Nested watershed hydrological model parameter optimization method

Abstract

The invention discloses a nested watershed hydrological model parameter optimization method, which comprises the following steps of: determining a research area of a watershed, layering the research area by utilizing elevation in the research area, and obtaining n sub-areas with different elevation levels; constructing a hydrological model equation of each elevation level sub-region; constructing a vector about dynamic parameters, performing crossover variation on the dynamic parameters in the vector by using a fitness function, and screening an optimal offspring individual vector; and sequentially optimizing the dynamic parameters of the hydrological model equation of each elevation level sub-region from top to bottom from the elevation level sub-region with the maximum elevation until the optimization of the dynamic parameters of the hydrological model equation of the research region is completed. According to the nested type watershed hydrological model parameter optimization method provided by the invention, the optimization time required by watershed parameter optimization is greatly shortened, and the parameter optimization efficiency is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of hydrological research, and in particular to a method for optimizing parameters of a nested watershed hydrological model. Background Art

[0002] In the process of hydrological research, the use of hydrological models to predict hydrological data is the key, and the optimization of model parameters is the key content of hydrological simulation research and application. The quality of model parameters directly affects the accuracy of model calculation results. How to make full use of existing observation data and improve optimization efficiency has always been an important research direction. At present, most studies only use data from export stations for parameter calibration, which makes it difficult to ensure the accuracy of hydrological forecasts. For basins with complex nested structures, existing methods often find it difficult to accurately consider the impact of interactions between basins at different levels on parameters, resulting in low accuracy of model simulation results, low efficiency when processing a large number of parameters, and easy to fall into local optimal solutions. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for optimizing parameters of a nested watershed hydrological model.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A nested watershed hydrological model parameter optimization method is provided, which comprises the following steps:

[0006] S1: Determine the study area of ​​the watershed, and stratify the study area using elevation to obtain n sub-areas of different elevation levels;

[0007] S2: construct the hydrological model equations for each elevation level sub-region;

[0008] S3: Extract the dynamic parameters that affect the calculation results of the hydrological model equation, construct a vector of the dynamic parameters, use the fitness function to perform crossover mutation on the dynamic parameters in the vector, screen the optimal offspring individual vector, use the individuals in the optimal offspring individual vector as the optimized dynamic parameters, and output the optimized dynamic parameters to update the hydrological model equation;

[0009] S4: According to the n elevation level sub-areas divided in the study area, step S3 is executed repeatedly, and the dynamic parameters of the hydrological model equation of each elevation level sub-area are optimized from top to bottom starting from the elevation level sub-area with the largest elevation, until the dynamic parameters of the hydrological model equation of the study area are optimized.

[0010] Furthermore, step S1 includes: determining the study area of ​​the watershed, obtaining the elevation interval [h min ,h max], set the number n of elevation level sub-areas, and calculate the elevation of each elevation level sub-area

[0011] Furthermore, the hydrological model equation for each elevation level sub-region in step S2 is:

[0012]

[0013] Among them, i is the number of the elevation level sub-area, t is the statistical time of the hydrological data, V i (t) is the water storage capacity at time t, P i (t) is the precipitation input at time t, E i (t) is the evaporation at time t, Qout,i(t) is the outflow at time t, j is the inflow path, Q in,j (t) is the inflow at time t, J i is the number of inflow pathways, P 0,i (t) is direct precipitation, α i is the precipitation distribution coefficient, k is the distribution path of the sub-region of the previous elevation level, K is the total amount of the distribution path, P k (t) is the outflow of distribution path k at time t, αk is the distribution coefficient of the sub-region of the previous elevation level, ΔR i is the radiation value received, ρ is the air density, c is the specific heat of air at constant pressure, e1 is the saturated water vapor pressure, e0 is the actual water vapor pressure, Δk is the slope of the saturated water vapor pressure-temperature curve, r′ is the air resistance, r0 is the surface resistance, γ is the resistance coefficient, β i is the influence coefficient of vegetation, n i is the Manning roughness coefficient, A i is the cross-sectional area of ​​water flow, R i is the hydraulic radius, S i is the slope of the water surface in the basin.

