A flood forecasting method and system based on support vector machine model

Abstract

The present invention discloses a flood forecasting method and system based on a support vector machine model, which belongs to the field of runoff forecasting; the method and system include S1, collecting hydrological and meteorological data of a study area; S2, determining the parameter CN value of a runoff generation module of an SCS-CN model based on the collected hydrological and meteorological data; S3, improving the runoff generation module of the SCS-CN model by using rainfall characteristic factors and the parameter CN value of the runoff generation module; S4, establishing an SCS-CN three-water source confluence model by coupling the runoff generation module improved in S3 with a three-water source confluence module; S5, constructing a dynamic confluence hysteresis parameter through a support vector machine model to improve the SCS-CN three-water source confluence model; S6, forecasting the flood process line of a watershed outlet section by using measured rainfall data; considering the influence of rainfall characteristic factors, improving the prediction ability of the model in areas with relatively dry climates; considering factors such as the spatial distribution of rainfall events and previous rainfall, so as to reduce the risk of prediction deviation of the hydrological model due to uneven spatial distribution of rainfall.

Description

Technical Field

[0001] The present invention belongs to the field of runoff forecasting, and in particular relates to a flood forecasting method and system based on a support vector machine model. Background Art

[0002] In recent years, machine learning models have rapidly developed and are widely used in flood variable prediction and flood susceptibility assessment. Because machine learning models do not require large amounts of data, they offer advantages in predicting flood variables. Previous research has successfully applied machine learning models to problems related to flood variable prediction.

[0003] The SCS-CN model is a conceptual model developed by the United States Department of Agriculture that predicts runoff using only two parameters: the number of curves and the initial loss rate. Due to its minimal input data requirements, relatively simple model structure, and high prediction accuracy, it is widely used to estimate surface runoff from rainfall events in small watersheds. However, this model is a conceptual model developed based on runoff data from US watersheds. When applied to study areas with different climate types, it is necessary to comprehensively consider factors such as local vegetation, topography, and soils. Through continuous improvements by domestic scholars, the SCS-CN model has achieved good results in domestic watersheds. However, the traditional SCS-CN model only has a runoff generation module and does not include a runoff confluence module. Therefore, after rainfall occurs, it is impossible to map the changing flow trend at the watershed outlet section.

[0004] Most conceptual hydrological models consider the watershed as a whole, studying the operation and changes in the basin's runoff generation mechanisms after a rainfall event. Because conceptual hydrological models lack the ability to calculate basin cells, when rainfall is unevenly distributed, the predicted peak flood times can deviate significantly from the actual peak flood times, especially in study areas with narrow, long basins or large drainage areas. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the object of the present invention is to provide a flood forecasting method and system based on a support vector machine model.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A flood forecasting method based on a support vector machine model comprises the following steps:

[0008] S1, collect hydrometeorological data of the study area;

[0009] S2, based on the collected hydrological and meteorological data, determines the CN value of the runoff module parameter of the SCS-CN model;

[0010] S3, using rainfall characteristic factors and runoff module parameter CN value to improve the runoff module of SCS-CN model;

[0011] S4, by coupling the improved runoff generation module and the three-water source confluence module in S3, the SCS-CN three-water source confluence model is established;

[0012] S5, constructing dynamic confluence hysteresis parameters through support vector machine model to improve the SCS-CN three-source confluence model;

[0013] S6, use measured rainfall data to predict the flood process line of the basin outlet section.

[0014] Furthermore, the hydrological and meteorological data collected in S1 include: rainfall intensity, runoff, soil type and land use type.

[0015] Furthermore, in said S2, the step of determining the flow generation module parameters is:

[0016] S21: Determine hydrological soil groups using soil type data from the study area and divide them according to infiltration capacity. Resample the land use types in the study area on the Arcgis platform to match them with the grids under each soil type. Calculate the CN value in each grid, then perform a weighted average of the CN values ​​for each grid to determine the overall CN value for the watershed.

