14 results about "Rainfall runoff" patented technology

This invention provides a multi-model fusion method, device, electronic device, and system for debris flow early warning. The method includes: collecting geographical and geological data, real-time rainfall, runoff, and auxiliary monitoring data; fusing a shallow flow numerical model, a moving least squares machine learning model, and a generalized regression neural network model, and obtaining a comprehensive runoff prediction through weighted least squares calculation; constructing critical runoff indicators based on the safety requirements of disaster-bearing bodies such as roadbeds, retaining walls, and culverts, and inversely calculating critical rainfall thresholds, then making differentiated adjustments based on the average gradient of the gully; generating a debris flow early warning report by comparing real-time predictions with thresholds, achieving final graded and zoned early warning for the warning area. This invention improves the accuracy and engineering relevance of early warnings by integrating physical mechanisms and data patterns, achieving refined early warning from "whether it will occur" to "when, where, and what the impact will be," while reducing the error of a single model and adapting to different terrains and disaster-bearing bodies along highways.

The application discloses a mountainous small watershed flood forecasting method based on rainfall scenario identification and terrain adaptation, belongs to the technical field of hydrology, selects a plurality of typical small watersheds with rich rainfall runoff observation data, constructs a mountainous small watershed rainfall sample set with multi-dimensional characteristics, and divides the rainfall sample set into sample subsets of multiple rainfall scenarios; the average flow velocity of the water flow in the flood process is analyzed; the characteristics of the underlying surface of each typical watershed are counted, the main underlying surface factors affecting the average flow velocity of the watershed are obtained through dimension reduction screening, and the correlation function between the underlying surface characteristics and the average flow velocity of the watershed in different rainfall scenarios is fitted; the spatial distribution of the average flow velocity of the unrecorded mountainous small watershed is batch calculated; the rainfall scenario most similar to the rainfall event in the prediction period is intelligently identified and predicted, and the concentration unit line in the rainfall scenario is automatically recommended; the runoff of the watershed is calculated based on the rainfall, and the flood process of the outlet of the batch mountainous small watersheds in the prediction period is calculated.

The application discloses a high-resolution flood simulation method coupling topological optimization motion wave and cellular automata, comprising the following steps: obtaining the basin data of a target region, and re-projecting the basin data to a high-resolution grid coordinate system to construct an upstream and downstream relationship grid graph; traversing the upstream and downstream relationship grid graph, taking the rainfall runoff obtained at a current time as a lateral inflow, combining a water flow state at a previous time step and an inflow state of a corresponding grid in an upstream to construct a one-dimensional motion wave equation, and solving the one-dimensional motion wave equation to obtain a river water level at the current time; taking the river water level as an initial condition, starting a two-dimensional diffusion process on the basis of the upstream and downstream relationship grid graph, and iteratively obtaining a submerged range and a water depth distribution graph at the current time step; and repeating the above steps to obtain the submerged range and the water depth distribution graph at all time steps. The method provided by the application can realize fast deduction of a large-scale, high-resolution and physically strictly conservative flood evolution process.

This invention relates to the field of Haematococcus pluvialis cultivation technology, specifically to a rainwater runoff collection and management system for an open Haematococcus pluvialis pond. The system includes a canopy, and several water collection mechanisms arranged side-by-side on both sides of the open pond. Each water collection mechanism includes a water collection pipe located at the bottom of the sloping surface of the canopy, a diversion pipe connected in the middle of the water collection pipe, and a water storage tank connected at the bottom of the diversion pipe. This invention achieves a normally closed seal of the water collection pipe by using an arc-shaped drain rod driven by a float to block the water collection trough, ensuring the quality of collected rainwater from the source. By using a diversion ball valve with a three-way diversion channel driven by a servo motor, combined with a water quality monitoring unit and a control unit, real-time judgment and diversion of water quality are achieved. This results in the automatic discharge of heavily polluted initial rainwater and the collection of only high-quality mid-to-late-stage rainwater, significantly improving the quality of recycled water resources.

