Real-time prediction and ponding process early warning method at urban flood points based on different deep learning methods

Abstract

Accurate prediction of urban floods is regarded as one of the critical means to prevent urban floods and reduce the losses caused by floods. However, due to the uncertainty and complexity of urban floods and waterlogging, it is very difficult to use simulation models to quickly and accurately predict and warn of urban floods. Therefore, it is necessary to develop new methods to support the rapid and accurate prediction of urban floods and waterlogging. In this study, a refined prediction and early warning method for urban flood and waterlogging processes based on deep learning methods is proposed from three aspects: research feasibility, method applicability, and method system. The spatial autocorrelation of rain and ponding points is analyzed by Moran's I. For each ponding point, the relationship model between the rainfall process and ponding process is constructed based on different deep learning methods, and the results are analyzed and verified by a statistical evaluation method. The results show that the gradient boosting decision tree algorithm has the highest accuracy and efficiency (with a root mean square error of 0.001) for ponding process prediction and is regarded as the most suitable method for ponding process prediction. Finally, the real-time prediction and early warning of urban floods and waterlogging processes driven by rainfall forecast data are realized, and the results are verified by the measured data. In addition, the results of the sensitivity analysis show that the rainfall peak and location coefficient have the greatest impact on ponding.