基于RBF神经元网络的风电功率短期预测

摘要

Accurate wind power outputs forecasting plays an important role in power system dispatching, power system stability,and wind farm operation. Based on historical data from an operating wind farm such as wind speed, environmental temperature, wind power and so on, a short-term wind power forecasting model based on the well-developed Radial Basis Function(RBF) neural network is presented for hour-ahead forecasting, and the predicted error is about 12％. The forecasting results are compared with actual wind power outputs, and this shows that the presented method can lead to acceptable and stable forecasting results.This work is supported by National Natural Science Foundation of China (No. 70673032).%准确地预测风力发电的输出功率对电力系统调度、电力系统稳定性和风电场运行都具有重要意义.从实际运行的风电场获得了相关风速、环境温度和风电功率的历史数据,建立了基于径向基函数( Radial Basis Function,RBF)神经元网络的短期风电功率预测模型.运用该模型进行了1 h后的风电输出功率预测,预测误差在12%附近.通过将预测结果和实际风电输出功率比较,表明该方法预测精度较高且比较稳定.