47 results about "Catchment hydrology" patented technology

The invention provides a method for monitoring hydrology and water quality of a river basin under influence of water projects based on Internet of Things. According to the method, various fixed and flowing sensors are additionally mounted so as to obtain real-time hydrology and water quality data of important water areas, in combination with a video technology, monitoring on a spot environment is achieved, the monitoring information is transmitted through the Internet of Things, intelligent prediction for hydrology and water quality of the river basin is carried out through a neural network, and finally, critical applications, such as current situation assessment, tendency estimation, implementation effect estimation and extreme event handling, are achieved through an expert system. The method provided by the invention can provide real-time, reliable and complete hydrology and water quality information for the river basin under influence of water projects, and can achieve cross-regional and multi-machine type integrated remote protection for the river basin.

The invention discloses an improved Xinanjiang model-based hydrological forecast method suitable for a hill region. The method comprises the steps of firstly, extracting geomorphologic data to build adigital elevation model (DEM); secondly, acquiring geomorphologic parameters, and establishing a geomorphologic instantaneous unit hydrograph (GIUH) according to the extracted geomorphologic parameters; and thirdly, replacing an original confluence calculation process of a Xinanjiang model with the GIUH, and carrying out hydrological forecast on the hill region after calibration by a constructedsemi-distributed watershed hydrological model. According to the method, global optimization is realized; the calculation efficiency is improved; the simulation precision is improved; and the requirements in multiple aspects of the hydrological forecast method are met.

The invention discloses a flood forecasting scheme real-time optimization method based on machine learning. The flood forecasting scheme real-time optimization method comprises the following steps: 1)collection and processing of basin hydrological data; 2) rainfall and flood session division and association; 3) generation of a rainfall flood event sample set; 4) flood grading; 5) construction ofa flood forecasting scheme; 6) classifier training based on machine learning; and 7) forecasting scheme real-time optimization based on early rainfall. According to the flood forecasting scheme real-time optimization method, the sample set is divided into subsets on the basis of flood levels, and the flood forecasting schemes are compiled respectively and associated with the early rainfall processthrough a machine learning method, and optimization of the flood forecasting schemes during real-time forecasting is achieved, and the real-time flood forecasting precision of the drainage basin canbe effectively improved.

The invention relates to a geological disaster prediction and forecast technology, and aims to provide a rainfall threshold analysis method causing basin landslide risks. The method comprises the following steps: establishing a three-dimensional grid according to DEM topographic data and soil layer thickness data of a target watershed, and generating boundary conditions and initial conditions required by an RIRM hydrological model; calculating space-time changes of hydrological elements in the drainage basin and drainage basin slope stability reference indexes of any three-dimensional positionin the drainage basin; drawing a relation curve that the safety factor SF, the landslide volume and the landslide area respectively change along with time under the conditions of different rainfall durations and different rainfall intensities; analyzing the relationship between landslide deformation damage and rainfall to obtain rainfall thresholds causing basin landslide risks under different conditions. According to the method, natural basin geomorphic hydrological response characteristics are reserved, and prediction errors can be effectively reduced; drainage basin hydrology and slope stability analysis can be carried out simultaneously, the actual occurrence process of drainage basin landslide disasters is met, and the prediction result is more real and reliable.

The invention discloses a hydrological simulation method based on Bayesian mode average fusion multi-source data. The method comprises the following steps of firstly, collecting meteorological data, hydrological series, satellite inversion rainfall and a reanalysis air temperature data set of a ground station in a scarce data area; respectively establishing correction models of ground observationdata and a simultaneous simulation meteorological data set in different months by adopting a quantile mapping-based daily deviation correction method, a regression correction method and an equal ratecorrection method; then adopting a seasonal Bayesian mode averaging method, optimizing the weight of each deviation correction scene through a posterior probability density function to acquire a corrected long-series meteorological data set; and calibrating a basin hydrological model and a long-short-term memory neural network model according to actual measurement data, and finally inputting the corrected long-series meteorological data set to realize runoff process simulation. Long-series runoff simulation of regions with scarce data can be realized, and an important reference basis with highoperability can be provided for basin water resource management and planning.

