A distributed flood forecasting method for data-free areas considering the impact of dam failure of small and medium-sized reservoirs

Abstract

This invention belongs to the field of distributed flood forecasting technology and discloses a distributed flood forecasting method for data-free areas that considers the impact of small and medium-sized reservoir dam failures. Addressing the current problem of the lack of hydrological models specifically considering reservoir dam failures in data-free areas, this invention constructs a grid-based distributed hydrological model, including a data processing module, a reservoir module, a runoff generation module, and a confluence module. The reservoir module determines the dam failure or storage status based on the rainfall carrying capacity of a virtual reservoir; the runoff generation module calculates surface runoff and groundwater runoff based on the principle of mixed runoff generation, and determines whether to consider the storage effect of small and medium-sized reservoirs based on the results from the reservoir module; the confluence module uses the results from the runoff generation module as input and determines whether to consider the dam failure of small and medium-sized reservoirs based on the results from the reservoir module, thereby obtaining the flow processes of each grid and outlet in the watershed. This invention provides a flood forecasting method for data-free areas that considers the dam failure of small and medium-sized reservoirs, and is expected to reduce the uncertainty of flood forecasting.

Description

Technical Field

[0001] This invention relates to the field of distributed flood forecasting technology, and in particular to a distributed flood forecasting method for data-free areas that takes into account the impact of small and medium-sized reservoir dam failures. Background Technology

[0002] Floods are among the most common and destructive natural disasters globally, especially in data-scarce areas where their prediction and management face enormous challenges. Small and medium-sized reservoirs, as crucial flood control facilities in these regions, pose a significant threat to downstream areas if they fail, and further exacerbate the uncertainty of flood forecasting. Therefore, a distributed flood forecasting method that can effectively account for the dam failure effects of small and medium-sized reservoirs in data-scarce areas is urgently needed.

[0003] Distributed flood forecasting models have significant advantages in simulating watershed rainfall-runoff processes. They can divide the watershed into several sub-regions and consider the influence of spatially heterogeneous factors such as rainfall, topography, and soil conditions within each sub-region. By incorporating the dam-break behavior of small and medium-sized reservoirs into distributed flood forecasting models, not only can the spatial resolution of the model be improved, but the nonlinear characteristics of the dam-break flood propagation process can also be captured more accurately, thereby significantly improving the accuracy of flood forecasts.

[0004] Therefore, this invention aims to develop a distributed flood forecasting method for data-free areas that considers the impact of dam breaks in small and medium-sized reservoirs. By comprehensively applying distributed modeling technology and simulation methods for the storage, release, and dam break effects of small and medium-sized reservoirs, high-precision flood forecasting under the influence of small and medium-sized reservoirs can be achieved in data-free areas. This method not only provides important technical support for flood control and disaster reduction in data-free areas but also provides a scientific basis for the safe management of small and medium-sized reservoirs. Summary of the Invention

[0005] To address the problems of existing technologies, this invention provides a distributed flood forecasting method for data-free areas that takes into account the impact of dam failures in small and medium-sized reservoirs.

[0006] The technical solution of the present invention is as follows: a distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures, which involves constructing and training a grid-based distributed flood forecasting model to perform distributed flood forecasting; the grid-based distributed flood forecasting model includes a data processing module, a reservoir module, a runoff generation module, and a runoff confluence module;

[0007] The data processing module divides the watershed into several grid units, and the data recorded by each station in the watershed is divided into grid format according to the corresponding grid unit.

[0008] The reservoir module uses a clustering method to aggregate numerous small and medium-sized reservoirs into a virtual reservoir and generalize its spatial location. The rainwater carrying capacity of the virtual reservoir is used as the dam failure criterion. If the current rainwater carrying capacity of the virtual reservoir exceeds its design standard, the runoff module initiates dam failure calculation. Otherwise, the reservoir operates normally, and the runoff generation module initiates flood interception calculation.

