：上海交通大学电气工程系;辽宁省电力有限公司;
【基金】：国家科技支撑计划(2015BAA01B02) 国家自然科学基金(51307107,51477098)~~
【分类号】：TM614;TM73

【相似文献】