The invention discloses an urban flood assessment method based on urban toughness and an urban watershed hydrological model. The method comprises the steps: acquiring satellite remote sensing data andurban meteorological data; calculating flood disaster occurrence areas, flood ponding areas and ponding depths of cities over the years; constructing an urban drainage basin hydrological framework model and an urban flood disaster response toughness framework model; calculating the toughness index of the city in response to flood disasters; carrying out contribution analysis on all elements of the city, and obtaining dimensions corresponding to the elements with the indexes lower than a first set value, wherein the lower the indexes of the elements are, the weaker the corresponding dimensionsare; calculating the drainage pressure and the ultimate drainage capacity of the urban drainage pipeline opening, and calculating the ponding risk of each area of the city; according to the water accumulation risk of each region of the city, obtaining the city region with the water accumulation risk greater than a second set value, wherein the higher the ponding risk is, the more easily the floodoccurs in the corresponding area; and evaluating the urban flood area and the urban weak dimension in the urban rainfall process.

The invention discloses a method for improving simulation precision of a high-altitude drainage basin hydrological process without data. The main purpose is to provide a method for improving simulation precision of a high-altitude drainage basin hydrological process without data. According to the method, a hydrological physical process, a neighboring area law and a vertical differentiation law are used as basic theory bases, and a parameter transfer relation is built for multiple groups of soil and water assessment tool (SWAT) parameters adjacent to a station mountain drainage basin in Xinjiang; based on the parameter transfer relation, a tool parameter set of a reference basin is converted into a tool parameter set of a target drainage basin, so that high-precision simulation of the high-altitude drainage basin hydrological process without data is achieved. The method is applicable to hydrological process simulation of medium and small rivers of a high-altitude region without data; after localization of tool parameters is achieved, the simulation precision of the drainage basin hydrological process without data can be improved, the method is beneficial to monitoring and early warning of a mountain torrent, and the risk of the mountain torrent disaster is reduced.

Artificial intelligence-based watershed hydrology analysis and management having of a network of weather stations and artificial drainage systems with artificial and natural reservoir management through locks and pumping stations. Methods and systems evaluate hydrologic risk in each area and analyse the consequences of future precipitations using neural network assisted simulation. Hydrographs calculated for each sub-basin, streams and rivers in the basin simulates the behavior of the basin under different scenarios corresponding to different types of management of the operation of locks and / or pumps and compares its results in terms of loss of flooded area, economic loss in each area, loss for flooding of urban areas, etc. Optimization of the simulation by artificial intelligence meta-heuristic algorithms, multi-layered neural network acts as a search engine to find mitigation solutions and best configurations of resource management controls to minimize socio-economic impacts on each basin.

The invention relates to a watershed water resource management method based on supply type hydrological ecosystem service, which comprises the following steps: analyzing basic information acquired bysimulating a mechanical watershed hydrological model to obtain a total water yield, and taking the total water yield as the supply type hydrological ecosystem service to obtain a supply type hydrological ecosystem service spatial distribution diagram layer; quantifying the economic value of the supply type hydroecological system service by adopting a market value method to obtain an economic valuespatial distribution diagram layer; and simulating watershed water resource protection and division of a potential development area according to the two spatial distribution map layers by adopting aspatial ecological planning model. According to the invention, the influence and relative importance of land utilization change, soil properties, climate, terrain and the like on space-time distribution of water yield are analyzed by utilizing a mechanical basin hydrological model, and a water resource area needing to be protected emphatically and a water resource area potentially developed are identified by combining a space ecological planning model, so that the comprehensiveness and effectiveness of water resource protection are improved.

The invention provides a runoff calculation and prediction method based on a watershed hydrological model. The method comprises the steps: building a watershed hydrological model, and constructing a feature region basic watershed database; dividing hydrological response units according to different characteristic area basic drainage basin data; inputting localized model data, and carrying out related data simulation; determining model parameters of the watershed hydrological model; verifying the correctness of the watershed hydrological model; and inputting meteorological data in the watershed in a future climate scene into the watershed hydrological model, and predicting the river runoff in the hydrological response unit. According to the rainfall runoff and rainfall runoff prediction method, the whole runoff production and confluence process in the drainage basin is considered, the water flow movement process of the whole drainage basin is simulated by inputting climate data, drainage basin data and land utilization data of the drainage basin, and the rainfall runoff and rainfall runoff prediction method can be well applied to simulation of river runoff of the drainage basin and is fast in rainfall runoff calculation and program response.