[0009] The runoff generation module uses the mixed runoff generation principle to calculate the surface runoff and groundwater runoff of each grid cell. If flood storage calculation is activated, the regulation and storage effect of small and medium-sized reservoirs in the watershed on the flood process needs to be considered to obtain the actual runoff of each grid cell. The actual runoff is the sum of surface runoff and groundwater runoff.

[0010] The confluence module uses the actual flow rate calculated by the runoff generation module as input for confluence calculation. If dam break calculation is initiated, the impact of dam break of small and medium-sized reservoirs in the basin on the flood process needs to be considered. The runoff generation results of each grid cell are used to calculate the confluence based on the topological relationship of the water system. The confluence results are superimposed with the calculated dam break flow rate to finally obtain the flow process of all grids in the basin and the basin outlet.

[0011] The calculation process for the reservoir module is as follows:

[0012] The design rainfall of the watershed is calculated by simulating the runoff through the designated watershed outlet under different design rainfalls. Based on the current water storage status and design standards of the virtual reservoir, the design rainfall corresponding to the closest runoff is found to determine the reservoir's rain-carrying capacity and whether the dam-break condition is met. The design rainfall of the watershed is calculated using the following formula.

[0013]

[0014] Where z is the observed value of the random variable, σ is the scale parameter, μ is the location parameter, and x is the shape parameter.

[0015] The runoff generation module comprises four parts: evapotranspiration, surface runoff, groundwater runoff, and reservoir impoundment, as detailed below;

[0016] (1) Evapotranspiration calculation

[0017] Evapotranspiration in the runoff generation module includes canopy wetland evaporation, vegetation transpiration, and bare soil evaporation;

[0018] The evaporation of the wet part of the canopy E c Calculate according to the following formula;

[0019]

[0020] In the formula, f is the proportion of the time required for the canopy to retain water for evaporation; P2 is the rainfall intensity; △t is the calculation time step; W represents the maximum wetted part of the canopy evaporation. iTotal amount retained by the canopy; W im E represents the maximum retention capacity of the canopy. p To determine the surface evaporation potential with stomatal resistance set to zero; r w r0 is the aerodynamic impedance for moisture transport; r0 is the surface evaporation impedance.

[0021] The vegetation transpiration E t Calculate using the following formula;

[0022]

[0023] r c Stomatal impedance of the leaf surface;

[0024] The bare soil evaporation E l The calculation formula is as follows;

[0025]

[0026] In the formula, A S i represents the percentage of bare soil area saturated with water; i0 represents the water storage capacity at a certain point; A represents the percentage of area with a water storage capacity less than i; b represents the water storage shape parameter; E represents the water storage shape parameter. p For potential evaporation;

[0027] When the soil is not adequately watered, the actual evaporation of the soil is βEp, and the potential evaporation is calculated using the Penman-Monteith formula; β is a function of soil moisture; for water storage capacity, the water storage capacity distribution curve is used for calculation, as shown in the following formula;

[0028] i = i m [1-(1-A) 1 / b (5)

[0029] In the formula, i represents the water storage capacity; m is the maximum water storage capacity; A is the area proportion where the water storage capacity is less than i; b is the water storage shape parameter.

[0030] The surface runoff is calculated using water storage capacity distribution curves and infiltration capacity distribution curves to obtain the surface runoff process, which simultaneously considers the mechanisms of full storage runoff and infiltration runoff, as well as the impact of subgrid non-uniformity of soil properties on runoff.

[0031] The water storage capacity distribution curve is described by formula (4), and the infiltration capacity distribution curve is described as follows;

[0032] f'=f m [1-(1-C) 1 / B (7)

[0033] In the formula, f' represents the infiltration capacity; f mf' represents the maximum infiltration capacity; C is the area proportion where the infiltration capacity is less than or equal to f'; B is the shape parameter of the infiltration capacity.