[0017] S22: Soil moisture conditions were divided into three levels: dry, normal, and wet based on the total rainfall in the previous five days. K-mean clustering was used to re-divide the intervals of soil moisture conditions. The total rainfall in the previous five days was the pre-influencing rainfall, and the threshold was set at 120 mm.

[0018] S23, convert the CN value according to the actual soil moisture conditions in the study area, interpolate the mutation intervals between CN1, CN2 and CN3, and obtain the CN value under different antecedent rainfall.

[0019] Furthermore, in S3, the improvement of the SCS-CN runoff model is to add the influence of rainfall characteristic factors on the basis of the original SCS-CN model. The specific steps are as follows:

[0020] S31, in the runoff module, the previous rainfall needs to supplement the initial water deficit of the basin. If the total rainfall does not meet the initial water deficit, the basin will not produce runoff. The runoff generated by the rainfall event is calculated as follows:

[0021]

[0022] Where: P is the total rainfall, mm; Q is the surface runoff, mm; I a is the initial loss, mm; S is the potential water storage capacity, mm;

[0023] S32, the calculation formula for initial loss and potential water storage capacity is:

[0024] I a =λS (2)

[0025]

[0026] Where: λ is the initial loss rate; CN is a dimensionless parameter;

[0027] S33, add characteristic factors such as rainfall intensity in the runoff generation module. The revised runoff calculation formula is as follows:

[0028]

[0029] Where: I 60 The maximum rainfall intensity in 60 minutes during the rainfall event, mm / h; is the average rainfall intensity of the rainfall event, mm / h; β is the rainfall intensity correction parameter; It is a comprehensive reflection of rainfall characteristic factors.

[0030] Furthermore, in S4, the process curve of the runoff depth in the study area can be obtained through the SCS-CN runoff generation module. The runoff depth generated in the study area in each period can be obtained by differential processing, and then input into the three water source confluence module for confluence calculation.

[0031] Furthermore, the three water sources refer to: surface runoff, subsoil flow and underground runoff, and are divided by parabolic free water storage curves and free water reservoirs.

[0032] Furthermore, in the above-mentioned runoff calculation, the unit line method is used for surface water runoff calculation, and the unit line is used to simulate the watershed as a series of n linear reservoirs; the linear reservoir method is used for subsurface flow and groundwater runoff runoff calculation; and the hysteresis algorithm is used for river network runoff calculation, and its calculation formula is:

[0033] Q3(I)=CS×Q3(I-1)+(1-CS)×QT3(IL) (5)

[0034] QT3(IL)=QS(IL)+QG(IL)+QI(IL) (6)

[0035] Where: Q3(I) is the unit area river network flow in the I period, m 3 / s; CS is the water receding coefficient of the river network; L is the confluence hysteresis parameter, h; QS is the surface runoff, m 3 / s; QG underground runoff, m 3 / s; QI soil midstream, m 3 / s; QT3(IL) is the sum of surface runoff, underground runoff and soil flow in the IL period, m 3 / s.

[0036] Further, In S5, The regression variables of the dynamic confluence hysteresis parameter include: the distance between the rainfall center and the basin section in the study area, the slope of the rainfall center, the total rainfall, the previous impact rainfall and the total river length; the regression function f(x) of the dynamic confluence hysteresis parameter is:

[0037]

[0038]

[0039] Where: (α i ,α i * ) is the Lagrange multiplier; b is the bias; K(x,x i ) is the radial basis function from the input space to the high-order feature space, that is, the kernel function; σ is the expansion constant of the radial basis function, which represents the radial range of the function.

[0040] Furthermore, in S6, first, the measured rainfall data and the runoff parameters of S2, S3, and S4 are input into the SCS-CN three-water source confluence model in S4 to obtain the runoff of each calculation grid in the study area. The runoff is then input into the improved three-water source confluence model in S5 to obtain the flood process line of the basin outlet section.

