A method for metamodelling of physical river models
Patent Information
- Application Number
- CN202610884355.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-18
- Publication Date
- 2026-09-29
AI Technical Summary
[0010]本发明针对现有技术中的上述问题，为弥补现有技术的不足，本发明提供一种用于变态河工物理模型率定的方法，针对当前完全利用物理模型反复试错会产生成本高、周期长的问题，此外，基于河道原型尺度数学模型的预调方案因尺度失配而难以准确指导变态物理模型调整；基于河道原型尺度数学模型与变态物理模型之间缺乏一致性，导致率定效率低下、精度不足，提出降低试错次数、人工成本和优化结果准确度的用于变态河工物理模型率定的方法
[0027]本发明提供了一种用于变态河工物理模型率定的方法，通过上述方法，本发明实现了数学模型与变态河工物理模型在几何尺度、边界条件和动力响应上的高度一致性。核心在于构建与河工物理模型几何尺度完全一致的数学模型，实现“同尺度数字孪生”，从而更加准确的指导河工物理模型的率定方案。具体而言，在几何尺度方面，两者关键特征尺寸的比例关系严格对应，消除了传统模型转换中的几何畸变；在边界条件方面，入流、出流的水力参数实现动态同步反馈，确保模型与原型之间的水力相似性；在动力响应方面，水流结构、阻力、泥沙运动等关键过程均达到良好的动力匹配。这种多维度的协同一致，使得模型试验结果能够真实反映原型的物理规律。
Smart Images
Figure CN122839893A_ABST
Abstract
Description
Technical Field
[0002] This invention relates to the field of water conservancy engineering technology, and in particular to a method for calibrating a physical model of a modified river engineering system. Background Technology
[0004] River engineering physical model tests are an important means of studying river channel evolution, flow structure, sediment transport, and the effectiveness of engineering control. Due to the enormous scale of actual rivers, "abnormal models" (i.e., models with different plane and vertical scales) are usually used to balance similarity and operability. However, abnormal models cannot fully satisfy the similarity in terms of gravity, resistance, and sediment movement, resulting in deviations between their hydrodynamic characteristics and the prototype. Therefore, calibration operations are required before formal testing, that is, by adjusting parameters such as roughness and boundary conditions, to make the model's flow characteristics as close as possible to the measured conditions.
[0005] Traditional calibration methods typically rely on repeated trial and error on physical models, involving parameter adjustment, water release testing, measurement results, and readjustment. However, this process is time-consuming, costly, and limited by model site availability, water supply capacity, and manpower. To improve efficiency, some studies have attempted to introduce mathematical models to assist in pre-simulation. However, current practices often involve simulating scenarios using mathematical models at the prototype scale, and then directly applying the results to the modified physical model. Because the geometric scale, vertical compression ratio, Reynolds number, and Froude number, among other dimensionless parameters, differ between the mathematical model's simulation results and the physical model's response, the pre-calibrated schemes perform poorly on the physical model, requiring extensive on-site corrections and failing to effectively conserve resources.
[0006] Therefore, there is an urgent need for a calibration method that can achieve a high degree of matching between mathematical models and abnormal physical models in terms of geometric and dynamic characteristics, so as to improve calibration accuracy and efficiency.
[0007] Based on this, the present invention designs a method for calibrating a physical model of a modified river engineering system to solve the above problems. Summary of the Invention
[0009] (1) Technical problems to be solved
[0010] To address the aforementioned problems in the prior art and overcome its shortcomings, this invention provides a method for calibrating a physical model of a modified river system. This method addresses the issues of high cost and long cycles resulting from repeated trial and error using physical models, the difficulty in accurately guiding the adjustment of the modified physical model due to scale mismatch in pre-adjustment schemes based on the river prototype-scale mathematical model, and the low efficiency and insufficient accuracy of calibration due to the lack of consistency between the river prototype-scale mathematical model and the modified physical model. Therefore, this invention proposes a method for calibrating a physical model of a modified river system that reduces the number of trials, labor costs, and the accuracy of optimization results.