The invention provides a monthly runoff prediction method based on a hydrothermal coupling balance principle, and belongs to the technical field of hydrological prediction. The method comprises the steps of generalizing the water storage space in the monthly-scale watershed hydrological cycle process into a soil water tank and an underground water tank under the natural condition; calculating themonthly-scale direct runoff of the soil water tank and the base flow of the underground water tank of a drainage basin; and adding the direct runoff of the soil water tank and the base flow of the underground water tank to obtain monthly runoff depth of the drainage basin, and obtaining monthly average flow volume of the drainage basin through unit conversion. According to the invention, a hydrothermal coupling balance principle is used as a basis, on the basis of analyzing a monthly-scale watershed hydrological cycle process mechanism, a proper hydrological cycle process description structureis selected, key processes and influence factors of a monthly-scale basin hydrological cycle process are grasped, a generalization method is adopted to express nonlinear relationships among monthly-scale hydrological processes, and nonlinear storage and discharge relationships between the soil and underground water storage structures are fully considered, so that monthly runoff prediction can beaccurately and effectively carried out.

The invention provides a method for calculating a river basin pollution source river entry coefficient based on a one-dimensional water quality model. The method comprises the steps of determining a river basin calculation range, and defining start and stop positions of upstream and downstream boundaries of a calculation area; collecting basin hydrological data and water quality data, and calculating pollution source emission statistical data in a regional range. Hydrological water quality data of upstream and downstream boundaries in the calculation area and emission data of pollutants in theregional range are obtained and substituted into an amphibious pollutant mass conservation equation, and a river entry coefficient of a basin pollution source can be calculated by utilizing a planning solving method. The model provided by the invention is simple in structure, and has the main characteristics of being simple and convenient, strong in operability and relatively high in simulation precision, and the river entering coefficient of a pollution source in the drainage basin can be well solved.

The invention discloses a watershed hydrological model parameter dynamic estimation method based on a digital twinborn technology. Constructing digital twin bodies of the watershed based on the physical characteristic data of the watershed and observation data such as historical hydro meteorology; sensors are arranged in the drainage basin to achieve real-time monitoring of multiple feature elements of the drainage basin, and real-time tracking and information synchronization of digital twins of the drainage basin are achieved; according to historical and real-time synchronous drainage basin digital twins, state variables and parameters in the simulation model are estimated in a mode of synchronously combining a data assimilation method and an optimization algorithm, and finally dynamic estimation of parameters of the drainage basin hydrological model is achieved. According to the method, the real-time monitoring data of the drainage basin can be fully utilized, the dynamic change presented by the characteristic conditions of the drainage basin under the changing environment background can be accurately reflected, the targets of intelligent management, analogue simulation, automatic optimization, real-time updating and reasonable prediction are achieved, and the runoff simulation precision of the drainage basin under the changing environment can be improved.

The invention relates to a hydrological model parameter calibration method based on multiple hydrological variables, and the method comprises the following steps: S1, obtaining long-time series hydrometeorological data, remote sensing inversion soil humidity and evaporation capacity data in a watershed range, and taking one year as a minimum unit, dividing the data period into a plurality of different sub-periods according to the conditions of annual precipitation, annual runoff and drainage basin underlying surface; s2, aiming at each sub-period, on the basis of remote sensing inversion evaporation capacity data and soil humidity index SWI data, calibrating hydrological model parameters in two modes of joint calibration and step-by-step calibration; and S3, by taking the King-Gupta efficiency coefficient as an evaluation index, analyzing and evaluating the flow process simulation effect of the model parameters in each rate stator sub-period and the flow process simulation effect of the model parameters transferred to other sub-periods. According to the method, the uncertainty of the model parameters can be effectively reduced, the mobility of the model parameters in different time periods is improved, and the reliability of the watershed hydrological model in production practice can be improved.

The invention discloses an automatic calibration method for hydrological model parameters in a data area. The automatic calibration method comprises the following steps: step 1, selecting hydrologicalmodel parameters needing to be calibrated; and step 2, adopting an artificial intelligence genetic algorithm to automatically calibrate hydrological model parameters needing to be calibrated in a data area. The technical problems that in the prior art, a traditional trial and error method is adopted for determining parameters of a hydrological model of a watershed with data, that is, parameter values of the hydrological model are continuously adjusted manually to meet the requirement for simulation precision, and the method is subjective, low in work repeatability, low in efficiency, extremely high in complexity, not beneficial to application and popularization of the hydrological model and the like are solved.

