Under the combined influence of climate change, accelerated urbanization, and inadequate urban flood defense standards, urban pluvial flooding has become an increasingly severe issue. This not only poses significant challenges to social stability and economic development but also makes accurate flood risk assessment crucial for improving urban flood control and drainage capabilities. This study uses Jinan, a typical foothill plain city in Shandong Province, as a case study to compare the performance of differential evolution (DE), genetic algorithm (GA), and particle swarm optimization (PSO) in calibrating the SWMM. By constructing a hydrological–hydrodynamic coupled model using the SWMM and LISFLOOD-FP, this study evaluates the drainage capacity of the pipe network and surface inundation characteristics under both historical and design rainfall scenarios. An agent-based model (ABM) is developed to analyze the dynamic risks and vulnerabilities of population and building agents under different rainfall scenarios, capturing macroscopic emergent patterns from individual behavior rules and analyzing them in both time and space dimensions. Additionally, using multi-source remote sensing data, dynamic population vulnerability, and flood hazard processes, a quantitative dynamic flood risk analysis is conducted based on cloud models. The results demonstrated the following: (1) PSO performed best in calibrating the SWMM in the study area, with Nash–Sutcliffe efficiency (NSE) values ranging from 0.93 to 0.69. (2) Drainage system capacity was low, with over 90% of the network exceeding capacity in scenarios with return periods of 1 to 100 years. (3) The vulnerability of people and buildings increased with higher flood intensity and duration. Most affected individuals were located on roads. In Event 6, 11.41% of buildings were at risk after 1440 min; in the 20-year flood event, 26.69% of buildings were at risk after 180 min. (4) Key features influencing vulnerability included the DEM, PND, NDVI, and slope. High-risk areas in the study area expanded from 36.54% at 30 min to 38.05% at 180 min.

Driven by climate change and rapid urbanization, pluvial flooding is increasingly endangering urban environments, prompting the widespread use of coupled hydrological–hydrodynamic models that enable more accurate urban flood simulations and enhanced pluvial flood forecasting. The simulation method for urban river flooding caused by heavy rainfall has garnered growing attention. However, existing studies primarily concentrate on prediction using hydrodynamic models or machine learning models, and there remains a dearth of a comprehensive prediction framework that couples both models to simulate the temporal evolution of river flood changes. This research proposes a novel framework for simulating urban pluvial river flooding by integrating physically based models with deep learning approaches. The sample set is enhanced through data augmentation and Generative Adversarial Networks, and scheduling control signals are incorporated into the encoder–decoder architecture to enable urban pluvial river flooding forecasting. The results demonstrate strong model performance, provided that the model’s structural complexity is aligned with the available training data. After incorporating scheduling information, the simulated water level process exhibits a “double-peak” pattern, where the first peak is noticeably lower than that under non-scheduling conditions. The current research introduces an innovative method for simulating and analyzing large-scale urban flooding, offering valuable perspectives for urban planning and flood mitigation strategies.

Climate change and urbanization are increasingly threatening urban environments through pluvial flooding, prompting the widespread use of coupled hydrological–hydrodynamic models. These models provide accurate urban flood simulations and forecasting capabilities, and they can analyze the benefits of low-impact development stormwater control measures in surface-flooding processes. However, most studies have primarily focused on analyzing the stormwater control effects for specific flood events, lacking an analytical framework that accounts for uncertainty. This research proposes a framework for evaluating uncertainty in urban pluvial-flood stormwater control, combining urban-scale simulation, stormwater control modeling, and uncertainty analysis, while constructing nonlinear dependencies between the features reflecting the surface-flood-control benefits. Based on uncertainty analysis and copula methods, this research aims to support sustainable urban planning and provide a sustainable decision-making approach for urban stormwater management. The results show that the uncertainty assessment method based on generalized likelihood uncertainty is effective. By comparing the posterior joint distribution with the prior joint distribution, for different governance performance metrics, the joint, synergistic, conditional, and combined governance effects all exhibit consistent trends as the metrics change. The current research presents an innovative method for simulating and analyzing stormwater control benefits at the urban scale, providing valuable insights for urban sustainable development and flood mitigation strategies.

The effects of climate change and increasing urbanization mean that urban areas are facing a greater risk of serious flooding. The paper aimed to adopt a data-driven approach to capture surface flood-prone features, providing a basis for surface flood susceptibility. This research developed an enhanced framework En-XGBoost, which consists of three modules: the core module, preprocessing module, and postprocessing module. Data augmentation, random extraction strategies, and local enhancement were introduced to improve the model’s performance. En-XGBoost was tested in Fuzhou, China. The main findings were as follows: (1) Neighborhood information extraction strategy outperformed information extraction strategy in extracting detailed flood-prone features, producing clearer boundaries between different flood susceptibility levels, and refining the flood risk areas. (2) Crucial explanatory variables were identified as major drivers of flood risk, with location-specific factors influencing the flood causes, necessitating localized analysis for specific sites. (3) The local enhancement, data augmentation, and random strategies improved model performance, with data augmentation proving more effective for stronger models and having limited impact on weaker ones. Model performance requires an appropriate alignment between data complexity and model complexity. En-XGBoost provided support for capturing surface flood-prone features.