论文部分内容阅读
基于最小二乘支持向量机理论,建立风速预测模型。同时,由于最小二乘支持向量机参数选取尚无有效方法,该文尝试采用蚁群算法理论来进行参数优化选择。选取某风场前四天的实测风速(采样间隔30min),应用所建立的风速预测模型,来预测第五天的48个风速值,其预测的平均绝对百分比误差仅为9.53%,预测效果较理想,验证了应用蚁群优化算法理论与最小二乘支持向量机理论进行风速预测的可行性,可为风电场规划选址和风力发电功率预测等提供理论支持。
Based on least-squares support vector machine theory, wind speed prediction model is established. At the same time, there is no effective method for parameter selection of LS-SVM. This paper attempts to use the ant colony algorithm theory to optimize parameters selection. The wind speed forecasting model established in the first four days of a certain wind farm (sampling interval 30min) was used to predict the 48th wind speed on the fifth day. The average absolute percentage error of prediction was only 9.53% The feasibility and feasibility of wind speed forecasting based on the theory of ant colony optimization and least square support vector machine theory are validated, which can provide theoretical support for the site selection of wind farm and wind power prediction.