A method for predicting groundwater depth, a readable storage medium and a device

The parameters of the groundwater depth prediction model were optimized by using the dung beetle optimization algorithm, which solved the problem of inaccurate prediction caused by inaccurate model parameter settings and improved the prediction accuracy and stability.

CN118673956BActive Publication Date: 2024-11-08NANJING UNIV OF INFORMATION SCI & TECH
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411164162.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-08
Estimated Expiration
2044-08-23

AI Technical Summary

Technical Problem

In existing technologies, the parameter settings of groundwater prediction models rely on experience, which may lead to model divergence, overfitting, or underfitting, affecting prediction accuracy.

Method used

The model parameters of the groundwater depth prediction model were optimized using a dung beetle optimization algorithm. The population was initialized through Chebyshev mapping, and the position updates of dung beetles were updated by perturbing them with rolling, reproduction, foraging, stealing, and adaptive T-distribution perturbations. The model parameters were optimized to improve the prediction accuracy.

Benefits of technology

It improves the prediction accuracy of the groundwater depth prediction model, reduces the model's instability, and enhances prediction performance indicators such as correlation coefficient and root mean square error.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118673956B_ABST
    Figure CN118673956B_ABST
Patent Text Reader

Abstract

The present invention discloses a groundwater depth prediction method, a readable storage medium and a device, and belongs to the field of groundwater prediction technology. The groundwater depth prediction method comprises: obtaining environmental data of a target area; inputting the environmental data of the target area into a groundwater depth prediction model trained by a dung beetle optimization algorithm, and obtaining a predicted value of the groundwater depth of the target area. The present invention uses the dung beetle optimization algorithm to optimize the parameters of the target groundwater depth prediction model, and inputs the optimal model parameters found by the optimization into the target groundwater depth prediction model, thereby improving the prediction accuracy of the prediction model and solving the problem that the model parameters of the current groundwater flow model are set by experience, resulting in the prediction result not being consistent with the actual situation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method for predicting groundwater depth, a readable storage medium and a device, and belongs to the technical field of groundwater prediction. Background Art

[0002] In recent years, physics-based numerical models, such as modular three-dimensional finite-difference groundwater flow models, have been used for groundwater prediction. However, numerical models require a large amount of accurate data, computing resources, and time. For groundwater prediction, it is difficult to obtain a large amount of accurate data, and machine learning models can make up for the shortcomings of numerical models in these aspects.

[0003] Different neural networks in machine learning models have different advantages. In groundwater prediction, the parameters in deep learning models are usually set based on experience, which may cause the model parameters to cross the optimal solution in each iteration, resulting in oscillation or divergence. This is called "oscillation phenomenon" or "instability". For neural networks, model parameters are also the key factor of "fitting". Inaccurate model parameter settings often lead to overfitting of the model or a simple model with the risk of underfitting. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a groundwater depth prediction method, a readable storage medium and a device. The model parameters of a target groundwater depth prediction model are optimized by a dung beetle optimization algorithm, and the optimal model parameters are added to the training of the target groundwater depth prediction model to improve the prediction accuracy of the prediction model. This solves the problem that the model parameters of the groundwater flow model are currently set based on experience, resulting in the prediction results being inconsistent with the actual situation.

[0005] In order to achieve the above purpose / solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0006] A first aspect of the present invention provides a method for predicting groundwater depth, comprising:

[0007] Obtain environmental data of the target area;

[0008] Input the environmental data of the target area into the groundwater depth prediction model trained by the dung beetle optimization algorithm to obtain the predicted value of the groundwater depth in the target area;

[0009] The training of the groundwater depth prediction model includes:

[0010] The model parameters of the target groundwater depth prediction model are optimized by using the dung beetle optimization algorithm to determine the optimal model parameters, where the model parameters of the target groundwater depth prediction model include learning rate, regularization parameter and number of hidden layer nodes;

[0011] The determined optimal model parameters are input into the target groundwater depth prediction model to obtain a trained groundwater depth prediction model.

[0012] Furthermore, the dung beetle optimization algorithm is used to optimize the model parameters of the target groundwater depth prediction model to determine the optimal model parameters, including:

[0013] The environmental data and the value range of the model parameters to be optimized are taken as known conditions;

[0014] Based on known conditions, setting operating parameters of the dung beetle optimization algorithm, the operating parameters including population size, maximum number of iterations and termination conditions;

[0015] Initialize the dung beetle population through Chebyshev mapping;

[0016] Determine the current fitness of the dung beetle population at its current location;

[0017] The dung beetle population was divided into rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, and the adaptive T distribution was introduced to perturb the dung beetle positions.

