大型风电场超短期风电功率预测

针对大规模风电场风电功率的非线性特性，采用最小二乘支持向量机（LS—SVM）的预测模型。由于LS—SVM的参数选择直接影响着模型的预测精度，于是采用一种基于量子粒子群优化方法来选择模型的超参数。为了弥补模型损失的鲁棒性，通过给每个样本误差不同的权系数，建立了具有良好泛化性能的WLS—SVM回归模型，从而进一步提高了模型预测的精度。本文提出一种基于量子粒子群优化（0uantum—behaved Particle Swarm Optimization，QPSO）参数选择的加权最小二乘支持向量机（Weighted Least Squares Support Vector Machine，WLS—SVM）的超短期风电功率预测模型。应用上述方法对内蒙古地区大型风电场进行了预测，结果证明了该方法的有效性。

Ultra-Short-Term Wind Power Prediction for Large-Scaled Wind Farms

For large-scale wind farm wind power nonlinear characteristics, we use least squares support vector machine prediction models. As the LS-SVM parameter selection directly affects the precision of the model, this paper uses a method based on quantum particle swarm optimization to choose the model of super parameters. To compensate loss of robustness of the model, by giving different weights to each sample error coefficient of this paper, a good generalization performance of the WLS-SVM regression model is established to further improve the model prediction accuracy. This paper presents an ultrashort term wind power prediction model which is based on the quantum particle swarm optimization （Quantum-behaved Particle Swarm Optimization, QPSO） parameters of the weighted least squares support vector machine （Weighted Least Squares Support Vector Machine, WLS-SVM）. The method as described in the paper has been applied in the wind power prediction in largescaled wind farms in Inner Mongolia, and the actual results prove effectiveness of the method.