风电外送断面极限输电能力的非参数回归估计

发布时间：2018-03-29 09:46

本文选题：风电　切入点：极限输电能力　出处：《电网技术》2017年11期

【摘要】：大规模风电的随机性和间歇性导致基于典型方式计算的通道极限输电能力(total transfer capability,TTC)有效性降低。提出一种TTC非参数回归估计技术,通过风电与负荷场景聚类形成代表性中心,采用二分法重复潮流计算各场景下含稳定约束的断面TTC值,提取各场景与所属中心场景间的属性偏差及TTC偏差作为特征数据样本,经过相关性检验与非参独立筛选后,利用基于三次B样条函数展开的Group Lasso算法对TTC偏差进行非参数回归估计。算例验证表明,该方法具备较强的非线性泛化能力,能以较高精度提取输电断面TTC运行规则的显性表达式,与传统方法相比具有更丰富的信息输出与更良好的解释性,可用于含风电外送断面电力系统TTC的在线快速估计。
[Abstract]:The randomness and intermittency of large-scale wind power result in the reduction of the efficiency of total transfer capabilityTTC based on typical calculation of channel limit transmission capacity. A non-parametric TTC regression estimation technique is proposed, which forms the representative center by clustering wind power with load scenes. The cross-section TTC value with stable constraints is calculated by using dichotomous repeated power flow method, and the attribute deviation and TTC deviation between each scene and its central scene are extracted as feature data samples. After correlation test and non-parametric independent screening, the cross-section TTC value is calculated. The Group Lasso algorithm based on cubic B-spline function expansion is used to estimate the nonparametric TTC deviation. An example shows that the method has strong nonlinear generalization ability. The explicit expression of TTC operating rules of transmission section can be extracted with high accuracy. Compared with the traditional method, it has more abundant information output and better explanation. It can be used to estimate the TTC of power system with wind power outputting section on line.
【作者单位】：四川大学电气信息学院;国网四川省电力公司成都供电公司;西南财经大学统计学院统计研究中心;
【分类号】：TM614

【相似文献】