[0034] The full-flow generation R1 occurs in the initial saturation area As and the portion that becomes saturated during the time period (A). s '-A s In terms of area, the excess runoff R2 occurs in the remaining area (1-As) and is redistributed within the entire area of ​​the excess runoff calculation; P represents the total rainfall over a period of time, including the saturated runoff R1, the excess runoff R2, and the total amount of water infiltrated into the soil ΔW, and the relationship between the three is as follows;

[0035] P = R1(y) + R2(y) + ΔW(y) (8)

[0036] y = R1(y) + △W(y) (9)

[0037] In the formula, y represents the vertical depth shown by the water storage capacity distribution;

[0038] According to formula (5), the formulas for calculating the runoff R1 and the change in soil moisture content ΔW are as follows;

[0039]

[0040] According to formula (12), the water input rate W is obtained. p The calculation formula is as follows;

[0041]

[0042] The excess runoff R2 is determined by the distribution curve of time period length versus infiltration capacity, soil infiltration capacity, and W. p The product of the areas enclosed by the three is calculated using the following formula;

[0043]

[0044] The vertical one-dimensional soil water movement is described using the ARNO model. The water vapor flux between different soil layers follows Darcy's law, and the groundwater runoff is calculated using the following formula.

[0045]

[0046] In the formula, D m D is the maximum base current; s For the current base current and D m The ratio; This represents the initial moisture content of the underlying soil. W represents the maximum water content of the underlying soil layer. s This represents the water content of the lower soil layer.

[0047] The reservoir impoundment is obtained by constructing a nonlinear equation between soil moisture content and reservoir water storage, thereby obtaining the reservoir water storage volume for each time period during the flood season. The specific formula is as follows:

[0048]

[0049] In the formula, V(t) is the reservoir capacity at time t; m and n are the linear and nonlinear parameters, respectively; W(t) is the soil moisture content at time t; W m V represents soil water storage capacity. e For the purpose of facilitating storage capacity; V d For dead storage capacity;

[0050] After deriving the time series of reservoir water storage during the flood, the reservoir capacity change value for each period is calculated according to ΔV(t)=V(t)-V(t-1), which represents the impact of the reservoir on the runoff of the basin. ΔV(t) is converted into the runoff above the reservoir to simulate the reservoir's water storage.

[0051] The confluence module comprises three parts: slope confluence, river confluence, and reservoir dam failure, and its calculation is as follows;

[0052] The slope runoff is calculated using a slope runoff unit line based on a two-parameter Gamma distribution. The shape of the unit line is controlled by the time scale parameter a and the shape parameter θ. The calculation formula for the Gamma distribution function is as follows;

[0053]

[0054] In the formula, t represents time; a is the time scale parameter of the distribution function; θ represents the shape parameter;

[0055] Using the results of the runoff generation module as input, the slope runoff at each time period is calculated using the Gamma distribution function. The calculation formula is as follows:

[0056]

[0057] In the formula, q is the flow rate at time step t; y t 'Indicates the flow rate calculated by the flow generation module; t max The maximum time duration of the Gamma distribution is represented by s; s represents the calculation period.

[0058] The flow rate of the river channel was calculated using the impulse response function method; the IRF is a one-dimensional diffusion wave equation derived from the one-dimensional Saint-Venant equation, and the calculation equation is as follows;

[0059]

[0060] In the formula, q is the flow rate of the cross-section; x is the distance along the river channel; C represents the flow velocity; D represents the diffusion coefficient; the river confluence flow rate is obtained by convolution integral of the equation, and the calculation formula is as follows;

[0061]

[0062] in,

[0063]

[0064] In the formula, U(ts) is the runoff depth generated at time ts;

[0065] The dam-break flow rate of the reservoir is calculated using the following formula;

[0066]

[0067] In the formula, H0 is the water level in front of the dam; Z is the dam length; g is the acceleration due to gravity; Q M The maximum flow rate at which the dam breaks is given; h = H0 - h', where h' is the residual height; and L is the breach length.