[0018] Updated dung beetle population locations;

[0019] Based on the current fitness of the dung beetle population at the current position, determine the updated global optimal position of the dung beetle population;

[0020] The optimal model parameters are output according to the updated global optimal position of the dung beetle population.

[0021] Furthermore, the initializing the dung beetle population by Chebyshev mapping includes:

[0022] ;

[0023] in: Indicates The location of the dung beetle, Indicates The location of a dung beetle.

[0024] Furthermore, the updating of the dung beetle population location includes:

[0025] Updated location of the rolling dung beetle:

[0026] ;

[0027] The boundary strategy is used to simulate the egg-laying area of ​​dung beetles, and the update position of the breeding dung beetles is:

[0028] ;

[0029] Updated location of the stealing dung beetle:

[0030] ;

[0031] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, Indicates the natural factors that cause the dung beetle to deviate from its original direction. The value is 1 or -1. Indicates no deviation, It means that the dung beetle has deviated from its original direction. Represents a random number between 0 and 1. Represents a random number between 0 and 1. represents the global worst position, Indicates the simulated light intensity change. The higher the value, the weaker the light source. Indicates the current local optimal position, represent independent random vectors, represents the lower bound of the breeding area, Indicates the upper limit of the breeding area; represents the global optimal position, Represents a random number, Represents a random matrix that follows a normal distribution.

[0032] Furthermore, the updating of the position of the dung beetle population further includes: updating the updated position of the foraging dung beetles by a spiral search strategy in a whale optimization algorithm:

[0033] ;

[0034] ;

[0035] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, represents random numbers that follow a normal distribution, represents independent random vectors, represents the lower bound of the dung beetle's foraging area, represents the upper bound of the dung beetle's foraging area, Represents a random number between -1 and 1, Parameters representing the dynamic spiral search shape, represents the natural index, represents a random matrix that conforms to the normal distribution, Indicates the maximum number of updates.

[0036] Furthermore, the updating position of the rolling ball dung beetle also includes:

[0037] When the rolling dung beetle encounters an obstacle and deflects, the dancing behavior is triggered, and the updated position of the dancing dung beetle is:

[0038] ;

[0039] in: It represents the angle at which the ball-rolling dung beetle deviates after encountering an obstacle. Indicated in The next location update The location of a dung beetle.

[0040] Furthermore, the step of determining the updated global optimal position of the dung beetle population based on the current fitness of the dung beetle population at the current position includes:

[0041] Obtain the updated optimal value of the position fitness of the dung beetle population;

[0042] Compare the updated optimal value of the dung beetle population position fitness with the current fitness. If the updated optimal value of the dung beetle population position fitness is better, the updated dung beetle population position replaces the original dung beetle position. Otherwise, the dung beetle position remains unchanged. The dung beetle population position is updated multiple times. , until the maximum number of updates meets the termination condition;

[0043] The position of the dung beetle population with the best fitness value is taken as the global optimal position of the dung beetle population.

[0044] Furthermore, the introducing of adaptive T distribution to disturb the position of dung beetles includes:

[0045] ;

[0046] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, represents the random disturbance term, Indicates the number of location updates is the degrees of freedom parameter of the T distribution.

[0047] A second aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting groundwater depth.

[0048] A third aspect of the present invention provides a computer device, comprising:

[0049] A memory for storing instructions;

[0050] The processor is used to execute the instructions so that the device implements the groundwater depth prediction method as described above.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. The present invention uses the dung beetle optimization algorithm to optimize the model parameters of the target groundwater depth prediction model, and inputs the optimal model parameters found by the optimization into the training of the target groundwater depth prediction model, thereby improving the prediction accuracy of the prediction model and solving the problem that the model parameters of the groundwater flow model are currently set by experience, resulting in the prediction results not being consistent with the actual situation.

