Rapid prediction of urban flooding is an important measure to reduce the risk of flooding and to protect people’s property. In order to meet the needs of emergency flood control, this paper constructs a rapid urban flood prediction model based on a machine learning approach. Using the simulation results of the hydrodynamic model as the data driver, a neural network structure combining convolutional neural network (CNN) and long and short-term memory network (LSTM) is constructed, taking into account rainfall factors, geographical data, and the distribution of the drainage network. The study was carried out with the central city of Zhoukou as an example. The results show that after the training of the hydrodynamic model and CNN−LSTM neural network model, it can quickly predict the depth of urban flooding in less than 10 s, and the average error between the predicted depth of flooding and the measured depth of flooding does not exceed 6.50%, which shows that the prediction performance of the neural network is good and can meet the seeking of urban emergency flood control and effectively reduce the loss of life and property.

At present, urban flood risk analysis and forecasting and early warning mainly use numerical models for simulation and analysis, which are more accurate and can reflect urban flood risk well. However, the calculation speed of numerical models is slow and it is difficult to meet the needs of daily flood control and emergency. How to use artificial intelligence technology to quickly predict urban flooding is a key concern and a problem that needs to be solved. Therefore, this paper combines a numerical model with good computational accuracy and an LSTM artificial neural network model with high computational efficiency to propose a new method for fast prediction of urban flooding risk. The method uses the simulation results of the numerical model of urban flooding as the data driver to construct the LSTM neural network prediction model of each waterlogging point. The results show that the method has a high prediction accuracy and fast calculation speed, which can meet the needs of daily flood control and emergency response, and provides a new idea for the application of artificial intelligence technology in the direction of flood prevention and mitigation.

Design rainfall is the basis for deriving design floods in areas where rainfall data are lacking and has a significant impact on the construction of water engineering facilities and municipal engineering designs. The Chicago rainfall pattern method has great applicability for urban short-duration design rainfall. In order to analyze the influence of design storm rainfall patterns on urban flooding, numerical models of hydrological and hydrodynamic processes were applied to simulate design rainfall with different recurrence periods and different rain peaks and were also used to compare and analyze the total amount of water accumulation and inundation extent by taking the central city of Zhoukou as an example. The results show that when the design rainfall recurrence period is less than 20 years, the total volume and inundation extent of waterlogging in design rainfall with a smaller peak ratio is larger. When the return period is greater than 20 years, the pattern is reversed. However, as the return period grows, the difference in peak inundation volume due to different peak rainfall amounts decreases. This study has certain guiding significance for urban flood forecasting and early warning efforts.

Soil compaction easily causes root hypoxia stress, resulting in poor root growth and the absorption of soil water and nutrients. We hypothesized that aerated irrigation (AI) could enhance nutrient uptake and utilization, thus unlocking the high yield potential by increasing soil aeration and root morphology indicators compared with that in the non-aeration treatment. A greenhouse experiment was conducted to investigate the effect of soil aeration and root morphology on the yield of greenhouse cucumbers. The dissolved oxygen (DO) in irrigation water at 10 mg L⁻¹ (A₁), 20 mg L⁻¹ (A₂), and non-aeration treatment (A₀) were applied via a subsurface drip irrigation system. The soil respiration rates, DO in soil water, root morphology, and crop yield were measured. The results showed that AI could significantly improve the soil respiration rate, DO in soil water, and root morphology compared with non-aeration treatment. The A₂ significantly increased soil respiration rate by 11.63% and 11.93%, respectively, compared with the A₁ and A₀ treatments. Under A₁ and A₂, the DO in soil water increased by 20.01% and 18.02%, respectively, compared with the A₀. Moreover, during the flowering and fruit set, the mature, and the late stages, the root surface area, root volume, root tip number, root forks, and root dry weight in the A₂ treatment significantly increased than that in the A₀ treatment. The soil respiration rate, DO in soil water, root length, and root forks were the main indexes correlated to the yield, respectively. The DO in soil water and root forks number significantly influenced the yield. The cucumber yield and economic benefits in A₂ peaked at 53.04 t ha⁻¹ and 3.95 × 10⁴ USD ha⁻¹, increased by 7.86% and 7.30% compared with that in the A₀ treatment, respectively (p < 0.05). The results could provide technical support and scientific knowledge for regulating greenhouse cucumbers under AI.

In order to investigate the influence of different design rainfall peaks on urban flooding characteristics based on the MIKE hydrodynamic model, the inundation process scenarios were extrapolated for different recurrence periods and three single- and double-peak rainfall types in Zhoukou city as an example, and the equivalent values of total inundation and inundation area were compared and analysed. The results show that bimodal rainfall has a higher risk of inundation than unimodal rainfall for the same rainfall ephemeris and return period. For unimodal rainfall, when the return period is less than 20 years, the more advanced the rainfall peak, the more severe the design rainfall inundation. When the return period is greater than 20 years, the further back the rainfall peak, the more severe the inundation of the design rainfall. The difference between the risk of inundation due to single- and double-peaked rainfall decreases as the return period increases.

By changing the alignment angle of spur dike, this study focused on the evolution of scour hole morphology in three alignments under clear-water scour conditions, including the 3-D structure of the scour hole, 2-D profile morphological evolution process and the evolution characteristics of the local bed shear stress. The results show that the plane area and volume of the scour hole both are power functions over time, which is similar to the evolution characteristics of scour depth. Local scour includes three stages: The initial stage, development stage and balance stage. The local bed shear stress presents successively: τ_b > τ_c, τ_b = τ_c and τ_b < τ_c. Based on this characteristic, the evolution mechanism between scour hole morphology and the local bed shear stress is further clarified. Furthermore, although the alignment of the spur dike significantly affects the longitudinal and vertical dimension erosion rates of the scour hole, the scour hole morphology is not only relatively constant but also has a specific proportion, and the evolution process is orderly in the whole process of evolution. To the scouring equilibrium state, the length of the scour hole on the upstream and downstream of the spur dike is approximately in line with the golden section feature. The related results provide technical support for scour parameter design and scour hole protection of spur dike in flood period.