[0068] The dam-break flood process is generalized into a fourth-order parabola, and the formula for calculating the reservoir emptying time T is as follows;

[0069]

[0070] In the formula, W represents the reservoir capacity in case of dam failure; Q M The maximum flow rate at which the dam breaks is denoted as ; K is a coefficient;

[0071] The dam-break flow process is represented by t / T on the X-axis and Q / Q. M The parabola is approximated using t / T as the vertical axis; when t / T equals 0, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9, Q / Q M The corresponding values ​​are 1, 0.62, 0.48, 0.34, 0.26, 0.207, 0.168, 0.13, 0.094, 0.061, and 0.03, respectively. When t / T equals 1, Q / Q M The corresponding value is Q0 / Q M Q / Q M It is 0.

[0072] The formula for calculating the ulcer length L is as follows;

[0073] 1) When the reservoir capacity is greater than 1 million m³ 3 hour;

[0074]

[0075] In the formula, k is the dam material coefficient; for clay, clay core wall or inclined wall, soil, stone, and concrete, k is 1.19; for homogeneous loam, k is 1.98.

[0076] 2) When the storage capacity is less than 1 million m³ 3 hour;

[0077]

[0078] In the formula, if the dam material is good, k is taken as 6.6; otherwise, k is taken as 9.1.

[0079] The beneficial effects of this invention are as follows: Addressing the lack of measured reservoir data in data-scarce areas, this invention proposes a flood forecasting method. By establishing a distributed hydrological model with grid-based units, the propagation process and spatiotemporal characteristics of dam-break floods are simulated, providing scientific support for risk assessment and emergency management, while effectively reducing the uncertainty of flood forecasting. This method not only improves flood forecasting capabilities in data-scarce areas but also promotes the application of distributed hydrological models in simulating complex hydrological processes, demonstrating significant theoretical innovation and practical application value. Attached Figure Description

[0080] Figure 1 This is a flowchart of the present invention.

[0081] Figure 2 This is a schematic diagram of the runoff generation module of a distributed hydrological model; (a) is the water storage capacity curve, and (b) is the infiltration capacity curve.

[0082] Figure 3 This is a schematic diagram of the rain gauge stations and hydrological zones in the study area.

[0083] Figure 4 This is a schematic diagram showing the distribution of reservoirs in the study area.

[0084] Figure 5 This is a land use diagram of the study area.

[0085] Figure 6 This is a 12.5m resolution DEM elevation map of the study area.

[0086] Figure 7 This is a schematic diagram of vegetation utilization in the study area. Detailed Implementation

[0087] Based on distributed hydrological modeling, this invention proposes a distributed flood forecasting method for data-free areas that considers the impact of dam failures in small and medium-sized reservoirs.

[0088] The present invention will be further described below through embodiments and in conjunction with the accompanying drawings.

[0089] The Huanren Hydropower Station is the leading hydropower station in the Hunjiang River cascade project, located approximately 4 km upstream of Huanren Town, Huanren Manchu Autonomous County, Liaoning Province. The dam site controls a drainage area of ​​10,364 km². 2 It accounts for 67.2% of the Hunjiang River basin area. The main task of the power station is power generation, while also taking into account the comprehensive utilization of downstream flood control, irrigation, and aquaculture. It plays a role in peak shaving, frequency regulation, and emergency backup in the Northeast Power Grid.

[0090] The first step is data processing.

