Real-Time Flood Classification Forecasting Based on k-Means++ Clustering and Neural Network

Abstract Floods are one of the most dangerous disasters that affect human beings. Timely and accurate flood forecasting can effectively reduce losses to human life and property as well as improve the utilization rate of flood resources. In this study, an F-K++ real-time flood classification forecasting framework based on k-means++, principal component analysis, and backpropagation (BP) neural network was constructed using the Jingle sub-basin, a tributary of the Yellow River Basin, as the study area. The model parameters of different flood types were obtained, and a flood classification and predictions were made based on the mature M-EIES model applied in the study area. The relationship between the hydrological model parameters and hydrological characteristics of the different flood types was analyzed. The results show that the flood real-time classification framework established in this study can be used for flood classifications and predictions in real time. The parameters of the flood classification and prediction model were consistent with the characteristics of the flood events. Compared with unclassified predictions, the accuracy of the classification and prediction models significantly improved. The Nash coefficient increased by 5%–11.62%, the relative error of the average flood peak was reduced by 6.08%–12.7%, the relative error of average flood volume was reduced by 5.74%–8.07%, and the time difference of the average peak was reduced by 43%–66%. The methodology proposed in this study can identify extreme flood events and provide scientific support for flood classifications and predictions, flood control and disaster reduction in river basins, and the efficient utilization of water resources.