[0011] (2) Technical solution
[0012] To achieve the above-mentioned technical objectives, the present invention provides the following technical solution: a method for calibrating a physical model of a modified river engineering system, comprising the following steps:
[0013] S1: Obtaining the design parameters of the river engineering physical model: After the river engineering physical model is completed, the three control coordinate points (x1, y1), (x2, y2), and (x3, y3) of the river engineering physical model are determined using real-time dynamic carrier phase differential technology (RTK), and the complete geometric information of the river engineering physical model is organized, including the planar scale λ. l Vertical scale λ h Riverbed elevation, shoreline outline, roughness zoning, etc.;
[0014] S2: Construct a two-dimensional mathematical model of water flow at the same scale: Based on the control coordinate points and the actual size of the river engineering physical model, delineate the boundaries, bridges, groynes, etc. of the two-dimensional mathematical model of water flow, and perform mesh generation to make it completely consistent with the river engineering physical model in terms of spatial coordinates, boundary positions, bridges, groynes, and roughness partitions.
[0015] S3: Convert the prototype river hydrological conditions required for calibration, including flow rate and water level, into the corresponding inflow and outflow boundary conditions of the river physics model according to the abnormal similarity law of the river physics model. The plane scale λ of the river physics model is known. l Vertical scale λ h It can be derived that the velocity ratio is... Flow ratio Roughness ratio That is, assuming the outflow boundary water level in the prototype river channel is H meters, then the water level in the model is H / λ. h In the prototype, the water flow velocity at a certain point is V m / s, and in the model, the velocity is V / λ. v The flow rate is m / s, the inflow rate in the prototype is Q cubic meters per second, and the flow rate in the model is Q / λ. Q cubic meters per second, the roughness of the river channel in the prototype is n, and the roughness of the model is n / λ. nThe converted inflow and outflow boundary conditions are used as inputs to the two-dimensional flow mathematical model.
[0016] S4: Conduct verification tests of the river engineering physical model, verify and analyze the water level and flow velocity of the stations measured in the test with the corresponding water level and flow velocity measured in the prototype river channel, and highlight the areas where the stations with poor fit are located.
[0017] S5: In the two-dimensional flow mathematical model of the same scale, the roughness value and the water level at the outflow boundary are adjusted. When the water level is too low, the roughness value is appropriately increased and the water level curve at the outflow boundary is appropriately increased according to the water level difference at the station, so that the simulated water level and velocity results are consistent with the water level and velocity results measured by the river engineering physical model verification test.
[0018] S6: For the areas where the water level and flow velocity fitting is poor in step S4, adjust the roughness, outflow boundary water level, etc. in the same scale two-dimensional water flow mathematical model in step S5 so that the simulated water level and flow velocity results are consistent with the target calibration standard of the river engineering physical model. Generally, the water level error should not exceed 3% and the flow velocity correlation coefficient should be greater than 0.85.
[0019] S7: The parameter combination verified by the two-dimensional water flow mathematical model of the same scale is directly applied to the river engineering physical model and water release verification is carried out to complete the high-efficiency determination.
[0020] S8: If there is a slight deviation between the measured results of the river engineering physical model and the prediction of the two-dimensional flow mathematical model at the same scale, the measured data will be fed back into the two-dimensional flow mathematical model at the same scale, and the parameters will be further adjusted. The computer will be used for rapid iteration until the data matches.
[0021] Preferably, in step S2, the actual size of the river engineering physical model is a non-prototype size.
[0022] Preferably, in step S2, the computational domain, topographic data, and initial conditions of the two-dimensional water flow mathematical model strictly correspond to the river engineering model entity.
[0023] Preferably, in step S7, the parameter combination includes the partition roughness value, the outflow boundary water level value, etc.
[0024] Preferably, in step S8, the overall feedback forms a closed-loop optimization mechanism of "digital pre-simulation - physical verification - data feedback - model update". Generally, the water level error should not exceed 3%, and the flow velocity correlation coefficient is greater than 0.85 to meet the verification requirements.
[0025] Preferably, in step S8, the MIKE software is used for rapid computer iteration.
[0026] (3) Beneficial effects
[0027] This invention provides a method for calibrating a modified river physics model. Through this method, the invention achieves a high degree of consistency between the mathematical model and the modified river physics model in terms of geometric scale, boundary conditions, and dynamic response. The core lies in constructing a mathematical model with a geometric scale completely consistent with the river physics model, achieving a "same-scale digital twin," thereby more accurately guiding the calibration scheme of the river physics model. Specifically, in terms of geometric scale, the proportional relationships of the key feature dimensions of the two models strictly correspond, eliminating geometric distortions in traditional model conversions; in terms of boundary conditions, the hydraulic parameters of the inflow and outflow are dynamically and synchronously fed back, ensuring the hydraulic similarity between the model and the prototype; in terms of dynamic response, key processes such as flow structure, resistance, and sediment movement all achieve good dynamic matching. This multi-dimensional synergistic consistency enables the model test results to truly reflect the physical laws of the prototype.