This invention discloses a method for predicting extreme rainfall runoff and infiltration by fusing confluence calculation and neural networks. The method includes: constructing a two-way coupling mechanism between surface runoff and soil infiltration processes; dynamically allocating rainfall through runoff over-infiltration determination; feeding the runoff results as source terms to a GPU parallel confluence calculation framework; dynamically feeding soil moisture back to infiltration capacity calculation; and iteratively solving the water balance equation using an implicit or semi-implicit scheme. This method achieves dynamic coupling and collaborative simulation of multiple water cycle processes under a unified framework, solving the problems of separation between surface water dynamics and infiltration models, non-conservation of mass, and insufficient physical consistency under extreme conditions in traditional methods.

This disclosure provides a method and system for predicting sediment concentration based on a freeze-thaw coupled physical constraint model, relating to the field of sediment concentration prediction technology. The method includes: acquiring and preprocessing various raw data such as hydrological data, meteorological data, and soil and vegetation data; constructing a baseline sediment concentration physical model based on the preprocessed raw data, and constructing an input feature vector; inputting the input feature vector into a sediment concentration prediction ANN model, and outputting the sediment concentration; wherein, during the training process of the sediment concentration prediction ANN model, the baseline sediment prediction of the physical model and the sediment mass conservation constraint are introduced into the ANN model loss function, and a freeze-thaw weighted strategy is used to enhance the loss function, resulting in a sediment concentration prediction ANN model that deeply integrates the physical model and the ANN model. This disclosure improves the stability and accuracy of predictions under extreme rainfall, runoff surges, or frequent freeze-thaw cycles.

The application discloses an intelligent interaction experience prediction scheme compiling method and system. The method comprises the following steps: constructing an experience prediction scheme compiling target section prediction system; selecting a historical flood sample in an intelligent interaction mode according to the target section prediction system and a basic hydrology database; selecting a rainfall start and end time period and a runoff start and end time period to carry out base flow segmentation in the intelligent interaction mode according to the selected historical flood sample, so as to obtain a rainfall runoff segmentation sample; generating a runoff scheme relationship line in the intelligent interaction mode according to the rainfall runoff segmentation sample and a unit runoff scheme of the target section prediction system; and completing each rainfall center scene compilation in the intelligent interaction mode according to the rainfall runoff segmentation sample and the runoff scheme relationship line. The intelligent interaction mode is used to replace a traditional manual operation mode, and the experience prediction scheme compiling convenience, objectivity and timeliness can be effectively improved.

This invention relates to the field of water environment monitoring and assessment technology, specifically a method and system for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics. The method includes acquiring water quality concentrations, flow velocity vectors, and regional rainfall-runoff data at monitoring nodes; identifying pollution types using the response time curves of water quality parameters and rainfall-runoff, and the concentration ratios between nodes; subsequently constructing a spatial transport matrix based on the velocity vector and spatial gradient distribution of concentration to quantify the mass load of pollutants migrating with the water flow per unit time; tracking the trajectory of the load evolution based on the matrix and calculating the self-purification degradation rate during the transport process; finally, establishing a spatiotemporal diffusion model by combining the load, degradation rate, and trajectory to simulate the coverage area and peak concentration of the pollution diffusion cloud within a preset time period, generating assessment results. This invention can achieve spatiotemporal dynamic tracking and accurate prediction of urban water pollution, providing a scientific basis for environmental governance decisions.

This invention discloses a method and system for constructing a rainfall-runoff model. The method involves collecting and preprocessing relevant daily sequence data within a watershed; constructing a hydrological model component library; pre-assembling the hydrological model structure; defining a parameter space; performing a collaborative search using genetic programming; simulating daily runoff using the optimal model structure and parameters obtained through the search and calibration; evaluating the model's generalization performance; and outputting the optimal model structure type, optimized parameter vector, calibration set, validation set performance indicators, and runoff simulation sequence. This invention can automatically generate rainfall-runoff models with adaptive model structures and interpretable hydrophysical mechanisms, effectively improving the automatic discovery capability of model structures and the interpretability of hydrophysical mechanisms, while reducing manual workload while avoiding overfitting.

