基于QPSO-LSSVM的风电场超短期功率预测

准确预测风电场的发电功率，有利于电网的经济和安全调度。为提高风电场超短期功率预测的精度，建立了基于最小二乘支持向量机（LSSVM）的风电场超短期功率预测模型，并采用量子粒子群算法（QPSO）对LSSVM中影响回归性能的参数进行优化。通过对福建某实际风电场超短期功率预测的应用表明，与BP神经网络和QPSO-LSSVM的预测结果相比，QPSO-LSSVM预测模型多种误差指标均较小，具有较高的预测精度和鲁棒性，是一种有效的风电场超短期功率预测方法。

Ultra-Short-Term Wind Power Forecasting Based on QPSO-LSSVM

An accurate forecast of wind power is beneficial to the economic and security dispatch of the power grid. To improve the accuracy of ultra-short-term wind power forecasting, this paper applies the quantum particle swarm optimization(QPSO) to optimize the parameters affecting the regression performance of the least squares support vector machine (LSSVM) in the QPSO-LSSVM model. Based on actual application in a wind farm, it is shown that the QPSO-LSSVM forecasting model has smaller errors on a variety of indicators than the BP neural network based model and PSO-LSSVM model. It has high forecasting accuracy and is robustness, and can be an effective ultra-short-term wind power forecasting methodology.