[0053] 2. The dung beetle optimization algorithm adopted in the present invention replaces the random initialization of the original dung beetle optimization algorithm with Chebyshev mapping, thereby improving the distribution of the initial population in the search space, updating the position of the dung beetle population in the foraging stage through a spiral search strategy and using the adaptive T distribution to perturb the position of the dung beetles, thereby optimizing the search direction and distance of the dung beetle individuals and avoiding the algorithm from falling into the local optimum during the foraging stage of the dung beetles. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flowchart of a method for predicting groundwater depth, a readable storage medium and a device according to an embodiment of the present invention;

[0055] Figure 2 It is a schematic diagram of a process flow of training a groundwater depth prediction model of a groundwater depth prediction method according to an embodiment of a groundwater depth prediction method, a readable storage medium and a device of the present invention;

[0056] Figure 3 It is a schematic diagram of the prediction results using the existing prediction model (DBO) based on the original dung beetle optimization algorithm;

[0057] Figure 4 The present invention is a schematic diagram of prediction results of a groundwater depth prediction method according to an embodiment of a groundwater depth prediction method, a readable storage medium and a device. DETAILED DESCRIPTION

[0058] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0059] Embodiment 1

[0060] like Figure 1 As shown, a method for predicting groundwater depth includes:

[0061] Obtain environmental data of the target area, where the environmental data includes local meteorological data and atmospheric circulation data.

[0062] The environmental data of the target area is input into the groundwater depth prediction model trained by the dung beetle optimization algorithm to obtain the predicted value of the groundwater depth in the target area.

[0063] like Figure 2 As shown in Figure 1, the training of the groundwater depth prediction model includes:

[0064] The dung beetle optimization algorithm is used to optimize the model parameters of the target groundwater depth prediction model to determine the optimal model parameters. The model parameters of the target groundwater depth prediction model include learning rate, regularization parameter and number of hidden layer nodes. Specifically:

[0065] The environmental data and the value range of the model parameters to be optimized are taken as known conditions.

[0066] Based on known conditions, set the operating parameters of the dung beetle optimization algorithm, including population size, maximum number of iterations and termination conditions;

[0067] Initialize the dung beetle population using the Chebyshev map, including:

[0068] ;

[0069] in: Indicates The location of the dung beetle, Indicates The location of a dung beetle.

[0070] Determine the current fitness of the dung beetle population at its current location.

[0071] The dung beetle population is divided into rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, and the adaptive T distribution is introduced to perturb the dung beetle positions, wherein the introduction of the adaptive T distribution to perturb the dung beetle positions includes:

[0072] ;

[0073] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, represents the random disturbance term, Indicates the number of location updates is the degrees of freedom parameter of the T distribution.

[0074] Updated dung beetle population locations, including:

[0075] Updated location of the rolling dung beetle:

[0076] ;

[0077] When the ball-rolling dung beetle encounters an obstacle and deflects (i.e. ), the dancing behavior is triggered, and the updated position of the dancing dung beetle is:

[0078] ;

[0079] It represents the angle at which the ball-rolling dung beetle deviates after encountering an obstacle. Indicated in The next location update The location of the dung beetle;

[0080] The boundary strategy is used to simulate the egg-laying area of ​​dung beetles and the updated position of breeding dung beetles:

[0081] ;

[0082] Updated location of the stealing dung beetle:

[0083] ;

[0084] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, Indicates the natural factors that cause the dung beetle to deviate from its original direction. The value is 1 or -1. Indicates no deviation, It means that the dung beetle has deviated from its original direction. Represents a random number between 0 and 1. Represents a random number between 0 and 1. represents the global worst position, Indicates the simulated light intensity change. The higher the value, the weaker the light source. Indicates the current local optimal position, represent independent random vectors, represents the lower bound of the breeding area, Indicates the upper limit of the breeding area; represents the global optimal position, Represents a random number, Represents a random matrix that follows a normal distribution.

[0085] Update the updated position of the foraging dung beetle through the spiral search strategy in the whale optimization algorithm:

[0086] ;

[0087] ;

[0088] in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, represents random numbers that follow a normal distribution, represents independent random vectors, represents the lower bound of the dung beetle's foraging area, represents the upper bound of the dung beetle's foraging area, Represents a random number between -1 and 1, Parameters representing the dynamic spiral search shape, represents the natural index, represents a random matrix that conforms to the normal distribution, Indicates the maximum number of updates.

[0089] Based on the current fitness of the dung beetle population at the current position, the updated global optimal position of the dung beetle population is determined, including:

[0090] Get the updated optimal value of the position fitness of the dung beetle population.