[0041] A flood forecasting system based on a support vector machine model, comprising:

[0042] Data collection unit: collects hydrological and meteorological data of the study area;

[0043] Runoff module parameter solving unit: Determine the CN value of the runoff module parameter of the SCS-CN model based on the collected hydrological and meteorological data;

[0044] Runoff module improvement unit: Use rainfall characteristic factors and runoff module parameter CN value to improve the runoff module of SCS-CN model;

[0045] SCS-CN three-water source confluence model construction unit: By coupling the improved runoff generation module and the three-water source confluence module, the SCS-CN three-water source confluence model is established;

[0046] SCS-CN three-source confluence model improvement unit: Improve the SCS-CN three-source confluence model by constructing dynamic confluence hysteresis parameters through the support vector machine model;

[0047] Flood forecasting unit: forecast flood process lines at the outlet section of the basin using measured rainfall data;

[0048] Beneficial effects of the present invention:

[0049] 1. This paper reclassifies the intervals for pre-annual soil moisture conditions using K-mean cluster analysis and employs cubic linear interpolation to derive CN values ​​for different pre-annual rainfall influencing amounts. It also considers the influence of rainfall characteristic factors such as rainfall intensity and average rainfall intensity, resolving the issue of rainfall intensity not being considered in the original runoff module and improving the model's predictive capabilities in arid regions.

[0050] 2. This paper uses a support vector machine model to construct dynamic basin runoff hysteresis parameters, taking into account factors such as the spatial distribution of rainfall events, previous rainfall, and underlying surface conditions. This reduces the risk of prediction bias in hydrological models due to uneven spatial distribution of rainfall, improves the ability of hydrological models to predict peak flood times, and, to a certain extent, alleviates the shortcomings and defects of conceptual hydrological models in the prediction process. This provides a solution for the application of conceptual hydrological models in complex research basins.

[0051] 3. The present invention obtains the flood process line of the basin outlet section by coupling the SCS-CN flow generation module and the three-water source confluence module. The SCS-CN three-water source confluence model has high accuracy in simulating peak flow and peak time, which expands the application scope of the SCS-CN model. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0053] Figure 1 A flow chart of the steps of the flood forecasting method;

[0054] Figure 2 This is a schematic diagram of the distribution of watershed stations in a village;

[0055] Figure 3 This is a distribution map of land use (left) and soil types (right) in a village watershed;

[0056] Figure 4 This is a simulation diagram of dynamic confluence hysteresis parameters based on the support vector machine model;

[0057] Figure 5 These are the simulation results of some events of the SCS-CN three-source confluence model. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0059] like Figure 1 As shown, a flood forecasting method based on a support vector machine model includes the following steps:

[0060] S1. Collect hydrological and meteorological data of the study area; including data on rainfall intensity, runoff, soil type and land use type.

[0061] S2. Determine the runoff module parameters of the SCS-CN model based on relevant data such as soil type and land use type in the study area. The specific steps are as follows:

[0062] S21: Hydrological soil groups were determined based on parameters such as soil texture and infiltration rate. Hydrological soil groups can be divided into four categories: A, B, C, and D, with decreasing infiltration capacity. Land use types in the study area were resampled using the ArcGIS platform to match the grids under each soil type. The CN (Curve Number) value was calculated for each grid, and the CN value for each grid was weighted averaged to determine the overall CN value for the watershed (assuming normal soil moisture).

[0063] S22, the SCS-CN model divides soil moisture conditions (AMC) into three levels based on the total rainfall in the previous five days, namely, drought (AMC1), normal (AMC2), and wet (AMC3);

[0064] The accuracy of determining the early soil moisture conditions will directly affect the prediction accuracy and stability of the hydrological model. This study statistically analyzed the relationship between the total rainfall in the five days before the flood and the suitable CN value, and used K-mean clustering to re-divide the interval of soil moisture conditions. The total rainfall in the first five days was the early influencing rainfall, and the threshold was set at 120 mm.