[0014] Further, step S3 includes:

[0015] S31: Extract dynamic parameters that affect the calculation results of the hydrological model equation, including the precipitation distribution coefficient α i , distribution coefficient α k , slope Δk, resistance coefficient γ, influence coefficient β i ; Construct a vector X about dynamic parameters, the number of elements in the vector X is m=5;

[0016] S32: construct a fitness function based on the water storage V(t,X) calculated by the hydrological model equation determined by the elements in the vector X and the actual water storage V0(t) at time t in the elevation level sub-area;

[0017]

[0018] Among them, F(X) is the fitness value calculated based on vector X, and T is the basin hydrological observation period; the fitness value represents the accuracy of the dynamic parameter genetic iteration process and is used to evaluate the adaptability of the genetic iteration process;

[0019] S33: During the observation period, based on the historical water storage observation and calculation results, construct any two individual vectors of the elements in vector X The observation times of two individual vectors are different, and the individuals in any two individual vectors are regarded as two different parents. is the mth dynamic parameter in the first parent, is the mth dynamic parameter in the second parent;

[0020] S34: Determine a crossover point u in the two individual vectors, and obtain a descendant individual vector after the crossover;

[0021]

[0022] in, is the kth individual in the first parent generation, k is the crossover point, is the kth individual in the second parent generation;

[0023] S35: setting a range for random mutation of each individual in the offspring individual vector, randomly selecting two corresponding offspring individuals in the two offspring individual vectors x′1 and x′2 to perform random mutation within the range of random mutation, substituting the two offspring individual vectors x′1 and x′2 after mutation into the hydrological model equation, calculating the fitness value, and calculating the acceptance probability p of the offspring individual vectors x′1 and x′2 after mutation according to the fitness value;

[0024]

[0025] Among them, w is the mutation iteration coefficient. The more mutation iterations there are, the smaller the mutation iteration coefficient is.

[0026] When p = 1, the offspring individual vector corresponding to the largest fitness value in F(x′1) and F(x′2) is selected as the optimal offspring individual vector;

[0027] When p≠1, set the receiving probability threshold p 阈值 ;

[0028] If p≤p 阈值 , then the corresponding offspring individual vector is taken as the optimal offspring individual vector;

[0029] Otherwise, re-mutate the offspring individuals in the offspring individual vectors x′1 and x′2 until the optimal offspring individual vector is output;

[0030] S36: The individuals in the optimal offspring individual vector are used as optimized dynamic parameters, and the optimized dynamic parameters are output to update the hydrological model equation.

[0031] The beneficial effects of the present invention are as follows: the present invention proposes a nested basin hydrological model parameter optimization method, which processes the study area of ​​the nested basin in layers to form multiple sub-areas with different elevations, establishes corresponding hydrological model equations for each layer of basin sub-areas, and iteratively optimizes the dynamic parameters of the model through an optimization algorithm; effectively overcomes the deficiency of the existing hydrological model parameter optimization that only uses the observation data of the basin outlet hydrological station, makes full use of the existing effective historical measured data, obtains hydrological parameters that can better represent the hydrological characteristics of the basin, and the results obtained by simulating these parameters can better represent the hydrological process occurring in the basin, which has important practical significance. The optimization time required for basin parameter optimization is greatly shortened, and the efficiency of parameter optimization is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Flowchart of a nested watershed hydrological model parameter optimization method. DETAILED DESCRIPTION

[0033] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0034] like Figure 1 As shown, a nested watershed hydrological model parameter optimization method includes the following steps:

[0035] S1: Determine the study area of ​​the watershed, stratify the study area by elevation, and obtain sub-areas of different elevation levels; Step S1 specifically includes: determine the study area of ​​the watershed, obtain the elevation interval [h min ,h max ], set the number n of elevation level sub-areas, and calculate the elevation of each elevation level sub-area This stratification method can reflect the differences in hydrological changes in areas of different altitudes, because elevation affects processes such as precipitation and runoff. Precipitation in high-altitude areas may be more diverse (such as snowfall), and water flows from high altitudes to low altitudes under the action of gravity, and the confluence characteristics of different elevation layers are different.

[0036] S2: construct the hydrological model equations for each elevation level sub-region;

[0037]

[0038] Among them, i is the number of the elevation level sub-area, t is the statistical time of the hydrological data, V i (t) is the water storage capacity at time t, P i (t) is the precipitation input at time t, E i (t) is the evaporation at time t, Qout,i(t) is the outflow at time t, j is the inflow path, Q in,j (t) is the inflow at time t, J i is the number of inflow pathways, P 0,i (t) is direct precipitation, α i is the precipitation distribution coefficient, k is the distribution path of the sub-region of the previous elevation level, K is the total amount of the distribution path, P k (t) is the outflow of distribution path k at time t, αk is the distribution coefficient of the sub-region of the previous elevation level, ΔR i is the radiation value received, ρ is the air density, c is the specific heat of air at constant pressure, e1 is the saturated water vapor pressure, e0 is the actual water vapor pressure, Δk is the slope of the saturated water vapor pressure-temperature curve, r′ is the air resistance, r0 is the surface resistance, γ is the resistance coefficient, β i is the influence coefficient of vegetation, n i is the Manning roughness coefficient, A i is the cross-sectional area of ​​water flow, R i is the hydraulic radius, S i is the slope of the water surface in the basin;

