论文部分内容阅读
为了给铁路运输提供准确的大风预警,文中将人工神经网络算法应用于新疆百里风区的风速预报,分别采用BP神经网络与Elman神经网络建立模型,对实际历史风速数据进行仿真预测。利用实时自动站资料预测未来20分钟瞬间风速并做预报对比检验。结果表明:与BP神经网相比,Elman人工神经网络模型具有更好的拟合效果,独立样本预报及实际预报的检验结果均达到了较为精确的效果,具有实际应用意义。
In order to provide accurate warning of wind gusts to railway transportation, the artificial neural network (ANN) algorithm is applied to forecast the wind speed of the Baili wind in Xinjiang. The BP neural network and the Elman neural network are respectively used to establish the model, and the actual historical wind speed data are simulated and predicted. Using real-time automated station data to predict the next 20 minutes of wind speed and forecast comparison test. The results show that compared with BP neural network, Elman artificial neural network model has a better fitting effect, the results of independent sample forecasting and actual forecasting have achieved more accurate results, and have practical significance.