[0065] S23, the CN value is converted according to the actual soil moisture conditions in the study area; because the SCS-CN model soil moisture delineation method has mutation points, the mutation intervals between CN1, CN2 and CN3 are interpolated. CN1 is the CN value when the soil is relatively dry, CN2 is the CN value when the soil is normal (that is, the CN value obtained by S21), and CN3 is the CN value when the soil is relatively moist.

[0066] S3, using characteristic factors such as rainfall intensity and average rainfall intensity to improve the runoff generation module of the SCS-CN model;

[0067] The improvement of the SCS-CN runoff model is to add the influence of rainfall characteristic factors on the basis of the original SCS-CN model;

[0068] The SCS-CN model is an empirical model based on the water balance equation and two basic assumptions: the first assumption is that the ratio of surface runoff to the maximum possible runoff is equal to the actual infiltration volume and the potential water storage capacity; the second assumption is that there is a certain proportional relationship between the initial loss and the potential water storage capacity.

[0069] In the runoff module, the previous rainfall needs to supplement the initial water deficit of the basin. If the total rainfall does not meet the initial water deficit, the basin will not produce runoff. The runoff generated by the rainfall is calculated as follows:

[0070]

[0071] Where: P is the total rainfall; Q is the surface runoff; I a is the initial loss; S is the potential water storage capacity; the calculation formula for the initial loss and potential water storage capacity is:

[0072] I a =λS (2)

[0073]

[0074] Where: λ is the initial loss rate; CN is a dimensionless parameter that can be obtained from S2. The larger the CN, the greater the runoff generation capacity of the basin. Its value is related to factors such as the land use type, vegetation cover type, and previous rainfall in the study area. At the same time, the CN value is also a relatively sensitive parameter in the runoff generation module. Changes in the CN value will have a significant impact on the prediction results of the study basin.

[0075] In semi-arid areas, soil water levels are low and soil water deficits are high. After a rainfall, it may be difficult for the watershed to reach field capacity. Runoff will only occur when the rainfall intensity exceeds the infiltration intensity. Therefore, rainfall intensity in this area plays a major role in runoff generation. To address this, characteristic factors such as rainfall intensity are added to the runoff generation module to improve the model's prediction accuracy in semi-arid areas. The revised runoff calculation formula is as follows:

[0076]

[0077] Where: I 60 The maximum rainfall intensity in 60 minutes during the rainfall event, mm / h; is the average rainfall intensity of the rainfall event, mm / h; β is the rainfall intensity correction parameter, which needs to be calibrated through the rainfall process of previous flood events; It is a comprehensive reflection of rainfall characteristic factors such as total rainfall and rainfall intensity.

[0078] S4, by coupling the improved runoff generation module and the three-water source confluence module in S3, the SCS-CN three-water source confluence model is established;

[0079] The SCS-CN runoff generation module can be used to obtain the process curve of the runoff depth in the study area. Differential processing can be performed on it to obtain the runoff depth generated in the study area in each period, which is then input into the three-water source confluence module for confluence calculation; the three water sources refer to surface runoff, subsurface flow and underground runoff, which are divided by parabolic free water storage curves and free water reservoirs.

[0080] In the confluence calculation, the unit line method is used for surface water confluence calculation. The unit line is used to simulate the watershed as a series of n linear reservoirs. The confluence calculation of subsurface flow and groundwater runoff adopts the linear reservoir method. The hysteresis algorithm is used for river network confluence. The calculation formula is:

[0081] Q3(I)=CS×Q3(I-1)+(1-CS)×QT3(IL) (5)

[0082] QT3(IL)=QS(IL)+QG(IL)+QI(IL) (6)

[0083] Where: Q3(I) is the unit area river network flow in the I period, m 3 / s; CS is the water receding coefficient of the river network; L is the confluence hysteresis parameter, h; QS is the surface runoff, m 3 / s; QG underground runoff, m 3 / s; QI soil midstream, m 3 / s; QT3(IL) is the sum of surface runoff, underground runoff and soil flow in the IL period, m 3 / s.