The application provides an intelligent interception method, device, storage medium and program product for rainfall runoff. The method comprises the following steps: obtaining rainfall forecast data of a target area, calling a rainfall type discrimination module, discriminating and processing the rainfall forecast data, and generating target rainfall type data; determining a target time grouping of the rainfall forecast data, calling a water quality simulation database, and matching the target rainfall type data and the target time grouping with the water quality simulation database, so as to obtain a target interception time and a target interception water volume, and intelligently intercepting the rainfall runoff according to the target interception time and the target interception water volume. The method improves the interception efficiency of the urban drainage system, realizes accurate interception of the rainfall runoff with high pollution value, improves the interception response capability of the urban drainage system by continuously optimizing the interception strategy, is suitable for the formulation of the interception strategy under different rainfall conditions and time changes, and reduces the water pollution risk.

The application provides a small watershed flood forecasting and early warning method and system, and belongs to the technical field of mountain torrent disaster prevention. The method comprises the following steps: collecting multi-source data and preprocessing; constructing a physically guided neural network model, embedding the hydrodynamic physical equation as a constraint into the network training process; constructing a dynamic adaptability evaluation index system, calculating the initial values of the rainfall-runoff response consistency, model structure applicability and prediction uncertainty indicators; constructing a reinforcement learning agent, executing a cyclic correction mechanism with the initial values of the three indicators as the starting state to obtain the final converged values of the three indicators; configuring the parameters of the physically guided neural network model according to the converged indicator values, executing flood forecasting by adopting a hierarchical progressive strategy, and generating hierarchical early warning information in combination with dynamic hazard body information. The application realizes self-adaptive optimization and dynamic parameter adjustment of the prediction model, and solves the problems of poor model adaptability, parameter calibration dependence on historical data and single early warning indicator in the prior art.

This invention relates to the field of pollution load prediction technology, and discloses a method, apparatus, equipment, medium, and program product for predicting rainfall runoff pollution. The method for predicting rainfall runoff pollution includes: dividing the rainfall runoff area into spatial grids and obtaining the basic hydrological data of each spatial grid unit; obtaining the target pollutant accumulation process and the target pollutant flushing process based on the basic hydrological data of each spatial grid unit; constructing a rainwater runoff pollution prediction model based on historical rainfall data, the target pollutant accumulation process, and the target pollutant flushing process; and predicting the rainfall runoff pollutant load for the prediction period based on the rainfall data and the rainwater runoff pollution prediction model, thereby obtaining the target rainfall runoff pollutant load. This invention improves the accuracy of rainfall runoff pollutant load prediction through the analysis of gridded pollutant accumulation and pollutant flushing processes.

This invention provides a mechanism-based data-driven hydrological runoff prediction method and system. The method first collects historical rainfall-runoff, temperature, wind speed, humidity, water level, and flow data. It then processes the data using a dynamic noise amplitude adaptive decomposition method, dividing it into high, low, and mid-frequency components based on sample entropy. The high-frequency components are then further refined and key features are extracted. A multi-scale network model integrating physical information is constructed to improve the evapotranspiration, runoff generation, water source distribution, and confluence links of the Xin'anjiang hydrological model, achieving mechanism- and data-driven runoff prediction. Combining water level prediction with a peak constraint penalty mechanism, a block-based adaptive optimization algorithm is used to jointly optimize the model and decomposition parameters. This invention can suppress mode aliasing, significantly improve decomposition accuracy, quantify complexity through entropy values, achieve data-driven partitioning of low, mid, and high frequencies, avoid the subjectivity of human thresholds, and improve the physical interpretability of subsequent modeling.

This invention discloses a method, storage medium, and terminal for predicting mixed runoff. First, a distributed rainfall-runoff model is constructed based on DEM data to obtain raster-scale runoff field characteristics, river convergence characteristics, and physical flow calculated from the confluence, thus providing a clear physical basis for runoff prediction. By unifying the temporal and spatial characteristics of multi-source features, features from different data sources are jointly modeled under the same time and spatial reference. A Transformer encoder is used to jointly model the spatial location and temporal evolution relationship, effectively capturing the spatial correlation between upstream and downstream areas within the watershed and the temporal variation of the runoff process. An LSTM decoder predicts the physical flow residual and superimposes it with the physical flow, correcting deviations under complex nonlinear conditions while preserving the overall trend of the physical model, thereby improving the accuracy and stability of the runoff prediction results.