CN113821999B - A Applicability Assessment Method of Hydrological Models for Inferring Design Floods - Google Patents

Abstract

The invention relates to a hydrological model applicability evaluation method for calculating design flood, which comprises the following steps: step 1, collecting and processing typical rainfall flood process data; step 2, selecting and rating a conceptual hydrological model; step 3, judging the applicability of the hydrological model in the research basin; step 4, substituting design rainstorm to obtain design flood; and 5, comparing the instantaneous unit line method and the original design result of the engineering to participate in the evidence. The method improves the empirical generalization in the calculation of the original design flood, analyzes the production convergence characteristic of the drainage basin from the drainage basin rainfall runoff yield theory, and deduces the design flood from a physical mechanism based on the applicability of the hydrological model. The method improves a design flood theory system, and is a novel method for calculating and designing the flood, which can be used for popularization and application.

Description

A Method for Applicability Assessment of Hydrological Models Used to Calculate Design Flood

technical field

The application date of this invention is November 21, 2017, and the application number is: 201711161934.3. It is a divisional application of an invention patent application titled "A Method for Calculating Flood in Watershed Design Based on a Conceptual Hydrological Model". The invention relates to the technical field of flood risk management, in particular to a flood estimation method for watershed design based on a conceptual hydrological model.

Background technique

Except for large watersheds or typical hydrological stations, the flood forecasting process in general watersheds and hydrological stations in my country still stays in the method of constructing the correlation relationship between rainfall and runoff. Correspondingly, there are relatively few theoretical achievements in designing floods. The reasoning formula method for small watersheds and the instantaneous unit line method with better universality. However, these two methods are only based on the statistical relationship of historical rainfall-runoff data, and have a certain degree of subjectivity in the derivation process of design floods. With the acceleration of the modernization process of water conservancy, the acquisition of water regime information is gradually improved, and the calculation of design flood also requires more objective and accurate methods.

The hydrological model is the product of people's deepening understanding of the water cycle process and flow generation and confluence in the basin, and the conceptual hydrological model is an important branch of it. The conceptual hydrological model, which is different from the black box model, is characterized in that it can describe the entire rainfall-runoff conversion process, and pays more attention to the physical meaning of parameters in the model structure. Different from the distributed and semi-distributed models that are in the exploratory stage, the theoretical basis of conceptual hydrological models is relatively mature, and has been widely used in flood risk management and forecasting operations around the world.

Contents of the invention

The present invention designs a hydrological model applicability evaluation method for deriving design floods, which solves the technical problem that the calculation results are distorted based on subjective experience in the current design flood derivation method and cannot reflect the limitations of real flood conditions.

In order to solve the above-mentioned technical problems, the present invention adopts the following scheme:

A flood derivation method for watershed design based on a conceptual hydrological model, including the following steps:

Step 1. Data collection and processing of typical storm and flood processes;

Step 2. Select and calibrate the conceptual hydrological model;

Step 3. Judging the applicability of the hydrological model in the research basin;

Step 4. Substituting the design rainstorm to obtain the design flood;

Step 5. Comparing the instantaneous unit line method and the original design results of the project for reference.

Further, the typical rain and flood process data collected in step 1 include daily rainfall data and time-period rainfall data, flood flow data, and evaporation data from rainfall stations in the watershed. Analyze the rationality of flood rainfall data, evaporation data and flood flow data, and delete unsuitable rain and flood processes;

Wherein, the daily rainfall data is used to calculate the previous impact rainfall and the rainfall of the previous five days;

Rainfall data for the period described are used as the main input of the hydrological model;

For the flood discharge data, select the flood process with a large peak height, and select a small flood process with a better peak shape appropriately. The selected small flood events should be controlled within 20%, and the number of floods used for simulation should be increased as much as possible if the data allow The number of events; the division of flood magnitude can be divided according to the flood return period or the magnitude of the flood peak in the basin; among them, the flood with level 1 is called a small flood, the level 2 is called a general flood, the level 3 is called a large flood, and the level 4 is called a large flood. Level 5 is called a major flood, level 5 is called a catastrophic flood, and level 6 or above is called an extraordinary flood;

For watersheds lacking data, the evaporation data is based on the data of nearby evaporation stations or based on the average evaporation of the upper watershed.

Further, the flow data collected in step 1 is preferentially used from the flow data of the participating hydrological station before entering the reservoir. If the participating station is downstream of the reservoir, the flow of the participating station needs to be restored and calculated.