The present invention provides a watershed hydrology and water quality monitoring method applicable to water conservancy projects under the influence of water conservancy projects based on the Internet of Things technology. By adding various fixed and flowing sensors, the real-time data of hydrology and water quality in key water areas can be obtained, and realized with video technology. For the monitoring of the on-site environment, the Internet of Things is used to realize the transmission of monitoring information, the intelligent prediction of the hydrology and water quality of the basin is carried out through the neural network, and finally the key applications such as status evaluation, trend evaluation, implementation effect evaluation, and extreme event processing are realized through the expert system. The invention can provide real-time, reliable and complete hydrological and water quality information for the river basin under the influence of the water conservancy project, and realize the integrated remote protection of cross-region and multi-machine models in the river basin.

The invention discloses a real-time optimization method of a flood forecasting scheme based on machine learning, comprising the following steps: 1) collection and processing of watershed hydrological data; 2) division and association of rainfall and flood events; 3) generation of a sample set of rainfall and flood events; 4 ) Flood classification; 5) Construction of flood forecasting scheme; 6) Classifier training based on machine learning; 7) Real-time optimization of forecasting scheme based on previous rainfall. The invention divides the sample set into subsets based on the flood level, compiles the flood forecast schemes respectively, and associates them with the previous rainfall process through the machine learning method, realizes the optimization of the flood forecast scheme during real-time forecasting, and can effectively improve the real-time flood of the basin. forecast accuracy.

The invention provides a watershed wetland flood regulation and storage function quantitative evaluation method. The method comprises the steps of 1, constructing a watershed hydrological model coupled with a wetland module; 2, performing basin hydrological process simulation under different wetland distribution situations according to the basin hydrological model, and obtaining corresponding riverway runoff based on different basin hydrological process simulation results; 3, determining a flood event according to a flood event threshold value and the corresponding river channel runoff under different wetland distribution situations; extracting flood characteristics according to the flood event; and 4, quantitatively evaluating the flood regulation and storage function of the watershed wetland according to the flood characteristics. Through the wetland flood regulation and storage function, the construction of the watershed hydrological model coupled with the wetland module is firstlycarried out, so that the hydrological process simulation precision is greatly improved, and the watershed wetland flood regulation and storage function and the space-time difference thereof can be finely and quantitatively evaluated.

The invention discloses a method for researching the split ratio and the sewage separation ratio and the evolution rule thereof of a bifurcated channel. The method comprises the following steps: surveying hydrological information of a drainage basin where the bifurcated channel is located; calculating the river flow of the drainage basin under different assurance rates; constructing a water environment mathematical model; analyzing a split ratio rule of the bifurcated channel; analyzing a sewage separation ratio rule of the bifurcated channel; and analyzing the evolution rule of the split ratio and the pollutant separation ratio of the bifurcated channel. The invention provides a research method for analyzing the splitting ratio and the sewage separation ratio and the evolution rule thereof of a bifurcated channel, aiming at the current situation that the current research on the sewage separation ratio, the splitting ratio and the sewage separation ratio evolution rule of the bifurcated channel is insufficient, and a scientific basis is provided for formulating environmental management policies such as flood control and drainage, ecological water supplementation, total pollutant amount control, water environment protection and water ecological restoration.

The invention discloses a small reservoir flood level forecasting method based on a two-dimensional-zero-dimensional coupling model. According to the current situation that hydrological data of a drainage basin where a small reservoir is located is deficient, the drainage basin where the reservoir is located is divided into two parts with the water surface boundary corresponding to the limited water level of the reservoir in the flood season as a dividing line, and the upstream land part adopts a two-dimensional hydrodynamic model to simulate the surface runoff process; the downstream reservoir area part adopts a zero-dimensional water conservation model to calculate the reservoir water level, the flow of the coupling edge of the two-dimensional hydrodynamic model is used as the inflow condition of the zero-dimensional water conservation model, the newly calculated reservoir water level is used as the control boundary of the next time step of the two-dimensional hydrodynamic model, and the two models are calculated in a rolling manner; therefore, the purpose of reservoir water level forecasting is achieved. According to the method, only two parameters which do not depend on historical hydrological data need to be considered, and a new technical solution can be provided for forecasting the water level of the small reservoir in a watershed lack of historical hydrological data.

