Method and system for optimizing wave parameters based on internal and external double-layer iteration

Abstract

The invention relates to the technical field of ocean, and provides a method and system for optimizing wave parameters based on inner and outer double-layer iteration. The method for optimizing the wave parameters based on inner and outer double-layer iteration comprises the following steps: constructing a wave model based on a wave mode taking a spectral energy balance equation as a core; the wave model comprises a plurality of empirical parameters and adjustment coefficients, namely to-be-optimized parameters, and then to-be-optimized parameter vectors are formed; searching an optimal parameter vector by using an inner and outer double-layer iterative optimization method, so that the error between the output wave data and the observation wave data of the wave model is minimum; the error between the output wave data and the observation wave data is represented by adopting a regularization error function, and the regularization error function is a normalized weighted mean square error introducing a priori constraint regularization term. According to the method, a wave model parameter optimization system with intelligent adjustment, closed-loop feedback and continuous evolution characteristics can be constructed, and the method can be widely applied to high-precision parameter optimization and automatic calibration of a wave mode.

Description

Technical Field

[0001] This invention relates to the field of marine technology, and in particular to a method and system for optimizing wave parameters based on internal and external double-layer iterative optimization. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Accurate simulation and forecasting of deep-sea waves are of vital importance for marine resource development, shipping safety, and disaster prevention and mitigation. Third-generation wave numerical models, represented by the SWAN model, are currently a crucial means of achieving high-precision ocean wave simulation. Their forecasting performance largely depends on the rational configuration of parameterization schemes for multiple physical processes within the model.

[0004] Currently, the methods for determining model parameters are either to directly use the model's default parameters or to calibrate them manually through trial and error based on limited local observation data. These traditional methods have limitations such as poor adaptability, low efficiency, and susceptibility to local optima. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for optimizing wave parameters based on internal and external double-layer iterative optimization, which can be widely applied to high-precision parameter optimization and automatic calibration of ocean wave models.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of the present invention provides a method for optimizing wave parameters based on an inner and outer two-layer iterative process.

[0007] In one or more embodiments, a method for optimizing wave parameters based on inner and outer two-layer iterative methods is provided, including: Based on the wave pattern with the spectral energy balance equation as the core, a wave model is constructed; the wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus constitute the parameter vector to be optimized. The optimal parameter vector is found by using an internal and external double-layer iterative optimization method to minimize the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is characterized by a regularized error function, which is a normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer two-layer iterative optimization method, the inner algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer algorithm adjusts the core hyperparameters of the inner algorithm by using the key statistical features of the inner algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

[0008] In one implementation, the parameter vector to be optimized includes the wind input coefficient, the white crown dissipation coefficient, and the bottom friction coefficient.

[0009] As one implementation method, the regularization error function for:

[0010]

[0011] in, This is the normalized weighted mean square error; For regularization weights; Let j be the parameter to be optimized; The parameter to be optimized is the empirical value; n is the number of parameters to be optimized. Allowable deviation; This represents the number of observation points; No. Data duration for each observation point; The first simulation of the model Each observation point is at The wave height or period value at any given moment; These are the corresponding measured values; The weighting coefficients for the observation points. ; To prevent small constants with a denominator of zero.

[0012] As one implementation method, the inner algorithm employs a genetic algorithm, treating a set of parameters to be optimized as an individual; the fitness function is set to... : ; in, Represents an individual The regularization error value.

[0013] As one implementation method, the outer layer algorithm adopts a reinforcement learning algorithm. The reinforcement learning algorithm uses the key statistical features of the inner layer algorithm in real time to form the environmental state and output actions to adjust the core hyperparameters of the inner layer algorithm.

[0014] As one implementation method, the environment state of a reinforcement learning algorithm consists of the optimal error, average error, error variance, and intergenerational improvement magnitude; the output action consists of the population size, crossover rate, mutation rate, and mutation intensity.

[0015] As one implementation method, when the reinforcement learning algorithm gets stuck in local convergence or the diversity decreases, the mutation rate or mutation intensity is increased to promote exploration. At the same time, the population size or crossover rate is increased to expand the search space or promote recombination and information exchange between individuals and improve the diversity of the population. When the error continues to decrease and the optimization direction is stable, the parameters such as population size, crossover rate, mutation rate and mutation intensity are reduced to improve search efficiency.

[0016] A second aspect of the present invention provides a system for optimizing wave parameters based on an inner and outer two-layer iterative process.

[0017] In one or more embodiments, a system for optimizing wave parameters based on an inner and outer two-layer iterative process includes: The wave model construction module is used to construct wave models based on wave patterns with the spectral energy balance equation as the core. The wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus form the parameter vector to be optimized. The parameter vector optimization module is used to find the optimal parameter vector using an internal and external double-layer iterative optimization method to minimize the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is characterized by a regularized error function, which is a normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer two-layer iterative optimization method, the inner algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer algorithm adjusts the core hyperparameters of the inner algorithm by using the key statistical features of the inner algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

[0018] A third aspect of the present invention provides a computer-readable storage medium.

[0019] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described above.

[0020] A fourth aspect of the present invention provides an electronic device.

[0021] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described above.