[0091] Compare the updated optimal value of the dung beetle population position fitness with the current fitness. If the updated optimal value of the dung beetle population position fitness is better, the updated dung beetle population position replaces the original dung beetle position. Otherwise, the dung beetle position remains unchanged. The dung beetle population position is updated multiple times. , until the maximum number of updates meets the termination condition.

[0092] The position of the dung beetle population with the best fitness value is taken as the global optimal position of the dung beetle population.

[0093] The optimal model parameters are output according to the updated global optimal position of the dung beetle population.

[0094] The determined optimal model parameters are input into the target groundwater depth prediction model to obtain a trained groundwater depth prediction model.

[0095] The model prediction performance test indicators are the correlation coefficient (R), root mean square error (RMSE) and residual prediction deviation (RPD). The higher the R and RPD, the better the performance of the model. At the same time, the smaller the RMSE, the more accurate the model prediction.

[0096] like Figure 3 and Figure 4 As shown, Figure 3 and Figure 4 The solid lines are all predicted values, the dotted lines are all observed values, and the vertical axis GWP represents the value of groundwater depth (in meters). Compared with the existing prediction model (DBO) based on the original dung beetle optimization algorithm, the R of the trained groundwater depth prediction model (IDBO) is 0.83, an increase of 2.6%, the RPD is increased from 2.4015 to 2.4381, and the RMSE is reduced from 0.0604 to 0.0595, and the prediction performance has been improved.

[0097] Embodiment 2

[0098] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the following method:

[0099] Obtain environmental data of the target area, including local meteorological data and atmospheric circulation data;

[0100] Input the environmental data of the target area into the groundwater depth prediction model trained by the dung beetle optimization algorithm to obtain the predicted value of the groundwater depth in the target area;

[0101] Among them, the training of groundwater depth prediction model includes:

[0102] The model parameters of the target groundwater depth prediction model are optimized by using the dung beetle optimization algorithm to determine the optimal model parameters, where the model parameters of the target groundwater depth prediction model include learning rate, regularization parameter and number of hidden layer nodes;

[0103] The determined optimal model parameters are input into the target groundwater depth prediction model to obtain a trained groundwater depth prediction model.

[0104] Embodiment 3

[0105] A computer device comprising:

[0106] A memory for storing instructions;

[0107] The processor is used to execute the instructions so that the device performs the steps of implementing the following method:

[0108] Obtain environmental data of the target area, including local meteorological data and atmospheric circulation data;

[0109] Input the environmental data of the target area into the groundwater depth prediction model trained by the dung beetle optimization algorithm to obtain the predicted value of the groundwater depth in the target area;

[0110] Among them, the training of groundwater depth prediction model includes:

[0111] The model parameters of the target groundwater depth prediction model are optimized by using the dung beetle optimization algorithm to determine the optimal model parameters, where the model parameters of the target groundwater depth prediction model include learning rate, regularization parameter and number of hidden layer nodes;

[0112] The determined optimal model parameters are input into the target groundwater depth prediction model to obtain a trained groundwater depth prediction model.

[0113] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take 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 codes.

[0114] The present application is described with reference to the flowcharts of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A function specified in a process or multiple processes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 The steps of a specified function in a process or multiple processes.

[0117] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms under the guidance of the present invention, which all fall within the protection of the present invention.