The invention discloses a remote sensing-based riverway water volume estimation method. The method comprises the following steps of obtaining a drainage basin boundary; acquiring a drainage basin remote sensing data image; cutting the drainage basin remote sensing data image: cutting the remote sensing data image in the step S2 by using the drainage basin boundary obtained in the step S1 in ENVI software to obtain the remote sensing data image of the whole drainage basin; extracting a basin water surface area; obtaining basin water volume data; and establishing a watershed water surface annualrunoff relationship: drawing a watershed water surface area and watershed water volume curve of the corresponding year, performing fitting to obtain a watershed runoff estimation formula and a correlation coefficient, and obtaining a river water volume estimation result through the watershed runoff estimation formula. According to the method, the water volume change of the river can be timely andquickly obtained, the problems of insufficient basin hydrological data and the like can be effectively solved, the dependence on actual measurement data is reduced, and the method has relatively goodprecision and extrapolation applicability.

The invention relates to a short-term power generation output prediction method for a distributed small hydropower station, and the method comprises the steps: carrying out the basin hydrological law analysis according to the historical electric quantity information and rainfall information of the small hydropower station, employing a distributed frame structure, and employing a large-scale data mining calculation method, and building a rainfall and power station electric quantity incidence relation calibration parameter; predicting the electric quantity of the power station in the future three days by adopting a chaos time sequence method through a future drainage basin rainfall prediction sequence, performing day-by-day process decomposition on the daily electric quantity by utilizing a similarity machine learning discrimination method, and finally obtaining an output process prediction sequence of the small hydropower station in the future three days. Therefore, the internal relation between the rainfall and the electric quantity of the small hydropower station without water regimen forecasting is accurately described, the daily electric quantity process is decomposed by using a recurrent neural network and similar data mining, and the daily output process forecasting of the small hydropower station without water regimen forecasting is realized. And the planned output compilation method of the water-regimen-free forecasting power station is provided for refined small hydropower dispatching.

The invention discloses a watershed hydrological model error correction method and device based on distributed evaporation error compensation. The method comprises the following steps: establishing a subunit evaporation error compensation vector model according to a watershed outlet section simulation flow process and a watershed subunit evaporation process; establishing a surface evaporation error compensation vector and an error compensation matrix model according to the subunit evaporation error compensation vector model; establishing a surface evaporation error estimation model and determining an evaporation error correction mode of the subunits; and calculating evaporation errors of the subunits to obtain a corrected evaporation process of the subunits, inputting the corrected evaporation process into a flow function of the drainage basin outlet section, namely a drainage basin hydrological model, and obtaining a corrected flow process of the drainage basin outlet section. The hydrological model feedback information is fully utilized, the method is decoupled from the hydrological model during error inversion, the method is theoretically suitable for any distributed hydrological model, and the method provides a new technical means for solving the watershed hydrological model daily scale evaporation input item error correction problem.

The present invention provides a multi-source precipitation product fusion method based on a dynamic heuristic algorithm, including: collecting limited station observation data in areas with scarce data, satellite inversion and reanalysis of precipitation data sets; using local magnitude scaling methods and equal rate The correction method corrects each data set and obtains the correction coefficient of each station; describes the ability of each data source to capture precipitation events through the assignment function, and calculates the state weight with space-time dynamic attributes; considers the ability of each data source to simulate the precipitation level , calculate the dynamic magnitude weight based on the cuckoo algorithm; use the ordinary Kriging interpolation method to dynamically map the correction coefficients, state weights and magnitude weights of all stations to the same spatial resolution, and obtain a long series of The raster quantitative precipitation product. The invention effectively integrates the advantages of each precipitation data source, corrects the system deviation of each precipitation data source, and provides a reference basis for river basin hydrological simulation and water resource planning.

The invention belongs to the technical field of hydrological analysis, and particularly relates to a runoff forecasting model construction method and device, electronic equipment and a medium. The method comprises the steps of obtaining a historical hydrological sequence of a target drainage basin and a target distribution function used for fitting a hydrological distribution condition of the target drainage basin; substituting the historical hydrological sequence into the target distribution function to obtain a first probability sequence; obtaining a probability prediction model, a second probability sequence and a prediction error sequence according to the first probability sequence; generating an error probability correlation function according to an error threshold sequence and the prediction error sequence; and according to the inverse function of the target distribution function, the probability prediction model and the error probability correlation function, generating a runoff prediction model for predicting the hydrological condition of the target drainage basin. According to the forecasting model construction method provided by the invention, through mutual cooperation of the target distribution function, the probability forecasting model and the error probability correlation function, the runoff forecasting model can have relatively high precision.

