Hydrological model parameter automatic correction method

Abstract

The invention discloses a hydrological model parameter automatic correction method, and belongs to the technical field of hydrological modeling and parameter optimization, and the method comprises the steps: obtaining real-time hydrological data; comparing a prediction result generated by the hydrological model under the current parameter with real-time hydrological data to obtain a prediction error; based on the prediction error, automatically correcting parameters of the hydrological model by adopting an adaptive optimization algorithm; comparing a new prediction result generated by the hydrological model under the corrected parameters with the real-time hydrological data to obtain a new prediction error; if the new prediction error is smaller than the error threshold value, correction is completed, and otherwise, according to the new prediction error, a self-adaptive optimization algorithm is adopted to conduct repeated correction on the parameters of the hydrological model till the latest prediction error is smaller than the error threshold value. The method can realize automatic correction of hydrological model parameters and improve the prediction precision of the hydrological model.

Description

Technical Field

[0001] The present invention relates to a method for automatically calibrating parameters of a hydrological model, belonging to the technical field of hydrological modeling and parameter optimization. Background Art

[0002] In the processes of flow prediction, flood forecasting, etc. by traditional hydrological models, parameters are usually set based on a large amount of historical data and expert experience. The existing technologies have the following deficiencies: 1. Strong parameter fixity: Parameters of traditional models are usually set statically, making it difficult to reflect the dynamic changes of the hydrological environment in a timely manner, resulting in insufficient prediction accuracy of the model; 2. Limited data sources: Most hydrological models only rely on data from fixed monitoring stations and fail to make full use of the real-time big data obtained from multiple channels such as satellites, drones, and the Internet of Things; 3. Dependence on manual intervention: Existing parameter calibration methods mainly rely on manual adjustment by experts, which is time-consuming and prone to introducing subjective factors, with a low level of automation.

[0003] With the rapid development of big data technology, the Internet of Things, and machine learning technology, it has become possible to collect and process a large amount of hydrological, meteorological, and environmental data in real time. There is an urgent need for a technical solution that can make full use of these data to achieve automatic calibration of model parameters, so as to improve the accuracy and timeliness of hydrological prediction. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for automatically calibrating parameters of a hydrological model, which can realize automatic calibration of hydrological model parameters and improve the prediction accuracy of the hydrological model.

[0005] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for automatically calibrating parameters of a hydrological model, including: Comparing the prediction result generated by the hydrological model under the current parameters with real-time hydrological data to obtain a prediction error; Based on the prediction error, automatically calibrating the parameters of the hydrological model by using an adaptive optimization algorithm; Comparing the new prediction result generated by the hydrological model under the calibrated parameters with real-time hydrological data to obtain a new prediction error; If the new prediction error is less than the error threshold, the calibration is completed; otherwise, based on the new prediction error, the parameters of the hydrological model are repeatedly calibrated by using an adaptive optimization algorithm until the latest prediction error is less than the error threshold.

[0006] In combination with the first aspect, further, real-time hydrological data is collected in real time through hydrological sensors, meteorological monitoring stations, satellite remote sensing, and unmanned aerial vehicle monitoring systems, and is transmitted to the central data processing platform through the Internet of Things for preprocessing.

[0007] In combination with the first aspect, further, the preprocessing includes: data cleaning, denoising, normalization, and format conversion of the real-time hydrological data.

[0008] In combination with the first aspect, further, the real-time hydrological data includes: rainfall, flow rate, water level, temperature, and humidity.

[0009] In combination with the first aspect, further, the hydrological models include distributed models and semi-empirical models.

[0010] In combination with the first aspect, further, the prediction errors include mean square error and mean absolute error.

[0011] In combination with the first aspect, further, the adaptive optimization algorithm is the gradient descent method; Based on the prediction error, automatically correcting the parameters of the hydrological model using the gradient descent method includes: Calculating the gradient of the error function with respect to the parameters of the hydrological model using the gradient descent method; According to the gradient of the error function with respect to the parameters of the hydrological model, with the goal of minimizing the prediction error, updating the parameters of the hydrological model; Repeating the iterative update until the prediction error is less than the error threshold or the update amplitude of the parameters of the hydrological model is less than the amplitude threshold, and completing the iterative update; Among them, the error function is: ; Among them, represents the error function, represents the parameters of the hydrological model, represents the th sample data input into the hydrological model, represents the prediction result generated by the hydrological model corresponding to under , represents corresponding real-time hydrological data, represents the total number of sample data; The gradient descent update formula is: ; Among them, represents the updated parameters of the hydrological model, represents the learning rate, represents the gradient of the error function with respect to the parameters of the hydrological model.