[0084] S5, based on the SCS-CN three-water source confluence model obtained in S4, dynamic confluence hysteresis parameters are constructed through the support vector machine model to improve the SCS-CN three-water source confluence model;

[0085] When rainfall in a study basin is spatially uneven or the basin is long and narrow, the basin runoff hysteresis often changes. In the past, most hydrological models kept the runoff hysteresis constant, which could easily lead to a large deviation between the peak flood time predicted by the hydrological model and the actual peak flood time.

[0086] Based on this phenomenon, the support vector machine model was used to construct the dynamic confluence hysteresis parameter (L). First, the factors that may affect the confluence hysteresis of the basin were screened out. Then, principal component analysis was performed to determine the main influencing factors. The final regression variables included the length of the rainfall center from the basin section, the slope of the rainfall center, the total rainfall, the previous impact rainfall, and the total river length.

[0087] The basic idea of ​​support vector machine is to use kernel function to map low-dimensional nonlinear space to high-dimensional feature space, so that low-dimensional nonlinear hydrological data can be separated in high-dimensional space. After a series of derivations, its regression function f(x) can be finally expressed as:

[0088]

[0089]

[0090] Where: (α i ,α i * ) is the Lagrange multiplier; b is the bias; K(x,x i ) is the radial basis function from the input space to the high-order feature space, that is, the kernel function; σ is the expansion constant of the radial basis function, which represents the radial range of the function.

[0091] After constructing the functional relationship between the confluence lag parameter and the regression variable through the support vector machine model, the confluence lag parameter of the flood is predicted based on the measured site data and underlying surface information (including slope, river channel length, etc.), and then substituted into equations (5) and (6) to obtain the improved SCS-CN three-source confluence model.

[0092] S6, using measured rainfall data to predict the flood process line of the basin outlet section;

[0093] First, the measured rainfall data and the runoff parameters of S2, S3, and S4 are input into the SCS-CN three-source confluence model in S4 to obtain the runoff of each calculation grid in the study area. The runoff is then input into the improved three-source confluence model in S5 to obtain the flood process line of the basin outlet section.

[0094] A flood forecasting system based on a support vector machine model, comprising: a data acquisition unit, a runoff module parameter solving unit, a runoff module improvement unit, an SCS-CN three-water source confluence model construction unit, an SCS-CN three-water source confluence model improvement unit, and a flood forecasting unit;

[0095] in:

[0096] Data collection unit: collects hydrological and meteorological data of the study area;

[0097] Runoff module parameter solving unit: Determine the CN value of the runoff module parameter of the SCS-CN model based on the collected hydrological and meteorological data;

[0098] Runoff module improvement unit: Use rainfall characteristic factors and runoff module parameter CN value to improve the runoff module of SCS-CN model;

[0099] SCS-CN three-water source confluence model construction unit: By coupling the improved runoff generation module and the three-water source confluence module, the SCS-CN three-water source confluence model is established;

[0100] SCS-CN three-source confluence model improvement unit: Improve the SCS-CN three-source confluence model by constructing dynamic confluence hysteresis parameters through the support vector machine model;

[0101] Flood forecasting unit: forecast flood process lines at the outlet section of the basin using measured rainfall data;

[0102] Example:

[0103] The following is a specific example of applying the above flood forecasting method using a certain place:

[0104] The study area is located in a village watershed. The schematic diagram of the study watershed is as follows: Figure 2 The hydrological station is located downstream of the watershed, with a catchment area of ​​745 km² and an average annual rainfall of approximately 500 mm. The village's watershed has five rainfall stations, with a density of approximately 149 km² per station. The average slope of the watershed is 9.19%, indicating a temperate monsoon climate and semi-arid conditions. Land use is primarily farmland and forest.

[0105] The hydrological and meteorological data of the village basin from 1957 to 2004 (flood season from June to September) were obtained, including data on rainfall, runoff, evaporation, and land use types. The data were obtained from the Provincial Hydrological Bureau. After reviewing the three properties of the rainfall and runoff data, 25 floods were finally selected, including the largest flood in the history of the village hydrological station (peak flow of 4520m3 / s and peak water level of 57.78m).