[0091] The model construction of this invention requires input data including watershed rainfall, soil, vegetation use, and DEM (Dual Element Model), as well as the location and reservoir capacity information of small and medium-sized water conservancy projects within the watershed. Specifically, this includes: DEM, reservoir distribution, soil type, vegetation type, precipitation, temperature, wind speed, and soil moisture content. DEM data of 12.5m resolution for the study area was downloaded from the 91weitu data download platform (https: / / www.91weitu.com / ); 1km resolution soil type spatial distribution data was used from the Food and Agriculture Organization of the United Nations (FAO); a global 1km resolution vegetation use dataset was used from the University of Maryland; and watershed rainfall, location and reservoir capacity information of small and medium-sized water conservancy projects were obtained from the watershed water management department. According to actual needs, the watershed was divided into several orthogonal grids of equal size, and underlying surface information such as soil properties, slope, river length, and river segment topology were extracted from the grid cells. The inverse distance weighting (IDW) method was used to interpolate the rainfall data from rain gauges into 0.05° grid data.

[0092] The second step, based on the data collected in the first step, divides the watershed above Huanren Reservoir into three parts: upper, middle and lower. Using the concept of aggregating reservoirs, the numerous small and medium-sized reservoirs in each part are aggregated into a virtual reservoir and its spatial location is generalized. The current rainwater carrying capacity of the reservoir is calculated and it is determined whether it exceeds its design standard. Then, a dam break is initiated in the runoff module or floodwater is impounded in the runoff generation module.

[0093] The third step involves calculating surface runoff using soil water storage capacity distribution curves and infiltration capacity distribution curves, and groundwater runoff using the ARNO model. If flood interception calculations are initiated, the interception capacity of small and medium-sized reservoirs during flood events is calculated, thereby obtaining the actual runoff for each grid.

[0094] The fourth step involves using the results from the runoff generation module as input, performing slope runoff calculations using unit lines based on the Gamma distribution, and performing river runoff calculations using the IRF method. If dam-break calculations are initiated, the flow process of small and medium-sized reservoirs within the basin is calculated. The upstream runoff calculation results of the reservoirs are then superimposed with the dam-break flow process to finally obtain the flow process of all grids within the basin and the basin outlet.

[0095] The fifth step involves discussing the inflow of Huanren Reservoir under the condition of dam failure in virtual reservoirs at different spatial locations. The dam failure flow process of the three virtual reservoirs above Huanren Reservoir is calculated under the condition of extreme precipitation. The results are then superimposed with the distributed flood forecast results of the designated sub-basin to obtain the flow process of all grids in the basin and the basin outlet under different extreme conditions.

[0096] Table 1. Flood inflow information for Huanren Reservoir under different scenarios.

[0097]

[0098] The above-described embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. A distributed flood forecasting method for data-free areas considering the impact of dam failures in small and medium-sized reservoirs, characterized in that, A grid-based distributed flood forecasting model is constructed and trained to perform distributed flood forecasting; the grid-based distributed flood forecasting model includes a data processing module, a reservoir module, a runoff generation module, and a runoff confluence module. The data processing module divides the watershed into several grid units, and the data recorded by each station in the watershed is divided into grid format according to the corresponding grid unit. The reservoir module uses a clustering method to aggregate numerous small and medium-sized reservoirs into a virtual reservoir and generalize its spatial location. The rainwater carrying capacity of the virtual reservoir is used as the dam failure criterion. If the current rainwater carrying capacity of the virtual reservoir exceeds its design standard, the runoff module initiates dam failure calculation. Otherwise, the reservoir operates normally, and the runoff generation module initiates flood interception calculation. The runoff generation module uses the mixed runoff generation principle to calculate the surface runoff and groundwater runoff of each grid cell. If flood storage calculation is activated, the regulation and storage effect of small and medium-sized reservoirs in the watershed on the flood process needs to be considered to obtain the actual runoff of each grid cell. The actual runoff is the sum of surface runoff and groundwater runoff; The confluence module uses the actual flow rate calculated by the runoff generation module as input for confluence calculation. If dam break calculation is initiated, the impact of dam break of small and medium-sized reservoirs in the basin on the flood process needs to be considered. The runoff generation results of each grid cell are used to calculate the confluence based on the topological relationship of the water system. The confluence results are superimposed with the calculated dam break flow rate to finally obtain the flow process of all grids in the basin and the basin outlet.

2. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 1, characterized in that, The calculation process for the reservoir module is as follows: The design rainfall of the watershed is calculated by simulating the runoff through the designated watershed outlet under different design rainfalls. Based on the current water storage status and design standards of the virtual reservoir, the design rainfall corresponding to the closest runoff is found to determine the reservoir's rain-carrying capacity and whether the dam-break condition is met. The design rainfall of the watershed is calculated using the following formula. （1）； Where z is the observed value of the random variable, For scale parameters, For position parameters, For shape parameters.

3. A distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures, as described in claim 1 or 2, characterized in that... The runoff generation module comprises four parts: evapotranspiration, surface runoff, groundwater runoff, and reservoir impoundment, as detailed below; The evaporation calculation is as follows; Evapotranspiration in the runoff generation module includes canopy wetland evaporation, vegetation transpiration, and bare soil evaporation; The evaporation of the wet part of the canopy E c Calculate according to the following formula; （2）； （3）； In the formula, f is the proportion of time required for canopy water interception and evaporation; P2 is the rainfall intensity; To calculate the time step; This represents the maximum evaporation rate of the wetted part of the canopy. W i W represents the total amount retained by the canopy. im E represents the maximum retention capacity of the canopy. p To determine the surface evaporation potential with stomatal resistance set to zero; r w r0 is the aerodynamic impedance for moisture transport; r0 is the surface evaporation impedance. The vegetation transpiration E t Calculate using the following formula; （4）； Stomatal impedance of the leaf surface; The bare soil evaporation E l The calculation formula is as follows; （5）； In the formula, A S i represents the percentage of bare soil area saturated with water; i0 represents the water storage capacity at a certain point. For the proportion of areas with a water storage capacity less than i, For water storage shape parameters; When the soil is not adequately watered, the actual evaporation of the soil is Potential evaporation was calculated using the Penman-Monteith formula; It is a function of soil moisture; for water storage capacity, it is calculated using the water storage capacity distribution curve, as shown in the following formula; （6）； In the formula, i represents the water storage capacity; m is the maximum water storage capacity; A is the area proportion where the water storage capacity is less than i; b is the water storage shape parameter.

4. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 3, characterized in that, The surface runoff is calculated using water storage capacity distribution curves and infiltration capacity distribution curves to obtain the surface runoff process, which simultaneously considers the mechanisms of full storage runoff and infiltration runoff, as well as the impact of subgrid non-uniformity of soil properties on runoff. The water storage capacity distribution curve is described by formula (4), and the infiltration capacity distribution curve is described as follows; （7）； In the formula, f' represents the infiltration capacity; f m f' represents the maximum infiltration capacity; C is the area proportion where the infiltration capacity is less than or equal to f'; B is the shape parameter of the infiltration capacity. The full flow generation R1 occurs in the initial saturated area As and the portion that becomes saturated during the time period. In terms of area, the excess infiltration runoff R2 occurs over the remaining area (1-As) and is redistributed across the entire area of ​​the excess infiltration runoff calculation; P represents the total rainfall over a period of time, including the saturated runoff R1, the excess infiltration runoff R2, and the total amount of water infiltrated into the soil. The relationship between the three is as follows; （8）； （9）； In the formula, y represents the vertical depth shown by the water storage capacity distribution; According to formula (5), the runoff generation R1 and the change in soil moisture content are calculated. The calculation formulas are as follows; （10）； （11）； According to formula (12), the water input rate W is obtained. p The calculation formula is as follows; （12）； The excess runoff R2 is determined by the distribution curve of time period length versus infiltration capacity, soil infiltration capacity, and W. p The product of the areas enclosed by the three is calculated using the following formula; （13）。 5. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 4, characterized in that, The vertical one-dimensional soil water movement is described using the ARNO model. The water vapor flux between different soil layers follows Darcy's law, and the groundwater runoff is calculated using the following formula. （14）； In the formula, D m D is the maximum base current; s For the current base current and D m The ratio; This represents the initial moisture content of the underlying soil. W represents the maximum water content of the underlying soil layer. s This represents the water content of the lower soil layer.

6. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 5, characterized in that, The reservoir impoundment is obtained by constructing a nonlinear equation between soil moisture content and reservoir water storage, thereby obtaining the reservoir water storage volume for each time period during the flood season. The specific formula is as follows: （15）； In the formula, Let be the reservoir capacity value during time period t; m and n are linear and nonlinear parameters, respectively; Let t be the soil moisture content during time period t; Soil water storage capacity; To improve storage capacity; For dead storage capacity; After deriving the time series of reservoir water storage during the flood, based on The changes in reservoir capacity over different time periods are calculated, representing the impact of the reservoir on watershed runoff. The flow generated above the reservoir is converted to simulate the reservoir's water retention capacity.

7. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 1, characterized in that, The confluence module comprises three parts: slope confluence, river confluence, and reservoir dam failure, and its calculation is as follows; The slope runoff is calculated using a slope runoff unit line based on a two-parameter Gamma distribution. The shape of the unit line is controlled by the time scale parameter a and the shape parameter θ. The calculation formula for the Gamma distribution function is as follows; （16）； In the formula, t represents time; a is the time scale parameter of the distribution function; θ represents the shape parameter. Using the results of the runoff generation module as input, the slope runoff at each time period is calculated using the Gamma distribution function. The calculation formula is as follows: （17）； In the formula, q is the flow rate at time step t; y t 'Indicates the flow rate calculated by the flow generation module; t max Indicates the maximum time length of the Gamma distribution; s represents the calculation period.

8. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 7, characterized in that, The flow rate of the river confluence is calculated using the impulse response function method; the IRF is a one-dimensional diffuse wave equation derived from the one-dimensional Saint-Venant equation, and the calculation equation is as follows; （18）； In the formula, q is the flow rate of the cross-section; x is the distance along the river channel; C represents the flow velocity; D represents the diffusion coefficient; the river confluence flow rate is obtained by convolution integral of the equation, and the calculation formula is as follows; （19）； in, （20）； In the formula, U(ts) is the runoff depth generated at time ts.

9. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 7, characterized in that, The flow rate of the reservoir dam failure is calculated using the following formula; （21）； In the formula, The water level in front of the dam is taken as Z; the dam length is Z; and g is the acceleration due to gravity. The maximum flow rate at which the dam breaks; ,in L represents the residual height; L represents the ulcer length. The dam-break flood process is generalized into a fourth-order parabola, and the formula for calculating the reservoir emptying time T is as follows; （22）； In the formula, W represents the reservoir capacity in case of dam failure; The maximum flow rate at which the dam breaks is denoted as ; K is a coefficient; The dam-break flow process is represented by t / T on the X-axis and Q / Q. M The parabola is approximated using t / T as the vertical axis; when t / T equals 0, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9, Q / Q M The corresponding values ​​are 1, 0.62, 0.48, 0.34, 0.26, 0.207, 0.168, 0.13, 0.094, 0.061, and 0.03, respectively. When t / T equals 1, Q / Q M The corresponding value is Q0 / Q M Q / Q M is 0.

10. The distributed flood forecasting method for data-free areas considering the impact of small and medium-sized reservoir dam failures according to claim 9, characterized in that, The formula for calculating the ulcer length L is as follows; 1) When the reservoir capacity is greater than 1 million m³; （23）； In the formula, k is the dam material coefficient; for clay, clay core wall or inclined wall, soil, stone, and concrete, k is 1.19; for homogeneous loam, k is 1.

98. 2) When the storage capacity is less than 1 million m³; （24）； In the formula, if the dam material is good, k is taken as 6.6; otherwise, k is taken as 9.1.

Images