[0012] In combination with the first aspect, further, the adaptive optimization algorithm is a genetic algorithm; Based on the prediction error, automatically correcting the parameters of the hydrological model using the genetic algorithm includes: Generating several groups of initial parameter combinations as a population, where each individual represents a group of hydrological model parameters; Running the hydrological model for each parameter combination, calculating the prediction error, and constructing a fitness function to evaluate the quality of each individual; Selecting individuals with good performance according to the fitness function; performing crossover operations on the selected individuals to generate new parameter combinations; randomly mutating some parameters to maintain population diversity; replacing the original population with a new generation of population and repeating the fitness evaluation; When after a fixed number of generations of iteration, the prediction error of the optimal individual in the population is less than the error threshold, selecting this parameter combination as the final corrected parameter of the hydrological model and using it to update the hydrological model; Among them, the fitness function uses error reciprocal transformation.

[0013] In the second aspect, the present invention provides a computer device, including: A storage medium for storing a computer program; A processor for executing the computer program to implement the method for automatically correcting the parameters of the hydrological model according to any one of the first aspect.

[0014] In the third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method for automatically correcting the parameters of the hydrological model according to any one of the first aspect.

[0015] In the fourth aspect, the present invention provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the method for automatically correcting the parameters of the hydrological model according to any one of the first aspect.

[0016] Compared with the prior art, the beneficial effects of the present invention are: The method for automatically correcting the parameters of the hydrological model provided by the present invention, by collecting multi-source hydrological data in real time and using an adaptive optimization algorithm to dynamically adjust the key parameters of the hydrological model, reduces the dependence on expert experience, realizes the real-time automatic correction of the parameters of the hydrological model, solves the defects of fixed parameters and manual intervention in traditional hydrological models, can reduce the prediction error, and improves the accuracy of the hydrological model in predicting complex hydrological processes such as flow, flood, and drought. It makes full use of multi-channel real-time data such as sensors, remote sensing, and drones, breaking through the limitations of traditional single data sources. Description of the Drawings

[0017] Figure 1It is the flowchart of the automatic calibration method for hydrological model parameters provided by the embodiments of the present invention; Figure 2 It is the flowchart of the adaptive parameter adjustment provided by the embodiments of the present invention. Detailed implementation manners

[0018] The technical solutions of the present application will be further described in detail below in conjunction with the detailed implementation manners.

[0019] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present application and should not be construed as a limitation to the present application. Without conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0020] The embodiments of the present application provide an automatic calibration method for hydrological model parameters, as Figure 1 shown, including: Obtain real-time hydrological data; Compare the prediction result generated by the hydrological model under the current parameters with the real-time hydrological data to obtain the prediction error; Based on the prediction error, automatically calibrate the parameters of the hydrological model by using an adaptive optimization algorithm; Compare the new prediction result generated by the hydrological model under the calibrated parameters with the real-time hydrological data to obtain the new prediction error; If the new prediction error is less than the error threshold, the calibration is completed; otherwise, according to the new prediction error, use the adaptive optimization algorithm to repeatedly calibrate the parameters of the hydrological model until the latest prediction error is less than the error threshold.

[0021] For the automatic calibration method for hydrological model parameters provided by the embodiments of the present application, the hydrological model is re-run with the calibrated parameters to generate a new prediction result, and then the new prediction result is compared with the real-time hydrological data to form a closed-loop feedback, continuously adjusting and optimizing the parameters of the hydrological model.

[0022] In this embodiment, the real-time hydrological data is collected in real time by various hydrological sensors, meteorological monitoring stations, satellite remote sensing and unmanned aerial vehicle monitoring systems deployed in water areas, river channels and basins, and is transmitted to the central data processing platform through the Internet of Things for preprocessing.

[0023] Specifically, the real-time hydrological data includes multi-source hydrological and environmental data such as rainfall, flow, water level, temperature and humidity.

