中国电机工程学报2017，Vol.37Issue(9)：2500-2506,7.DOI:10.13334/j.0258-8013.pcsee.160389
考虑风电功率条件相关性的广义椭球不确定集合建模
Modeling Ellipsoidal Uncertainty Set Considering Conditional Correlation of Wind Power Generation
摘要
Abstract
Robust optimization has shown appealing performance in power system dispatch in recent years. One of the critical problems in robust optimization is to determine the uncertainty set of the wind power. However, existing methods have difficulties in updating the uncertainty sets according to upcoming day-ahead wind power forecast. In view of such difficulties, this paper proposed the ellipsoidal uncertainty set for conditional correlated wind power generation based on conditional normal copula model. The proposed model takes into consideration the temporal dependence and non-Gaussian marginal distribution of wind power forecast error as well as the conditional dependence between the wind power forecast and forecast error. This paper presented the analytical form and effective sampling method of the conditional normal copula. The proposed method updates the uncertainty set according to the day-ahead wind power forecast. As a result, it improves the goodness of fitness and reduces the degree of conservatism of the uncertainty set. The effectiveness of the proposed method is validated based on the historical data from a wind farm cluster. The simulation results also demonstrate the effect of the statistic characteristics of the wind power forecast error on the uncertainty set.关键词
风电功率/normalcopula/条件分布/采样/自适应更新/不确定集合Key words
wind power/normal copula/conditional distribution/sampling/adaptive updating/uncertainty sets分类
信息技术与安全科学引用本文复制引用
吴巍,汪可友,李国杰,葛延峰..考虑风电功率条件相关性的广义椭球不确定集合建模[J].中国电机工程学报,2017,37(9):2500-2506,7.基金项目
国家科技支撑计划(2015BAA01B02) （2015BAA01B02）
国家自然科学基金(51307107,51477098). National Key Technology R&D Program of China (2015BAA01 B02) （51307107,51477098）
Project Supported by National Natural Science Foundation of China(51307107,51477098). （51307107,51477098）
