精确的风速预测对提高风能利用效率、推动电力市场的运行具有重要的 意义，但风速的突变性和非线性性等特点使得风速预测的挑战性大大增加。现有的风速预测方法研究逐渐以组合模型预测方法的创新为主流，本文尝试将一个新的组合模型应用于风速预测，即灰色模型与BP神经网络模型串联式组合，结合二者分别在数据依赖与非线性建模能力的优点来预测风速。此外，在考虑风速的时间相关性时，通过灰色关联分析更进一步地将气象因素也纳入研究范围，最终得到风速预测结果。经过在美国国家环境信息中心提供的2021-2023两年的实测逐日风速数据集上的实验验证，结果表明，GM-BP组合预测模型较单一预测模型的预测结果更为准确，验证了该方法的可行性。

Accurate wind speed prediction is of great significance to improve the efficiency of wind energy utilization and promote the operation of the power market, but the mutability and nonlinearity of wind speed make wind speed prediction much more challenging. This paper attempts to apply a new combined model to wind speed prediction, that is, the combination of gray model and BP neural network model, and combines the advantages of data dependence and nonlinear modeling ability to predict wind speed. In addition, when considering the temporal correlation of wind speed, the grey correlation analysis is used to further incorporate meteorological factors into the research scope, and obtain the final wind speed prediction results. After experimental verification on the measured daily wind speed dataset provided by the NCEI in the United States from 2021 to 2023, the results show that the GM BP combined prediction model is more accurate than that of the single prediction model, which verifies the feasibility of the method.