[0106] We downloaded the village watershed DEM data from the Geospatial Data Cloud and used ArcGIS software to perform operations such as projection, resampling, depression filling, flow direction extraction, and flow volume extraction to obtain a geospatial distribution map of the study area. We then resampled the land use types within the study area to match the grids under each soil type.

[0107] Based on the soil saturation infiltration rate, minimum infiltration rate, and soil texture data of the village watershed, the village watershed was determined to be Class B hydrological soil group. The initial CN value of the study area was then determined based on the land use type of the study area (soil moisture was normal). Soil data was derived from the FAO World Soil Type Database with a spatial resolution of 1000m*1000m. Land use data was derived from the Global Land Cover Product (GlobeLand30) with a spatial resolution of 30m*30m. Their distribution is shown in Figure 3 .

[0108] The SCS-CN model determines soil moisture conditions based on the total rainfall in the previous five days, which may result in sudden changes. Therefore, this paper re-divides the soil moisture condition intervals of the original model using K-mean clustering and interpolates the sudden change intervals between CN1, CN2, and CN3. CN1 is the CN value when the soil is relatively dry, while CN3 is the CN value when the soil is relatively moist. The CN values ​​of the study area under different antecedent rainfall amounts are shown in Table 1:

[0109] Table 1. CN value of SCS-CN model

[0110]

[0111] After determining the CN value of different sessions, the session runoff can be calculated by the following formula:

[0112]

[0113] Where: P is the total rainfall, mm; Q is the surface runoff, mm; I a is the initial loss, mm; S is the potential water storage capacity, mm.

[0114] The initial loss and potential water storage capacity in the runoff module can be calculated using the following formula:

[0115] I a =λS (2)

[0116]

[0117] Where: λ is the initial loss rate.

[0118] In semi-arid areas, the soil water level is low and the soil water shortage is large. After a rainfall, the watershed may find it difficult to reach the field water holding capacity. Runoff will only occur when the rainfall intensity exceeds the infiltration intensity. Therefore, it is necessary to add characteristic factors such as rainfall intensity to the runoff module to improve the prediction accuracy of the model in this village watershed. The revised runoff calculation formula is:

[0119]

[0120] Where: I 60 The maximum rainfall intensity in 60 minutes during the rainfall event, mm / h; is the average rainfall intensity of the rainfall event, mm / h; β is the rainfall intensity correction parameter; It is a comprehensive reflection of rainfall characteristic factors such as total rainfall and rainfall intensity.

[0121] Through the above steps, the process curve of the runoff depth in the study area can be obtained. Differential processing can be performed to obtain the runoff depth generated by the village basin in each period, and then input it into the three water source confluence module for confluence calculation.

[0122] The constant term (B) and initial loss rate in the empirical relationship between potential water storage capacity and CN value have a certain impact on the runoff module. Therefore, in this example, the constant term and initial loss rate in the empirical relationship between potential water storage capacity and CN value in the runoff module were calibrated. At the same time, the sensitive parameters KI, KG, CS, CI, CG, and SM in the confluence module were calibrated. Sixteen floods from 1957 to 1979 were used to calibrate the model parameters, and the remaining nine floods were used to verify the model. The parameter calibration results are shown in Table 2, and the simulation results are shown in Table 3.

[0123] Table 2. SCS-CN model parameter values

[0124]

[0125] Table 3. Statistical indicators of the SCS-CN model simulation results for the village watershed

[0126]

[0127] The total runoff is divided into surface runoff, subsurface runoff, and groundwater runoff using a parabolic free-storage curve, and each is calculated using the confluence module. Furthermore, based on the concept of full-storage runoff generation, runoff can only occur within the runoff-generating area. All rainfall within the runoff-generating area generates runoff and enters the reservoir, becoming the free-water reservoir recharge.