[0039] S3: Extract dynamic parameters that affect the calculation results of the hydrological model equation, including the precipitation distribution coefficient α i , distribution coefficient α k , slope Δk, resistance coefficient γ, influence coefficient β i ; Construct a vector X about dynamic parameters, the number of elements in the vector X is m=5;

[0040] S4: Construct a fitness function based on the water storage V(t,X) calculated by the hydrological model equation determined by the elements in the vector X and the actual water storage V0(t) at time t in the elevation level sub-area;

[0041]

[0042] Among them, F(X) is the fitness value calculated based on vector X, and T is the basin hydrological observation period; the fitness value represents the accuracy of the dynamic parameter genetic iteration process and is used to evaluate the adaptability of the genetic iteration process.

[0043] S5: During the observation period, construct any two individual vectors of the elements in vector X based on the historical water storage observation and calculation results The observation times of two individual vectors are different, and the individuals in any two individual vectors are regarded as two different parents. is the mth dynamic parameter in the first parent, is the mth dynamic parameter in the second parent;

[0044] S6: Determine the intersection point u in the two individual vectors and obtain the offspring individual vector after the crossover;

[0045]

[0046] in, is the kth individual in the first parent generation, k is the crossover point, is the kth individual in the second parent generation;

[0047] S7: Set the range of random mutation for each individual in the offspring individual vector, randomly select two corresponding offspring individuals in the two offspring individual vectors x′1 and x′2 to perform random mutation within the range of random mutation, substitute the two offspring individual vectors x′1 and x′2 after mutation into the hydrological model equation, calculate the fitness value, and calculate the acceptance probability p of the offspring individual vectors x′1 and x′2 after mutation according to the fitness value;

[0048]

[0049] Among them, w is the mutation iteration coefficient. The more mutation iterations there are, the smaller the mutation iteration coefficient is.

[0050] When p = 1, the offspring individual vector corresponding to the largest fitness value in F(x′1) and F(x′2) is selected as the optimal offspring individual vector;

[0051] When p≠1, set the receiving probability threshold p 阈值 ;

[0052] If p≤p 阈值 , then the corresponding offspring individual vector is taken as the optimal offspring individual vector;

[0053] Otherwise, re-mutate the offspring individuals in the offspring individual vectors x′1 and x′2 until the optimal offspring individual vector is output;

[0054] S8: The individuals in the optimal offspring individual vector are used as the optimized dynamic parameters, and the optimized dynamic parameters are output to update the hydrological model equation;

[0055] S9: According to the n elevation level sub-areas divided in the study area, steps S2-S7 are executed repeatedly, and the dynamic parameters of the hydrological model equation of each elevation level sub-area are optimized from top to bottom starting from the elevation level sub-area with the largest elevation, until the dynamic parameters of the hydrological model equation of the study area are optimized.

[0056] The present invention proposes a nested basin hydrological model parameter optimization method, which processes the study area of ​​the nested basin in layers to form multiple sub-areas with different elevations, establishes corresponding hydrological model equations for each layer of basin sub-areas, and iteratively optimizes the dynamic parameters of the model through an optimization algorithm; effectively overcomes the deficiency of the existing hydrological model parameter optimization that only uses the observation data of the basin outlet hydrological station, makes full use of the existing effective historical measured data, and obtains hydrological parameters that can better represent the hydrological characteristics of the basin. The results obtained by simulating these parameters can better represent the hydrological process occurring in the basin, which has important practical significance. The optimization time required for basin parameter optimization is greatly shortened, and the efficiency of parameter optimization is effectively improved.