Further, the reduction calculation adopts the water balance method, and the water balance method is to calculate the inflow flow by checking the relationship between the water level, storage, storage and discharge of the built reservoir, and the restored flow is the average inflow flow during the calculation period; the inflow flow is according to the following formula 1 and Equation 2 calculate:

In the formula: Δt is the calculation period length, the unit is s;

q _t is the average outbound flow rate of the time period, the unit is m ³ /s;

Q _t is the average inflow flow in a time period, the unit is m ³ /s;

V _t , V _t+1 is the storage capacity of the reservoir at the beginning and end of the period, and the unit is m ³ ;

q _t , q _t+1 is the outbound flow at the beginning and end of the period, and the unit is m ³ /s.

Further, the step 2 comprehensively considers the different rainfall and runoff characteristics of the watershed, and selects any one of the Xin'anjiang model, the water tank model or the SCS model as the most widely used conceptual hydrological model with simple parameters as an alternative.

Further, in step 3, the rainfall data, flood flow data and evaporation data collected and processed in step 1 are substituted into step 2 to select and calibrate the conceptual hydrological model for simulation, and the simulation results are compared with the measured flow process data, The relative error of simulated flood peak, flood volume and certainty coefficient are used as the evaluation criteria for the applicability of the model in the application basin.

Further, the relative error in the step 3 is calculated according to the following formula 3 and formula 4:

In the formula: ε ₁ ——relative error, %; _Qreal ——actual flood peak flow, m ³ /s; Q _model ——simulated peak flow, m ³ /s;

In the formula: _ε2 is the relative error, %; R _is the actual quantity, mm; R _model is the simulated quantity, mm;

In the step 3, the coefficient of certainty is calculated by formula 5:

为实测值的均值；n为资料序列长度；以相对误差30％内，确定性系数50％以上计算合格率并选择模拟效果最佳的模型作为流域设计洪水计算最优模型。In the formula: DC is the coefficient of certainty (taking 2 decimal places); y _0(i) is the measured value; y _c(i) is the predicted value;

is the mean value of the measured value; n is the length of the data sequence; the pass rate is calculated with a relative error of 30% and a certainty coefficient of more than 50%, and the model with the best simulation effect is selected as the optimal model for basin design flood calculation.

Further, in step 4, the design rainstorm is selected by using the regional rainfall pattern given in the hydrological atlas of the region or selecting historical typical floods (for example, the largest flood in the history of the basin with flow records), and using the ratio of rainfall to secondary floods during the period as the distributed rainfall. Type, the obtained time-period design rainstorm value is used as the input item of the optimal model selected in step 3, and the evaporation data is brought into the optimal model selected in step 3 for simulation using the average evaporation value of the flood season.

Furthermore, step 5 aims to further verify the conclusions obtained from the typical flood simulation in step 3. During engineering design, the design flood value obtained by the hydrological model algorithm and the traditional engineering algorithm can be cross-referenced and verified.

Further, in step 5, the instantaneous unit line method commonly used in designing floods in small and medium watersheds is selected, and the calculation data required by the instantaneous unit line method are obtained from the hydrological atlas of the corresponding area. The hydrological model algorithm is combined with the instantaneous unit line algorithm and engineering The original design flood values are cross-referenced and confirmed.

The flood calculation method for watershed design based on the conceptual hydrological model has the following beneficial effects:

The derivation method involved in the present invention selects a conceptual hydrological model with perfect theory and wide application, and applies it to watersheds with different rainfall and runoff characteristics to simulate the design flood, which changes the empirical generalization in the original design flood calculation. From the perspective of runoff production theory, analyzing the flow generation and confluence mechanism of the watershed, and calculating the design flood from the physical process improves the accuracy of the design flood, which is conducive to improving the theoretical system of the design flood. It is a new method for calculating the design flood that can be extended.

Description of drawings

Fig. 1: The present invention is based on the flow chart of the design flood calculation method of small and medium watersheds based on the conceptual hydrological model;

Fig. 2: the present invention divides the flood map according to the flood return period;

Fig. 3: the present invention divides the flood map according to the flood peak;

Fig. 4: the Xin'anjiang model structure of two water sources used in the present invention;

Fig. 5: the water tank model structure that uses among the present invention;

Figure 6: SCS model structure used in the present invention;

Fig. 7: the typical flood simulation result figure of 3 kinds of models in verification period in the present invention;

Fig. 8: Results of watershed design flood simulation in the present invention;

Figure 9: Schematic diagram of a comparison table of flood peak and runoff in a watershed once in a century in the present invention.