[0128] In the confluence calculation, the unit line method is used for surface water confluence calculation. The unit line is used to simulate the watershed as a series of n linear reservoirs. The confluence calculation of subsurface flow and groundwater runoff adopts the linear reservoir method. The hysteresis algorithm is used for river network confluence. The calculation formula is:

[0129] Q3(I)=CS×Q3(I-1)+(1-CS)×QT3(IL) (5)

[0130] QT3(IL)=QS(IL)+QG(IL)+QI(IL) (6)

[0131] Where: Q3(I) is the unit area river network flow in the I period, m 3 / s; CS is the water receding coefficient of the river network; L is the confluence hysteresis parameter, h; QS is the surface runoff, m 3 / s; QG underground runoff, m 3 / s; QI soil midstream, m 3 / s; QT3(IL) is the sum of surface runoff, underground runoff and soil flow in the IL period, m 3 / s.

[0132] If the study area is set in a narrow and long basin or a large basin area, it is often necessary to consider the spatial distribution of rainfall. In this example, the support vector machine model is used to construct dynamic confluence hysteresis parameters. The relevant variables include the distance from the rainfall center to the basin section, the slope of the rainfall center, the total river length, the total rainfall, and the previous rainfall influencing hydrological and meteorological characteristic parameters. After a series of derivations, the regression function can be finally expressed as:

[0133]

[0134]

[0135] Where: (α i ,α i * ) is the Lagrange multiplier; b is the bias; K(x,x i ) is the radial basis function from the input space to the high-order feature space, that is, the kernel function; σ is the expansion constant of the radial basis function, which represents the radial range of the function.

[0136] Support vector machine models are often used to solve machine learning problems with small samples and can simplify classification and regression problems. Since the support vector machine model is a supervised machine learning method, it requires certain previous data to build a database. Therefore, some previous flood data is used to build the model. The actual value of the model is the confluence lag parameter when the flood peak time simulated by the hydrological model is the same as the actual flood peak time. The closer the predicted value is to the 1:1 straight line, the better the model prediction result. The prediction results are shown in Figure 4 .

[0137] In summary, the SCS-CN three-source confluence model can be constructed through the above steps. The simulation results of some events are shown in Figure 5 .

[0138] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0139] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. A flood forecasting method based on a support vector machine model, characterized in that: The following steps are involved: S1, collect hydrometeorological data of the study area; S2, based on the collected hydrological and meteorological data, determines the CN value of the runoff module parameter of the SCS-CN model; S3, using rainfall characteristic factors and runoff module parameter CN value to improve the runoff module of SCS-CN model; S4, by coupling the improved runoff generation module and the three-water source confluence module in S3, the SCS-CN three-water source confluence model is established; S5, constructing dynamic confluence hysteresis parameters through support vector machine model to improve the SCS-CN three-source confluence model; S6, using measured rainfall data to predict the flood process line of the basin outlet section; In S4, the process curve of the runoff depth in the study area can be obtained through the SCS-CN runoff generation module. The runoff depth generated in the study area in each period can be obtained by differential processing, and then input into the three water source confluence module for confluence calculation; The three water sources are: surface runoff, subsoil runoff and underground runoff, and are divided by parabolic free water storage curves and free water reservoirs; In the above-mentioned runoff calculation, the unit line method is used for surface water runoff calculation, and the unit line is used to simulate the watershed as a series of n linear reservoirs. The linear reservoir method is used for subsurface flow and groundwater runoff runoff calculation. The hysteresis algorithm is used for river network runoff calculation, and its calculation formula is: Q3(I)=CS×Q3(I-1)+(1-CS)×QT3(IL) (5) QT3(IL)=QS(IL)+QG(IL)+QI(IL) (6) Where: Q3(I) is the unit area river network flow in the I period, m 3 / s; CS is the water flow recession coefficient of the river network; L is the confluence hysteresis parameter, h; QS is the surface runoff, m 3 / s; QG underground runoff, m 3 / s; QI soil midstream, m 3 / s; QT3(IL) is the sum of surface runoff, underground runoff and soil flow in the IL period, m 3 / s; The regression variables of the dynamic confluence hysteresis parameter include: the distance between the rainfall center and the basin section in the study area, the slope of the rainfall center, the total rainfall, the previous impact rainfall and the total river length; the regression function f(x) of the dynamic confluence hysteresis parameter is: Where: (α i ,α i * ) is the Lagrange multiplier; b is the bias; K(x,x i ) is the radial basis function from the input space to the high-order feature space, that is, the kernel function; σ is the expansion constant of the radial basis function, which represents the radial range of the function; In S6, first, the measured rainfall data and the runoff parameters of S2, S3, and S4 are input into the SCS-CN three-water source confluence model in S4 to obtain the runoff of each calculation grid in the study area. The runoff is then input into the improved three-water source confluence model in S5 to obtain the flood process line of the basin outlet section.

