Damage caused by flash floods is increasing due to urbanization and climate change, thus it is important to recognize floods in advance. The current physical hydraulic runoff model has been used to predict inundation in urban areas. Even though the physical calculation process is astute and elaborate, it has several shortcomings in regard to real-time flood prediction. The physical model requires various data, such as rainfall, hydrological parameters, and one-/two-dimensional (1D/2D) urban flood simulations. In addition, it is difficult to secure lead time because of the considerable simulation time required. This study presents an immediate solution to these problems by combining hydraulic and probabilistic methods. The accumulative overflows from manholes and an inundation map were predicted within the study area. That is, the method for predicting manhole overflows and an inundation map from rainfall in an urban area is proposed based on results from hydraulic simulations and uncertainty analysis. The Second Verification Algorithm of Nonlinear Auto-Regressive with eXogenous inputs (SVNARX) model is used to learn the relationship between rainfall and overflow, which is calculated from the U.S. Environmental Protection Agency’s Storm Water Management Model (SWMM). In addition, a Self-Organizing Feature Map (SOFM) is used to suggest the proper inundation area by clustering inundation maps from a 2D flood simulation model based on manhole overflow from SWMM. The results from two artificial neural networks (SVNARX and SOFM) were estimated in parallel and interpolated to provide prediction in a short period of time. Real-time flood prediction with the hydraulic and probabilistic models suggested in this study improves the accuracy of the predicted flood inundation map and secures lead time. Through the presented method, the goodness of fit of the inundation area reached 80.4% compared with the verified 2D inundation model.

Because of their enclosed nature, underground spaces are more vulnerable to flooding than above-ground structures. As a result, flooding disasters have the potential to cause widespread casualties. Major domestic plan manuals such as ‘Waterproof Standards Plan Manual for Underground Space Flood Prevention’ suggest that the expected flood height is determined and measures are established to ensure safety for underground facilities located in areas where flooding is expected, but no clear methods and standards are presented. In this study, an inundation prediction chart was prepared by performing a one-dimensional (EPA-SWMM) model analysis using the probability rainfall, respectively. Then, data on the depth of inundation where the underground facility is located were extracted using a two-dimensional (FLO2D) model analysis. Using inundation depth and weather data, the probability of inundation for underground facilities and the expected inundation height were quantitatively evaluated through @RISK. Based on the quantitative assessment, it was determined that there was a 7.7% possibility of inundation in an underground space in the case of a 2 h time frame (average inundation depth: 0.1557 m), and a 13.3% possibility in the case of a 3 h time frame (average inundation depth: 0.1655 m). It would be possible to contribute to the development of countermeasures and manuals for underground space flooding using the quantitative assessment described above.

As data analysis becomes increasingly important in the field of flood disasters, a data-based analytical model that considers hydraulic factors has emerged as an alternative to flood analysis. Data-driven modeling requires complex and extensive data that depend on actual and diverse hypothetical scenarios. Thus, preliminary data processing is required to utilize extensive input data efficiently and obtain accurate results. In this study, a gamma test was applied as a preliminary data processing technique to the efficient prediction of flood disasters in urban areas. Specifically, probability rainfall was calculated for the Daejeo and Dadae districts in Busan, and manhole overflow was estimated by inputting relevant data into a storm water management model (SWMM) with a Huff distribution. A preliminary uncertainty and sensitivity analysis was performed with the gamma test using combined time-series data for probability rainfall and manhole overflow, and then an optimal combination of data was selected for a simulation using correlativity analysis (Gamma, Gradient, Standard error, and V-ratio). The proposed method employs quantitative criteria for efficient selection of optimal input data, thus saving a greater amount of time and effort than when using the traditional method that derives results from repeated execution of models.