Detailed ways

Below in conjunction with Fig. 1 to Fig. 9, the present invention will be further described:

The present invention improves the empirical generalization of the original design flood calculation method, starts from the basin rainfall runoff theory, analyzes the flow generation and confluence characteristics of the basin, and calculates the design flood from the physical mechanism based on the applicability of the model. This method perfects the theoretical system of design flood and is a new method for calculating design flood that can be used for popularization and application.

As shown in Figure 1, a conceptual hydrological model-based flood calculation method for watershed design includes the following steps: Step 1, data collection and processing of typical storm-flood processes; Step 2, selection and calibration of conceptual hydrological models; Step 3. Judging the applicability of the hydrological model in the research basin; Step 4. Substituting the design storm to obtain the design flood; Step 5. Comparing the instantaneous unit line method and the original design results of the project for reference.

Embodiment: select a watershed, and adopt the method provided by the present invention to simulate the design flood process once in a hundred years, specifically comprising the following steps:

Step 1. Select the daily rainfall and time-period rainfall data and the flow data from 1980 to 2012 in a certain watershed. The evaporation data is the evaporation data from 1980 to 2012 from the evaporation stations in the watershed, and the time period is 2 hours; Downstream, before the actual flood simulation, the initial flow of the reservoir was restored according to Formula 1 and Formula 2, and a total of 16 flood events with large flood volume and good continuity were selected as calibration and verification data for the hydrological model. Due to the small area of the watershed and limited historical data, the floods at the survey stations in the watershed have large errors in classification according to the return period. Using Figure 3 to classify according to the flood peak, and after converting according to the watershed area, we selected 3 major floods of level 3 and general floods of level 2 10 games, 3 games of level 1 small floods.

As shown in Figure 2, the division of flood magnitude can also be divided according to the flood return period; the flood with level 1 is called a small flood, the level 2 is called a general flood, the level 3 is called a large flood, and the level 4 is called a flood. It is a severe flood, level 5 is called a catastrophic flood, and level 6 or above is called an extraordinary flood.

Step 2. According to the characteristics of rainfall and runoff in the basin, three hydrological models are selected: Xin'anjiang model with two water sources, water tank model and SCS model. The 13 floods in step 1 were used to calibrate the parameters of the hydrological model. Among them, the Xin'anjiang model adopts a two-water source structure, and the structure is shown in Figure 4. The water tank model uses two layers of water tanks connected in series, and the structure is shown in Figure 5. The structure of the SCS model is shown in Figure 6.

Specifically, the water source part of the Xin'anjiang model selected by the present invention is divided into surface and underground water sources by stabilizing the infiltration rate. flood process. WU, WL, and WD represent the initial upper-layer tension water storage, the lower-layer tension water storage, and the deep-layer tension water storage respectively; EU, EL, and ED represent the upper-layer evapotranspiration, lower-layer evapotranspiration, and deep-layer evapotranspiration, respectively. Emission; IMP is the ratio of impermeable area to the total watershed area; Ew is the evaporation of the watershed during a period.

The water tank model adopts a series structure of two layers of water tanks. In the upper water tank: the upper hole simulates ground runoff, the lower hole simulates soil flow, and the bottom hole is used to simulate the infiltration of water; in the lower tank: the side hole simulates underground runoff, and the bottom hole is used to simulate The amount of infiltrated water (the bottom hole of the lower water tank in a large closed basin can be closed). In the model, P is the period rainfall, E is the period evaporation, x is the storage depth of the water tank, y is the period runoff, z is the period infiltration, and α and β are the corresponding outflow coefficients.

In the SCS model, S is the possible retention of the watershed at that time, mm; I _a is the initial loss, CN is the curve index, and F is the area of the watershed.

In step 3, the rainfall and evaporation data collected and processed in step 1 are respectively substituted into the three hydrological models calibrated in step 2 for simulation. Select the remaining three floods in step 1 to verify the simulation effect of the model. The verification results of the two floods with better simulation effects are shown in Figure 5. Calculate the relative error and certainty coefficient according to formulas 3, 4 and 5, calculate the qualified rate with the peak error within 30% and the certainty coefficient above 50%, and select the model with the best simulation effect as the optimal model for basin design flood calculation. The pass rate of the simulated Xin'anjiang model is 92.3%, the pass rate of the water tank model is 64.2%, and the pass rate of the SCS model is 50%. Therefore, the Xin'anjiang model is selected as the optimal model for basin design flood calculation.

Step 4. Select the 1-in-100-year rainfall pattern given in the hydrological atlas, and use the obtained time-period design rainstorm value as the input of the optimal model selected in step 4. The design flood encountered is deduced, and the results are shown in Figure 6.

