风电场短期风速的改进Volterra自适应预测法

Short-term Wind Speed Forecasting in Wind Farms Using Volterra Adaptive Filter

欲提高含大量风电机组电网的安全稳定性及降低其运行成本,高精度的短期风速预测是一种有效的手段。首先在对具有混沌属性的风速时间序列进行相空间重构的基础之上,使用一种混沌时间序列的Volterra自适应滤波预测法对风速进行了预测;然后针对该方法滤波系数不易收敛及预测结果存在时延的缺点,改进了自适应算法的系数更新方法,从而加快了收敛并提高了预测精度。算例分析结果验证了该方法的可行性和有效性。

To ensure the security and stability, and also to reduce the operation cost of a power grid which has a great portion of wind power, a precise forecasting of the short -term wind speed would be an effective method. Firstly, the phase space re- construction is carried out for the wind speed time series, and then a Voltcrra adaptive forecasting method is used to forecast the short - term wind speed. Aiming at the shortages of this method that it is not easy to converge and there exists delay in the forecasting result, a modification in the updating method of the filtering coefficients is proposed, by which the convergence speed of the training progress of the filtering coefficients is accelerated and the precision of the forecasting result is improved. Finally, the forecasting result of an example case shows the feasibility and validity of the proposed forecasting method.