2. A flood forecasting method based on a support vector machine model according to claim 1, characterized in that: The hydrological and meteorological data collected in S1 include: rainfall intensity, runoff, soil type and land use type.

3. A flood forecasting method based on a support vector machine model according to claim 2, characterized in that: In said S2, the step of determining the parameters of the flow generation module is: S21: Determine hydrological soil groups using soil type data from the study area and divide them according to infiltration capacity. Resample the land use types in the study area on the Arcgis platform to match them with the grids under each soil type. Calculate the CN value in each grid, then perform a weighted average of the CN values ​​for each grid to determine the overall CN value for the watershed. S22, classifies soil moisture conditions into three levels: dry, normal, and wet based on the total rainfall in the previous five days; K-mean clustering was used to re-divide the interval of soil moisture conditions. The total rainfall in the first five days was the pre-influencing rainfall, and the threshold was set at 120 mm. S23, convert the CN value according to the actual soil moisture conditions in the study area, interpolate the mutation intervals between CN1, CN2 and CN3, and obtain the CN value under different antecedent rainfall.

4. A flood forecasting method based on a support vector machine model according to claim 3, characterized in that: In S3, the improvement of the SCS-CN runoff model is to add the influence of rainfall characteristic factors on the basis of the original SCS-CN model. The specific steps are as follows: S31, in the runoff generation module, the previous rainfall needs to supplement the initial water deficit of the basin. If the total rainfall does not meet the initial water deficit, the basin will not generate runoff. The runoff generated by the rainfall event is calculated as follows: Where: P is the total rainfall, mm; Q is the surface runoff, mm; I a is the initial loss, mm; S is the potential water storage capacity, mm; S32, the calculation formula for initial loss and potential water storage capacity is: I a =λS (2) Where: λ is the initial loss rate; CN is a dimensionless parameter; S33, add characteristic factors such as rainfall intensity in the runoff generation module. The revised runoff calculation formula is as follows: Where: I 60 The maximum rainfall intensity in 60 minutes during the rainfall event, mm / h; is the average rainfall intensity of the rainfall event, mm / h; β is the rainfall intensity correction parameter; It is a comprehensive reflection of rainfall characteristic factors.

5. A flood forecasting system based on a support vector machine model, executing the flood forecasting method based on a support vector machine model according to any one of claims 1 to 4, characterized in that: include: Data collection unit: collects hydrological and meteorological data of the study area; Runoff module parameter solving unit: Determine the CN value of the runoff module parameter of the SCS-CN model based on the collected hydrological and meteorological data; Runoff module improvement unit: Use rainfall characteristic factors and runoff module parameter CN value to improve the runoff module of SCS-CN model; SCS-CN three-water source confluence model construction unit: By coupling the improved runoff generation module and the three-water source confluence module, the SCS-CN three-water source confluence model is established; SCS-CN three-source confluence model improvement unit: Improve the SCS-CN three-source confluence model by constructing dynamic confluence hysteresis parameters through the support vector machine model; Flood forecasting unit: Use measured rainfall data to forecast the flood process line at the outlet section of the basin.

Images