基于灰色神经网络的兆瓦级风电机组输出功率预测研究＊

摘要

With the increasing of the wind power installation capacity and as regard to the power grid dispatching, operating and maintenance, it is very important to accurate predict the output power of the wind power. The grey neural network forecasting model is established. The grey theory and the neural network are combined together to realize the forecast of the unit output power in the whole region. In order to show that the proposed method effective and correct, the simulative experiments are conducted by taking the data from Northwest Jiuquan Wind Power Base as the samples. The 1.5 MW wind turbine prediction model is established for verification of the scheme, and the quantitative analysis of the prediction error is also given. The experimental results show that the grey neural network has higher prediction accuracy to the MW class wind turbine output power and has a certain researching significance.%随着风电装机容量的增加，准确预测风力发电输出功率对电网的调度、风电的运营维护有重要作用。建立灰色神经网络预测模型，将灰色理论和神经网络这两种控制方法结合在一起，实现整个区域的机组输出功率预测。采用西北酒泉风力发电基地的数据作为样本进行模拟实验，建立1．5兆瓦级风电机组的预测模型验证所提方法的有效正确性，并定量分析了预测误差，结果表明灰色神经网络对兆瓦级风电机组输出功率具有较高的预测精度，具有一定的研究意义。