基于机器学习的降雨量雷达回波数据建模与预测

Rainfall modeling and prediction by radar echo data based on machine learning

以浙江省2016年1-10月的雷达回波强度数据为基础,分别应用随机森林模型、BP神经网络模型、卷积神经网络模型来预测降雨量并进行对比.建模分析结果表明,随机森林模型预测效果精确度较低,容易低估较大的降雨强度,而BP神经网络和卷积神经网络预测的效果都比随机森林好,特别是卷积神经网络,其预测值与真实值更加接近,且对较大的降雨强度拟合较好.

The rainfall is modeled and predicted based on the radar echo intensity data during January to October of 2016 in Zhejiang province,and the prediction results are compared between random forest method,BP neural network model,and convolutional neural network(CNN) model.The results show that the random forest model is relatively low in accuracy,and is easy to underestimate large rainfall intensity.The BP neural network and the CNN method perform better than random forest method,especially the convolutional neural network model.Compared with the other two machine learning methods,the CNN is better in prediction accuracy and large rainfall intensity fitting.