[0028] Building upon this foundation, the present invention significantly improves the efficiency and accuracy of model calibration. The number of parameter iterations during calibration is drastically reduced. Simultaneously, due to enhanced consistency among models, the number of trial and error attempts in river physics models is significantly reduced, avoiding numerous invalid experiments and thus substantially reducing water consumption, equipment energy consumption, and manual operation costs during the testing process. For large-scale, long-duration river physics models, these cost-saving effects are particularly pronounced, significantly shortening the testing cycle and effectively reducing overall operating costs by approximately [amount missing].
[0029] Therefore, this invention is particularly suitable for the rapid calibration and multi-scheme comparison of large and complex river engineering physical models. It can achieve a calibration method that highly matches the mathematical model and the abnormal river engineering physical model in terms of geometric and dynamic characteristics, thereby improving calibration accuracy and efficiency, and has significant technical and economic advantages. Attached Figure Description
[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments. Wherein:
[0032] Figure 1 A diagram of the overall river engineering physical model of the lower reaches of the Minjiang River;
[0033] Figure 2 for Figure 1 Partial river channel diagram from the physical model of the China River Engineering Corporation;
[0034] Figure 3 A schematic diagram of a two-dimensional mathematical model of water flow of the same size based on a river engineering physical model;
[0035] Figure 4 for Figure 3 Schematic diagram of local mesh generation for a two-dimensional mathematical model of water flow of the same size, including groynes and bridges;
[0036] Figure 5A roughness partitioning diagram of a two-dimensional mathematical model of water flow of the same size;
[0037] Figure 6 The results are the initial verification of the water level and flow velocity of the river engineering physical model and prototype.
[0038] Figure 7 The results of matching water level and flow velocity between a two-dimensional mathematical model of water flow and a river engineering physical model of the same size;
[0039] Figure 8 The results of the verification of water level and flow velocity for a two-dimensional water flow mathematical model and a prototype of the same size are presented.
[0040] Figure 9 The results are used to verify the water level and flow velocity of the river engineering physical model and prototype. Detailed Implementation
[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0043] Example 1:
[0044] To achieve the above objectives, the present invention provides the following technical solution: a method for calibrating a physical model of a modified river engineering system, comprising the following steps:
[0045] S1: Obtain the design parameters of the river engineering physical model. After the river engineering physical model is completed, extract its complete geometric information, including the planar scale λ. L Vertical scale λ H Riverbed elevation, shoreline outline, roughness zoning, and inlet / outlet boundary location, etc.
[0046] S2: Construct a two-dimensional mathematical model of water flow at the same scale. Based on the actual dimensions (not the prototype dimensions) of the above-mentioned river engineering physical model, establish a two-dimensional mathematical model of water flow that is completely consistent with the river engineering physical model in terms of spatial coordinates, boundary positions, bridges, buttresses, roughness zones, etc. The computational domain, topographic data, and initial conditions of this mathematical model strictly correspond to the entities of the river engineering physical model.
[0047] S3: Convert the prototype hydrological conditions required for calibration, including but not limited to flow rate and water level, into the corresponding inflow and outflow boundary conditions of the river physics model according to the abnormal similarity law of the river physics model, such as: plane scale. Vertical scale Then the flow velocity is greater than the scale. Flow ratio =450*100*10=450000, and used as input for a two-dimensional water flow mathematical model of the same scale.
[0048] S4: Conduct verification tests of the river engineering physical model, and verify and analyze the water level and flow velocity of the stations measured in the test with the water level and flow velocity of the prototype stations, and highlight the areas where the stations with poor fit are located.
[0049] S5: In the two-dimensional mathematical model of water flow at the same scale, by adjusting the roughness value and the outflow boundary water level, the simulated water level and velocity results are made to match the water level and velocity results measured by the verification test of the river engineering physical model. Based on the adjusted mathematical model, the river engineering physical model is corrected.
[0050] S6: For stations in areas where the water level and velocity fitting in step S4 is poor, adjust the roughness value and outflow boundary water level in the mathematical model in step S5 to ensure that the simulated water level and velocity results match the target calibration standard of the river physics model. This process can be rapidly iterated on a computer without using the river physics model.