[0024] The preprocessing includes: cleaning, denoising, normalizing, and format conversion of real-time hydrological data to ensure data quality and compatibility, and providing uniformly standard data input for subsequent processing and analysis.

[0025] In this embodiment, the hydrological model includes distributed models, semi-empirical models, etc. The prediction errors include error calculation indicators such as mean square error and mean absolute error.

[0026] In a possible embodiment, the adaptive optimization algorithm is the gradient descent method.

[0027] In this embodiment, as Figure 2 shown, based on the prediction error, automatically correcting the parameters of the hydrological model using the gradient descent method specifically includes the following steps: Step 1: Initial parameter setting: Set a group of initial parameters as the initial state of the hydrological model; Step 2: Use the gradient descent method to calculate the gradient of the error function with respect to the parameters of the hydrological model; In this embodiment, the error function is: ; where represents the error function, represents the parameters of the hydrological model, represents the th sample data input into the hydrological model, represents the prediction result generated by the hydrological model corresponding to under , represents corresponding real-time hydrological data, represents the total number of sample data.

[0028] Step 3: According to the gradient of the error function with respect to the parameters of the hydrological model, with the goal of minimizing the prediction error, update the parameters of the hydrological model; In this embodiment, the gradient descent update formula is: ; where represents the updated parameters of the hydrological model, represents the learning rate, represents the gradient of the error function with respect to the parameters of the hydrological model.

[0029] Step 4: Repeat the iterative update until the prediction error is less than the error threshold or the update amplitude of the parameters of the hydrological model is less than the amplitude threshold, and complete the iterative update.

[0030] In a possible embodiment, the adaptive optimization algorithm is the genetic algorithm.

[0031] In this embodiment, based on the prediction error, the genetic algorithm is used to automatically correct the parameters of the hydrological model, which specifically includes the following steps: Step 1: Generate several groups of initial parameter combinations as a population, and each individual represents a group of hydrological model parameters; Step 2: Run the hydrological model for each parameter combination, calculate the prediction error, and construct a fitness function to evaluate the quality of each individual; In this embodiment, the fitness function uses the reciprocal transformation of the error.

[0032] Step 3: Select the individuals with good performance according to the fitness function; perform crossover operations on the selected individuals to generate new parameter combinations; randomly mutate some parameters to maintain population diversity; replace the original population with the new generation population, and repeat the fitness evaluation; Step 4: When after a fixed number of generations of iteration, the prediction error of the optimal individual in the population is less than the error threshold, select this parameter combination as the final corrected parameter of the hydrological model and use it to update the hydrological model.

[0033] The method for automatically correcting the parameters of the hydrological model provided by the embodiment of the present application uses the corrected parameters to run the hydrological model again, obtains new prediction results, compares the new prediction results with real-time hydrological data, obtains new prediction error feedback, and continuously optimizes the parameters of the hydrological model until long-term stable prediction accuracy is achieved.

[0034] Through experimental verification, after adopting the method for automatically correcting the parameters of the hydrological model provided by this embodiment, the prediction accuracy and response speed of the hydrological model in complex hydrological processes such as floods and droughts have been significantly improved. At the same time, the method for automatically correcting the parameters of the hydrological model provided by this embodiment has good automation and integration characteristics, is convenient for popularization and application in practical projects such as water conservancy disaster prevention and resource scheduling, and has broad practical value and market prospects.

[0035] The embodiment of the present application provides a system for automatically correcting the parameters of a hydrological model, which integrates each module for implementing the method for automatically correcting the parameters of the hydrological model provided by the embodiment of the present application through a software platform to construct an automatic control system, and realizes data interfaces, algorithm modules, feedback control, and online monitoring. The front-end interface of the system displays data collection, parameter update, and error trends in real time, providing an intuitive basis for water conservancy scheduling and disaster prevention decision-making.

[0036] The system for automatically correcting the parameters of the hydrological model provided by this embodiment can execute the method for automatically correcting the parameters of the hydrological model provided by any embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method.

[0037] The embodiment of the present application provides a computer device, including: A storage medium for storing a computer program; A processor for executing a computer program to implement the automatic calibration method for hydrological model parameters provided in any embodiment of the present application.

[0038] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the automatic calibration method for hydrological model parameters provided in any embodiment of the present application is implemented.