This application relates to a hydrological model parameter calibration method based on multiple hydrological variables, including the following steps: S1, obtaining long-term hydrometeorological data within the watershed range, and remote sensing inversion of soil moisture and evaporation data, with 1 year as the minimum unit , the data period is divided into several different sub-periods according to the conditions of annual precipitation, annual runoff, and the underlying surface of the watershed; S2. For each sub-period, based on remote sensing retrieval evaporation data and soil moisture index SWI data, The parameters of the hydrological model are calibrated in two ways: joint calibration and step-by-step calibration; S3, taking the Kling-Gupta efficiency coefficient as the evaluation index, analyzing and evaluating the flow rate of the model parameters in each stator period and transferred to other sub-periods Process simulation effect. The invention can effectively reduce the uncertainty of model parameters, improve the transferability of model parameters in different time periods, and is beneficial to improve the reliability of watershed hydrological models in production practice.

The invention discloses a data-free region hydrological parameter calibration method based on an adversarial neural network. The method comprises the following steps: collecting soil texture, vegetation coverage, land utilization rate, topographic data, runoff coefficient, annual total evaporation amount, gradient and gradient data; dividing the calibration region into calculation units within 30square kilometers or less; determining the underlying surface and meteorological correlation factors of each parameter of each calculation unit; adopting an adversarial neural network GAN to automatically calibrate the hydrological parameters of the watershed with the data to obtain the optimal hydrological parameters of each unit; training a unified parameter generator based on an adversarial neural network (GAN) by adopting the optimal hydrological parameters of all computing units in the area with data; determining hydrological parameters of the data-free region through the trained parameter generator. The technical problems that in the prior art, work repeatability and efficiency are low, complexity is extremely high, and application and popularization of a hydrological model are not utilized are solved.

The invention discloses a watershed hydrological similarity estimation method based on time series nuclear distance, comprising the following steps: S1. Obtaining hydrological observation data of the watershed to be studied; S2. Determining the precipitation process to be analyzed according to the hydrological observation data of the watershed to be studied and the analysis period of the hydrological response process; standardize the hydrological response time series of different analysis stations in the watershed to be studied to obtain the standardized sequence of the hydrological response process; S3. S4. Use the calculation method described in S3 to calculate the nuclear distance between the hydrological response processes of two different analysis stations, obtain the nuclear distance matrix D of the hydrological response process of different analysis stations in the watershed, and finally use the hierarchical clustering analysis method Analyze its hydrological similarities. The invention overcomes the disadvantage of high cost in the prior art, improves the utilization rate of the existing hydrological observation data, and reduces the estimation cost of the hydrological similarity of the existing watershed.

The invention provides an ensemble rainfall forecast and real-time flood probability forecast coupling method. The method comprises the steps of collecting hydro-meteorological basic materials of drainage basins and ensemble forecast rainfall data and building a hydrological model to obtain ensemble flood certainty forecasts; on the basis of real-time correction of error, converting corrected ensemble flood certainty forecasts into ensemble flood probability forecasts; performing weight fusion on the obtained ensemble flood probability forecasts to obtain comprehensive real-time flood probability forecasts. The method couples ensemble rainfall forecast information and Copula-BFS and is capable of considering the comprehensive influence of uncertainties such as rainfall forecast input, hydrological model parameters and model structures and estimating the total uncertainty of flood forecasts quantitatively. By performing error real-time correction on corresponding flood certainty forecasts before flood probability forecasts, the flood forecast precision can be increased and the performance of flood uncertainty forecast intervals is improved.

The invention discloses a runoff calculation method considering the probability distribution of the maximum depression flow storage depth, and relates to the field of hydrological model parameter setting; after considering the uncertainty of the parameter of the maximum depression flow storage depth, the method assumes The depth of flow storage in depressions on the cushion surface presents a certain probability distribution within a certain range, so that the improved runoff model can effectively solve the problem that the calculation unit cannot produce flow when the flow generation model of the basin has a small storage flow in the depression, and allows the model's The mechanism is closer to the actual runoff process, which provides a new way of thinking for the mechanism research and parameter setting research of the watershed hydrological model.