Claims

1. A nested watershed hydrological model parameter optimization method, characterized in that: The following steps are involved: S1: Determine the study area of ​​the watershed, and stratify the study area using elevation to obtain n sub-areas of different elevation levels; S2: construct the hydrological model equations for each elevation level sub-region; S3: Extract the dynamic parameters that affect the calculation results of the hydrological model equation, construct a vector of the dynamic parameters, use the fitness function to perform crossover mutation on the dynamic parameters in the vector, screen the optimal offspring individual vector, use the individuals in the optimal offspring individual vector as the optimized dynamic parameters, and output the optimized dynamic parameters to update the hydrological model equation; S4: According to the n elevation level sub-areas divided in the study area, step S3 is executed repeatedly, and the dynamic parameters of the hydrological model equation of each elevation level sub-area are optimized from top to bottom starting from the elevation level sub-area with the largest elevation, until the dynamic parameters of the hydrological model equation of the study area are optimized.

2. The nested watershed hydrological model parameter optimization method according to claim 1, characterized in that: The step S1 includes: determining the research area of ​​the watershed, obtaining the elevation interval [h min ,h max ], set the number n of elevation level sub-areas, and calculate the elevation of each elevation level sub-area 3. The nested watershed hydrological model parameter optimization method according to claim 2, characterized in that: The hydrological model equation of each elevation level sub-region in step S2 is: Among them, i is the number of the elevation level sub-area, t is the statistical time of the hydrological data, V i (t) is the water storage capacity at time t, P i (t) is the precipitation input at time t, E i (t) is the evaporation at time t, Q out,i (t) is the outflow at time t, j is the inflow path, Q in,j (t) is the inflow at time t, J i is the number of inflow pathways, P 0,i (t) is direct precipitation, α i is the precipitation distribution coefficient, k is the distribution path of the sub-region of the previous elevation level, K is the total amount of the distribution path, P k (t) is the outflow of distribution path k at time t, α k is the distribution coefficient of the sub-area of ​​the previous elevation level, ΔR i is the radiation value received, ρ is the air density, c is the specific heat of air at constant pressure, e1 is the saturated water vapor pressure, e0 is the actual water vapor pressure, Δk is the slope of the saturated water vapor pressure-temperature curve, r′ is the air resistance, r0 is the surface resistance, γ is the resistance coefficient, β i is the influence coefficient of vegetation, n i is the Manning roughness coefficient, A i is the cross-sectional area of ​​water flow, R i is the hydraulic radius, S i is the slope of the water surface in the basin.

4. The nested watershed hydrological model parameter optimization method according to claim 3, characterized in that: The step S3 comprises: S31: Extract dynamic parameters that affect the calculation results of the hydrological model equation, including the precipitation distribution coefficient α i , distribution coefficient α k , slope Δk, resistance coefficient γ, influence coefficient β i ; Construct a vector X about dynamic parameters, the number of elements in the vector X is m=5; S32: construct a fitness function based on the water storage V(t,X) calculated by the hydrological model equation determined by the elements in the vector X and the actual water storage V0(t) at time t in the elevation level sub-area; Among them, F(X) is the fitness value calculated based on vector X, and T is the basin hydrological observation period; the fitness value represents the accuracy of the dynamic parameter genetic iteration process and is used to evaluate the adaptability of the genetic iteration process; S33: During the observation period, based on the historical water storage observation and calculation results, construct any two individual vectors of the elements in vector X The observation times of two individual vectors are different, and the individuals in any two individual vectors are regarded as two different parents. is the mth dynamic parameter in the first parent, is the mth dynamic parameter in the second parent; S34: Determine a crossover point u in the two individual vectors, and obtain a descendant individual vector after the crossover; in, is the kth individual in the first parent generation, k is the crossover point, is the kth individual in the second parent generation; S35: setting a range for random mutation of each individual in the offspring individual vector, randomly selecting two corresponding offspring individuals in the two offspring individual vectors x′1 and x′2 to perform random mutation within the range of random mutation, substituting the two offspring individual vectors x′1 and x′2 after mutation into the hydrological model equation, calculating the fitness value, and calculating the acceptance probability p of the offspring individual vectors x′1 and x′2 after mutation according to the fitness value; Among them, w is the mutation iteration coefficient. The more mutation iterations there are, the smaller the mutation iteration coefficient is. When p = 1, the offspring individual vector corresponding to the largest fitness value in F(x′1) and F(x′2) is selected as the optimal offspring individual vector; When p≠1, set the receiving probability threshold p 阈值 ; If p≤p 阈值 , then the corresponding offspring individual vector is taken as the optimal offspring individual vector; Otherwise, re-mutate the offspring individuals in the offspring individual vectors x′1 and x′2 until the optimal offspring individual vector is output; S36: The individuals in the optimal offspring individual vector are used as optimized dynamic parameters, and the optimized dynamic parameters are output to update the hydrological model equation.

Images