It can be seen from Figure 7 that when simulating the once-in-a-century flood process in the basin, the original design value of the project is used as the standard to evaluate the design flood calculation results of the Xin'anjiang model, and the flood peak error and flood volume error are -5.3% and -5.4% respectively %, the error is controlled within 10%, both of which are smaller than the calculation results of the instantaneous unit line method -13.4% and 14.3%.

Accompanying drawing has carried out exemplary description to the present invention, and obviously the realization of the present invention is not limited by above-mentioned mode, as long as adopt the various improvements that the method concept of the present invention and technical scheme are carried out, or the conception of the present invention is unimproved And technical solutions that are directly applied to other occasions are within the protection scope of the present invention.

Claims (3)

1. A hydrological model suitability assessment method for use in inferring design floods, comprising the steps of:

step 1, collecting and processing typical rainfall flood process data;

the flow data collected in the step 1 preferably adopts the flow data of the reference hydrological station before warehousing, and if the reference hydrological station is located at the downstream of the reservoir, the flow of the reference hydrological station needs to be subjected to reduction calculation processing;

step 2, selecting and rating a conceptual hydrological model;

step 3, judging the applicability of the hydrological model in the research basin;

in the step 3, the rainfall data, the flood flow data and the evaporation data collected and processed in the step 1 are substituted into the conceptual hydrological model selected and calibrated in the step 2 for simulation, a simulation result is compared with actually measured flow process data, and a simulated flood peak, a flood relative error and a certainty coefficient are used as evaluation standards of the applicability of the model in the application basin; calculating the qualified rate by using the relative error within 30% and the certainty coefficient of more than 50%, and selecting a model with the best simulation effect as an optimal model for basin design flood calculation;

step 4, substituting design rainstorm to obtain design flood;

the selection of the rainstorm designed in the step 4 is transferred to the regional rain type given in the regional hydrological chart set or the selected historical typical field sub-flood, the drainage basin historical maximum flood with the flow record is used, the time-period rainfall sub-flood ratio is used as the distribution rain type, the obtained time-period design rainstorm value is used as the input item of the optimal model screened in the step 3, and the evaporation data is simulated by adopting the average time-period evaporation value in the flood season and being brought into the optimal model screened in the step 3;

step 5, comparing the instantaneous unit line method and the original design result of the engineering to participate in the certification;

and 5, further verifying the conclusion obtained by the typical flood simulation in the step 3, wherein the design flood value obtained by the hydrological model algorithm and the engineering traditional algorithm can be mutually referred to and verified in engineering design.

2. The method of claim 1, wherein the method comprises: the typical rainfall flood process data collected in the step 1 comprise daily rainfall data and time-interval rainfall data of rainfall stations in the drainage basin, flood flow data and evaporation data, the rationality of the rainfall data, the evaporation data and the flood flow data of the selected drainage basin typical flood is analyzed according to the lag and correlation among the rainfall flows, and the non-conforming rainfall flood process is deleted;

wherein the daily rainfall data is used for calculating early-stage influence rainfall and rainfall in the previous five days;

the rainfall data in the time period is used as a main input item of the hydrological model;

the flood flow data selects a flood process with a large peak height, and can properly select a small flood process with a better peak type, the selected small flood field number is controlled within 20 percent, and the flood field number used for simulation is increased as much as possible under the permission of the data; the division of the flood magnitude can be divided according to the magnitude of a flood recurrence period or a flood peak in a drainage basin; wherein, flood with grade 1 is called small flood, grade 2 is called general flood, grade 3 is called larger flood, grade 4 is called large flood, grade 5 is called super large flood, grade 6 is called extraordinary flood;

for the watershed with the lack of data, the evaporation data uses the data of the adjacent evaporation station or the average evaporation condition of the upper watershed.

3. The method of claim 1, wherein the method comprises: the reduction calculation in the step 1 adopts a water quantity balance method, the water quantity balance method is to calculate the warehousing flow by checking the storage and discharge relation of the water level reservoir of the built reservoir, and the flow after reduction is the average warehousing flow in the calculation period; the warehousing flow rate is calculated according to the following formula 1 and formula 2:

formula 1;

formula 2;

in the formula:Δtfor calculating the time period length, the unit is s;

is the average warehouse-out flow in m³/s；

Q _t Is the average warehousing flow in a time interval and has the unit of m³/s；

V _t ，V _t+1 The storage capacity of the reservoir is at the beginning and the end of the time interval and the unit is m³；

q _t ，q _t+1 The flow rate of the warehouse outlet at the beginning and the end of the time interval is m³/s。

Images