[0039] An embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, the automatic calibration method for hydrological model parameters provided in any embodiment of the present application is implemented.

[0040] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0041] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more of the processes Figure 1 or blocks or the combination of blocks.

[0042] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one or more of the processes Figure 1 or blocks or the combination of blocks.

[0043] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide means for implementing the functions in the processFigure 1 one process or multiple processes and / or boxes Figure 1 steps of functions specified in one box or multiple boxes

[0044] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present application.

Claims

1. An automatic calibration method for hydrological model parameters, characterized in that It includes: Obtain real-time hydrological data; Compare the prediction results generated by the hydrological model under the current parameters with the real-time hydrological data to obtain the prediction error; Based on the prediction error, use an adaptive optimization algorithm to automatically correct the parameters of the hydrological model; Compare the new prediction results generated by the hydrological model under the corrected parameters with the real-time hydrological data to obtain the new prediction error; If the new prediction error is less than the error threshold, the correction is completed; otherwise, according to the new prediction error, use the adaptive optimization algorithm to repeatedly correct the parameters of the hydrological model until the latest prediction error is less than the error threshold.

2. The automatic calibration method for hydrological model parameters according to claim 1, characterized in that, The real-time hydrological data is collected in real time through hydrological sensors, meteorological monitoring stations, satellite remote sensing, and unmanned aerial vehicle monitoring systems, and is transmitted to the central data processing platform through the Internet of Things for preprocessing.

3. The automatic calibration method for hydrological model parameters according to claim 2, characterized in that The preprocessing includes: data cleaning, denoising, normalization, and format conversion of the real-time hydrological data.

4. The automatic calibration method for hydrological model parameters according to claim 2, wherein The real-time hydrological data includes: rainfall, flow rate, water level, temperature, and humidity.

5. The automatic calibration method for hydrological model parameters according to claim 1, characterized in that The hydrological model includes a distributed model and a semi-empirical model.

6. The automatic calibration method for hydrological model parameters according to claim 1, characterized in that The prediction error includes the mean square error and the mean absolute error.

7. The automatic calibration method for hydrological model parameters according to claim 1, characterized in that The adaptive optimization algorithm is the gradient descent method; Based on the prediction error, using the gradient descent method to automatically correct the parameters of the hydrological model includes: Use the gradient descent method to calculate the gradient of the error function with respect to the parameters of the hydrological model; According to the gradient of the error function with respect to the parameters of the hydrological model, with the goal of minimizing the prediction error, update the parameters of the hydrological model; Repeat the iterative update until the prediction error is less than the error threshold or the update amplitude of the parameters of the hydrological model is less than the amplitude threshold, and complete the iterative update; Among them, the error function is: ； Among them, represents the error function, represents the parameters of the hydrological model, represents the th sample data input into the hydrological model, represents the predicted result generated by the hydrological model corresponding to below, and represents corresponding real-time hydrological data, represents the total number of sample data; The gradient descent update formula is: ； Among them, represents the parameters of the updated hydrological model, represents the learning rate, represents the gradient of the error function with respect to the parameters of the hydrological model.

8. The automatic calibration method of hydrological model parameters according to claim 1, characterized in that The adaptive optimization algorithm is the genetic algorithm; Based on the prediction error, using the genetic algorithm to automatically correct the parameters of the hydrological model includes: Generate several groups of initial parameter combinations as a population, and each individual represents a set of hydrological model parameters; Run the hydrological model for each parameter combination, calculate the prediction error, and construct a fitness function to evaluate the quality of each individual; Select the individuals with good performance according to the fitness function; perform crossover operations on the selected individuals to generate new parameter combinations; randomly mutate some parameters to maintain population diversity; replace the original population with the new generation population and repeat the fitness evaluation; When after a fixed number of generations of iteration, the prediction error of the best individual in the population is less than the error threshold, select this parameter combination as the final corrected parameter of the hydrological model and use it to update the hydrological model; Among them, the fitness function uses the reciprocal transformation of the error.

9. A computer device, characterized in that, It includes: A storage medium for storing a computer program; A processor for executing the computer program to implement the hydrological model parameter automatic correction method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the hydrological model parameter automatic correction method according to any one of claims 1 to 8.

Images