Claims

1. A method for predicting groundwater depth, characterized in that: include: Obtain environmental data of the target area; Input the environmental data of the target area into the groundwater depth prediction model trained by the dung beetle optimization algorithm to obtain the predicted value of the groundwater depth in the target area; The training of the groundwater depth prediction model includes: The model parameters of the target groundwater depth prediction model are optimized by using the dung beetle optimization algorithm to determine the optimal model parameters, where the model parameters of the target groundwater depth prediction model include learning rate, regularization parameter and number of hidden layer nodes; The method of using the dung beetle optimization algorithm to optimize the model parameters of the target groundwater depth prediction model to determine the optimal model parameters includes: The environmental data and the value range of the model parameters to be optimized are taken as known conditions; Based on known conditions, setting operating parameters of the dung beetle optimization algorithm, the operating parameters including population size, maximum number of iterations and termination conditions; pass chebyshev Map initializes the dung beetle population; Determine the current fitness of the dung beetle population at its current location; The dung beetle population was divided into rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, and adaptive T Distribution disturbance dung beetle location; Updating the position of the dung beetle population; the updating of the position of the dung beetle population also includes: updating the updated position of the foraging dung beetles through the spiral search strategy in the whale optimization algorithm: ; ; in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, represents random numbers that follow a normal distribution, represents independent random vectors, represents the lower bound of the dung beetle's foraging area, represents the upper bound of the dung beetle's foraging area, Represents a random number between -1 and 1, Parameters representing the shape of the dynamic spiral search, represents the natural index, represents a random matrix that conforms to the normal distribution, Indicates the maximum number of updates; Based on the current fitness of the dung beetle population at the current position, determine the updated global optimal position of the dung beetle population; Output the optimal model parameters according to the updated global optimal position of the dung beetle population; The determined optimal model parameters are input into the target groundwater depth prediction model to obtain a trained groundwater depth prediction model.

2. The method for predicting groundwater depth according to claim 1, characterized in that: Said through chebyshev Mapping initializes the dung beetle population, including: ; in: Indicates The location of the dung beetle, Indicates The location of a dung beetle.

3. The method for predicting groundwater depth according to claim 1, characterized in that: The updating of the dung beetle population location includes: Updated location of the rolling dung beetle: ; The boundary strategy is used to simulate the egg-laying area of ​​dung beetles, and the update position of the breeding dung beetles is: ; Updated location of the stealing dung beetle: ; in: Indicated in The next location update The location of the dung beetle, Indicated in The next location update The location of the dung beetle, Indicates the natural factors that cause the dung beetle to deviate from its original direction. The value is 1 or -1. Indicates no deviation, It means that the dung beetle has deviated from its original direction. Represents a random number between 0 and 1. Represents a random number between 0 and 1. represents the global worst position, Indicates the simulated light intensity change. The higher the value, the weaker the light source. Indicates the current local optimal position, represent independent random vectors, represents the lower bound of the breeding area, Indicates the upper limit of the breeding area; represents the global optimal position, Represents a random number, Represents a random matrix that follows a normal distribution.

4. The method for predicting groundwater depth according to claim 3, characterized in that: The updated position of the rolling ball dung beetle also includes: When the rolling dung beetle encounters an obstacle and deflects, the dancing behavior is triggered, and the updated position of the dancing dung beetle is: ; in: It represents the angle at which the ball-rolling dung beetle deviates after encountering an obstacle. Indicated in The position of the dung beetle at the time of the position update.

5. The method for predicting groundwater depth according to claim 1, characterized in that: The step of determining the updated global optimal position of the dung beetle population based on the current fitness of the dung beetle population at the current position includes: Obtain the updated optimal value of the position fitness of the dung beetle population; Compare the updated optimal value of the dung beetle population position fitness with the current fitness. If the updated optimal value of the dung beetle population position fitness is better, the updated dung beetle population position replaces the original dung beetle position. Otherwise, the dung beetle position remains unchanged. The dung beetle population position is updated multiple times. , until the maximum number of updates meets the termination condition; The position of the dung beetle population with the best fitness value is taken as the global optimal position of the dung beetle population.

6. The method for predicting groundwater depth according to claim 1, characterized in that: The introduction of adaptive T Distributed disturbance of dung beetle locations, including: ; in: Indicated in The position of the dung beetle at the time of the position update, Indicated in The next location update The location of the dung beetle, represents the random disturbance term, Indicates the number of location updates for T The degrees of freedom parameter of the distribution.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the groundwater depth prediction method as described in any one of claims 1 to 6 is implemented.

8. A computer device, characterized in that: include: A memory for storing instructions; The processor is used to execute the instruction so that the device implements the groundwater depth prediction method as described in any one of claims 1 to 6.

Citation Information

Patent Citations

  • A method for optimizing groundwater depth of BP neural network based on genetic algorithm

    CN109409568A

  • Structure parameter optimization method for permanent magnet adsorption device of wall-climbing robot

    CN116362055A

  • Photovoltaic power short-term prediction method and system based on improved dung beetle optimization algorithm

    CN117039894A

  • Short-term wind power prediction method based on DBO-ELM

    CN117217374A

  • Unknown radar waveform identification method based on dung beetle optimization-open set differential distribution alignment

    CN117390544A