[0051] S7: Apply the parameter combinations (such as zone roughness values and local adjustment measures) that have been verified as effective by the two-dimensional water flow mathematical model at the same scale directly to the corresponding area of the river engineering physical model, and conduct one-time or a small number of water release verifications to complete the high-efficiency determination.
[0052] S8: If there is a slight deviation between the measured results of the river engineering physical model and the prediction of the mathematical model, the measured data can be fed back to the two-dimensional water flow mathematical model of the same scale to further fine-tune the parameters and form a closed-loop optimization mechanism of "digital pre-performance - physical verification - data feedback - model update". Generally, the water level error should not exceed 3% and the flow velocity correlation coefficient should be greater than 0.85 to meet the verification requirements.
[0053] Specifically, this embodiment uses the overall river engineering physical model of the lower reaches of the Minjiang River as an example. The model simulates a natural river channel of approximately 165 km in length, from the Shuikou Hydropower Station to the estuary, using measured topographic data of the Minjiang River. The planar scale of the river engineering physical model is 450, the vertical scale is 100, and the total length of the model is approximately 217 m. Figure 1 As shown, the river engineering physical model includes complex terrain such as the south and north river branches, sandbars, bridges, and spur dikes, etc. Figure 2 As shown, this river engineering physical model is used to study the impact of river regulation projects on water flow characteristics such as flow regime. The roughness of the river channel in different river sections is determined by the parameters designed in the river engineering physical model.
[0054] Based on the completed river physics model, three control points were selected: point 1 (2909553.857, 40383012.69), point 2 (2909591.745, 40383203.22), and point 3 (2909611.414, 40383121.97). The boundaries, bridges, and groynes of the two-dimensional flow mathematical model were delineated, and a mesh was created to establish a two-dimensional flow mathematical model that is completely consistent with it in terms of spatial coordinates, boundary positions, bridges, groynes, and roughness partitions. The computational domain, topographic data, and initial conditions of this two-dimensional flow mathematical model strictly correspond to the entities in the river physics model. The two-dimensional flow mathematical model is as follows: Figure 3 As shown, the boundary, local groynes, and bridge grids are as follows: Figure 4 As shown, the channel roughness is as follows Figure 5 As shown.
[0055] The prototype hydrological conditions required for calibration, including but not limited to flow rate and water level, are converted into the corresponding inflow and outflow boundary conditions of the river engineering physical model according to the abnormal similarity law of the river engineering physical model. Among them, the plane scale is... =450, vertical scale =100, then the flow velocity ratio is... =100 1 / 2 =10, flow rate scale =450*100*10=450000, roughness ratio =100 2 / 3 / 450 1 / 2 =1.016, Through a river engineering physical model verification experiment, the water level and flow velocity measured at the experimental station were initially verified and analyzed with the water level and flow velocity measured at the prototype station. The simulation results showed that the average error of the water level was 22%, and the correlation coefficient of the flow velocity was 0.89. The water level verification did not meet the requirements. The verification results are as follows. Figure 6 As shown.
[0056] In a two-dimensional mathematical model of water flow at the same scale, by adjusting the roughness value and the outflow boundary water level, the simulated water level and velocity results were made consistent with the water level and velocity results measured in the verification experiment of the river engineering physical model. The adjusted simulation results showed an average water level error of 2.62% and a velocity correlation coefficient of 0.98, meeting the requirements. Based on the adjusted mathematical model, the river engineering physical model was further corrected, such as... Figure 7 As shown.
[0057] For stations with poor water level and velocity fitting (such as the Zhuqi hydrological station), adjustments were made to the roughness value and outflow boundary water level in the mathematical model to ensure that the simulated water level and velocity results matched the target calibration standard of the river physics model. This process can be rapidly iterated on a computer without modifying the river physics model. After adjustment, the simulation results showed an average water level error of 2.97% and a velocity correlation coefficient of 0.99, meeting the requirements. The verification results are as follows: Figure 8 As shown.
[0058] The parameter combinations (such as zone roughness values and local adjustment measures) validated by the same-scale two-dimensional flow mathematical model were directly applied to the corresponding areas of the river engineering physical model. One-time or limited-number water discharge verifications were conducted to achieve high-efficiency determination. The simulation results after adjustment showed an average water level error of 2.06% and a flow velocity correlation coefficient of 0.99, meeting the requirements. The results are as follows: Figure 9 As shown.
[0059] If there is a slight deviation between the measured results of the river engineering physical model and the prediction of the mathematical model, the measured data can be fed back to the two-dimensional water flow mathematical model of the same scale for further fine-tuning of parameters, forming a closed-loop optimization mechanism of "digital pre-performance - physical verification - data feedback - model update".
