Power output of wind farm is affected by wind speed, temperature, pressure, humidity and other uncertain factors, resulting in that the prediction accuracy of wind power is lower. Therefore, combined forecasting model of fuzzy theory and support vector machine theory is put forward in this pa-per. The combination forecasting method calculates fuzzy membership values corresponding to the sample data by using fuzzy membership function, and then generates fuzzy sample set. Then model of fuzzy sam-ple set is done by using support vector machine (SVM). Finally prediction model is used for wind power prediction in the next 24 hours. The simulation experiment shows that the combined forecasting method can effectively improve forecasting accuracy of the wind power and has comparatively strong practicality in engineering.%由于风电场输出功率受风速、气温、气压和湿度等不确定性因素的影响,导致风电功率的预测精度偏低。因此,提出了模糊理论与支持向量机理论相结合的组合预测模型。该组合预测方法用模糊隶属函数计算样本数据的模糊隶属度值,生成模糊样本集。然后用支持向量机对模糊样本集训练建模,最后利用预测模型预测未来24小时的风电功率。仿真实验表明,该组合预测方法有效地提高了风电功率预测精度,具有很强的工程实用性。
展开▼