基于自适应神经网络的风光互补电站出力预测研究

Research on Output Forecasting of Wind Power and Photovoltaic Complementary Power Plant Based on Adaptive Neural Network

为提升风光互补电站出力预测准确度,研究了风光互补电站的出力预测方法。分析了风光互补电站机组出力情况,建立了广义回归神经网络模型和径向基神经网络模型用于训练历史数据。提出改进动态组群合作优化求解算法,利用该算法对风光互补电站出力进行预测,并利用仿真分析论证了提出模型的有效性,说明提出的方法能够有效降低预测误差,改善预测精度。

In order to improve the output prediction accuracy of wind-solar hybrid power plants,an output prediction model of wind-solar hybrid power plants based on adaptive neural network was proposed.The unit output of the wind-solar hybrid power plant is analyzed,and a generalized regression neural network model and a radial basis neural network model are established.A solution model of joint dynamic group cooperative optimization algorithm and sine-cosine optimization algorithm is proposed,and the model is used to calculate the wind-solar hybrid forecast,and the effectiveness of the model proposed in this paper is demonstrated by simulation analysis.