基于稀疏矢量自回归概率模型的超短期风电功率预测算法

ULTRA SHORT TERM WIND POWER FORECASTING BASED ON THE SPARSE VECTOR AUTOREGRESSION

可再生能源的概率预测被广泛认为是电力系统优化的必要条件。提出一种在大量地点进行超短期的参数化风电概率预测的时空方法,其基于logit-normal的参数框架,将多个风电场的位置参数建模为一个向量值时空过程,并采用改进的指数平滑法跟踪尺度参数,采用一种先进的稀疏向量自回归模型拟合技术,对定位参数进行建模,并与传统的向量自回归模型相对比。以澳大利亚22个风电场的每5分钟平均风力发电数据集为例进行了测试,验证了该算法的有效性。

The probability prediction of renewable energy is widely considered as a necessary condition for power system optimization. A sptio-temporal method of ultra short term parametric wind power probabilistic forecasting in a large number of locations is proposed. Based on the parametric framwork of logit-normal, the location parameter for multiple wind farms was modeled as a vector-valued spatiotemporal process, and the scale parameter was tracked by modified exponential smoothing. A state-of-the-art technique for fitting sparse vector autoregressive models was employed to model the location parameter and demonstrate numerical advantages over conventional vector autoregressive models. The proposed method was tested on a dataset of 5 min mean wind power generation at 22 wind farms in Australia. The experimental results verify the effectiveness of this method.