基于时序-支持向量机的风电场发电功率预测

准确的风电场风电功率预测可以有效地减轻风电场对电力系统的不利影响,同时提高风电在电力市场中的竞争力。基于时间序列法和支持向量机法, 对风电功率预测进行研究,提出预测风电功率的时序-支持向量机预测方法。该方法用时间序列法建模,选取影响风电功率最大的参数作为支持向量机预测模型的输入变量;为提高预测精度,提出基于时间点运动轨迹演化的方法选取与预测时刻功率相似的样本作为模型的训练样本。实例验证结果表明,该方法有效地提高了风电功率预测精度。

Wind power forecasting based on time series and SVM

The accurate wind power forecasting can relieve the adverse effects of wind power plants on power systems and enhance the competitive ability of wind power plants in electricity markets.A wind power forecasting method is proposed based on time series method and support vector machine(SVM).The mathematical model was built by time series method.The factors which have significant impacts on the wind power were selected as SVM’s inputs.In order to improve forecasting accuracy,a method based on time series trace evolution was used to find SVM training samples which similar to the power at the forecasting point.The actual examples prove the improvement of wind power forecasting accuracy by using the time series-SVM method.