[0060] The above embodiments are preferred embodiments of the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, the embodiments of the present invention are not limited to the foregoing embodiments. For those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for calibrating a physical model of a modified river engineering system, characterized in that: Includes the following steps: S1: Obtaining the design parameters of the river engineering physical model: After the river engineering physical model is completed, the three control coordinate points (x1, y1), (x2, y2), and (x3, y3) of the river engineering physical model are determined using real-time dynamic carrier phase differential technology (RTK), and the complete geometric information of the river engineering physical model is organized, including the planar scale λ. l Vertical scale λ h Riverbed elevation, shoreline outline, and roughness zoning; S2: Construct a two-dimensional mathematical model of water flow at the same scale: Based on the above control coordinate point positions and the actual size of the river engineering physical model, delineate the boundary, bridge, and groynes of the two-dimensional mathematical model of water flow, and perform mesh generation to make it completely consistent with the river engineering physical model in terms of spatial coordinates, boundary positions, bridges, groynes, and roughness partitions. S3: Convert the prototype river hydrological conditions required for calibration, including flow rate and water level, into the corresponding inflow and outflow boundary conditions of the river physics model according to the abnormal similarity law of the river physics model. The plane scale λ of the river physics model is known. l Vertical scale λ h It can be deduced that: the velocity ratio Flow ratio Roughness ratio That is, assuming the outflow boundary water level in the prototype river channel is H meters, then the water level in the model is H / λ. h In the prototype, the water flow velocity at a certain point is V meters per second, and in the model, the flow velocity is V / λ. v The flow rate is m / s, the inflow rate in the prototype is Q cubic meters per second, and the flow rate in the model is Q / λ. Q cubic meters per second, the roughness of the river channel in the prototype is n, and the roughness of the model is n / λ. n The converted inflow and outflow boundary conditions are used as inputs to the two-dimensional flow mathematical model. S4: Conduct verification tests of the river engineering physical model, verify and analyze the water level and flow velocity of the stations measured in the test with the corresponding water level and flow velocity measured in the prototype river channel, and highlight the areas where the stations with poor fit are located. S5: In the two-dimensional flow mathematical model of the same scale, the roughness value and the water level at the outflow boundary are adjusted. When the water level is too low, the roughness value is appropriately increased and the water level curve at the outflow boundary is appropriately increased according to the water level difference at the station, so that the simulated water level and velocity results are consistent with the water level and velocity results measured by the river engineering physical model verification test. S6: For the areas where the water level and flow velocity fitting is poor in step S4, adjust the roughness and outflow boundary water level in the same-scale two-dimensional water flow mathematical model in step S5 so that the simulated water level and flow velocity results are consistent with the target calibration standard of the river engineering physical model. Generally, the water level error should not exceed 3% and the flow velocity correlation coefficient should be greater than 0.
85. S7: The parameter combination verified by the two-dimensional water flow mathematical model of the same scale is directly applied to the river engineering physical model and water release verification is carried out to complete the high-efficiency determination. S8: If there is a slight deviation between the measured results of the river engineering physical model and the prediction of the two-dimensional flow mathematical model at the same scale, the measured data will be fed back into the two-dimensional flow mathematical model at the same scale, and the parameters will be further adjusted. The computer will be used for rapid iteration until the data matches.
2. The method for calibrating a physical model of a modified river engineering system according to claim 1, characterized in that, In step S2, the actual size of the river engineering physical model is a non-prototype size.
3. The method for calibrating a physical model of a modified river engineering system according to claim 1, characterized in that, In step S2, the computational domain, topographic data, and initial conditions of the two-dimensional water flow mathematical model strictly correspond to the river engineering model entity.
4. The method for calibrating a physical model of a modified river engineering system according to claim 1, characterized in that, In step S7, the parameter combination includes the partition roughness value and the outflow boundary water level value.
5. The method for calibrating a physical model of a modified river engineering system according to claim 1, characterized in that, In step S8, the overall feedback forms a closed-loop optimization mechanism of "digital pre-simulation - physical verification - data feedback - model update". Generally, the water level error should not exceed 3%, and the flow velocity correlation coefficient is greater than 0.85 to meet the verification requirements.
6. The method for calibrating a physical model of a modified river engineering system according to claim 1, characterized in that, In step S8, the MIKE software is used for rapid computer iteration.