[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention utilizes a dual-layer iterative optimization method to find the optimal parameter vector. The inner-layer algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer-layer algorithm adjusts the core hyperparameters of the inner-layer algorithm based on real-time key statistical features. Through continuous iteration, the nested mechanism of outer-layer control and inner-layer search works collaboratively to search for the optimal parameter vector that minimizes the error between the output wave data and the observed wave data. This constructs a dual-layer closed-loop interactive system and a wave model parameter optimization system with intelligent adjustment, closed-loop feedback, and continuous evolution characteristics. It achieves complete automation, self-learning, and high-efficiency intelligence in the parameter optimization process and can be widely applied to high-precision parameter optimization and automatic calibration of ocean wave models. Attached Figure Description

[0023] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0024] Figure 1 This is a flowchart of the method for optimizing wave parameters based on internal and external double-layer iterative optimization according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the system structure of wave parameters based on internal and external double-layer iterative optimization according to an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0027] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0028] To improve the intelligence level of parameter optimization, existing technologies have introduced the intelligent optimization method of Genetic Algorithm (GA) into wave model parameter calibration. While GA possesses good global search capabilities, it is prone to getting trapped in local optima, and its search performance is heavily dependent on the settings of hyperparameters such as crossover rate, mutation rate, and population size, thus affecting the accuracy of wave model parameter optimization results. Reinforcement learning utilizes the cyclical interaction between the agent and the environment, learning optimal behavioral strategies through a reward-driven mechanism, making it particularly suitable for adaptive optimization of complex and nonlinear systems. However, directly searching for wave model parameters using reinforcement learning faces challenges due to its high dimensionality and computational cost, thereby reducing the efficiency of wave model parameter optimization.

[0029] Figure 1 A schematic diagram of the wave parameter optimization method based on inner and outer double-layer iterative optimization according to an embodiment of the present invention is provided. Figure 1 The method for optimizing wave parameters based on internal and external double-layer iterative optimization in this embodiment may include the following steps S101~S102.

[0030] The specific implementation process of steps S101 to S102 is as follows: Step S101: Based on the wave pattern with the spectral energy balance equation as the core, construct a wave model; the wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus constitute the parameter vector to be optimized.

[0031] It should be noted here that the parameter vector to be optimized includes, but is not limited to, the wind input coefficient, the white crown dissipation coefficient, and the bottom friction coefficient.

[0032] In this embodiment of the invention, the wave model adopted is the SWAN (Simulating Waves Nearshore) wave model, which is a wave model with the spectral energy balance equation as its core. The basic equation is:

[0033] In the formula For wave action density spectrum, To find the partial derivative in the horizontal direction; Group velocity; The amplitude of the wave group; The discretization parameter for frequency; The discretization parameter is for the direction; The wave frequency; The direction of the wave; The source and sink terms include physical processes such as wind energy input, white-hat dissipation, bottom friction, and nonlinear wave-wave interaction.

[0034] Source and Exchange Represented as:

[0035] in, This represents the wave energy input under wind action; and These represent the nonlinear wave energy transfer from three-wave and four-wave sources, respectively. Indicates the dissipation term of white waves; This represents the bottom friction dissipation term; This represents the dissipation term caused by wave breaking.

[0036] The wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized, thus forming a parameter vector to be optimized:

[0037] For example It can be: For wind input coefficient, For the dissipation coefficient of the white crown, The coefficient of friction is set as the base, and each parameter satisfies the corresponding constraints based on empirical parameters.

[0038] The goal of wave parameter optimization is to find the optimal parameter set. This minimizes the error between the model output and the observed data.

[0039] Step S102: Use the inner and outer double-layer iterative optimization method to find the optimal parameter vector that minimizes the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is represented by a regularized error function, which is the normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer double-layer iterative optimization method, the inner layer algorithm searches for the optimal parameter vector that minimizes the regularized error function, and the outer layer algorithm adjusts the core hyperparameters of the inner layer algorithm through the key statistical features of the inner layer algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

[0040] In this embodiment of the invention, the regularization error function for:

[0041]

[0042] in, For Normalized Weighted Mean Square Error (MSE); For regularization weights; Let j be the parameter to be optimized; The parameter to be optimized is the empirical value; n is the number of parameters to be optimized. Allowable deviation; This represents the number of observation points; No. Data duration for each observation point; The first simulation of the model Each observation point is at The wave height or period value at any given moment; These are the corresponding measured values; The weighting coefficients for the observation points. ; To prevent small constants with a denominator of zero.

[0043] This prevents the parameters from deviating excessively from the physically reasonable range and measures the degree of matching between the model output and the measured wave data.

[0044] In one or more embodiments, the inner algorithm employs a genetic algorithm, treating a set of parameters to be optimized as an individual.

[0045] Genetic algorithms undertake the task of global search of the parameter space, and their core objective is to minimize the regularization error function. This allows for the acquisition of the optimal parameter combination. The evolutionary process of a genetic algorithm includes individual representation, fitness calculation, selection, crossover, and mutation, among other steps. Its hyperparameters (including population size) are crucial for achieving this. Cross rate Variation rate Variation intensity (etc.) is dynamically regulated by the upper-level reinforcement learning agent to balance the exploratory and convergent aspects of the search at different stages.

[0046] Set a set of model parameters as a single entity:

[0047] in, Indicates the first Individual (parameter combination); Represents the first individual The gene (parameter) corresponds to the in the wave model. One parameter to be optimized; For parameter dimensions.

[0048] Real-valued encoding is used to ensure the physical meaning of the parameters and the continuity of calculation. Each parameter value must satisfy the physical boundary constraints. After individual initialization and crossover / mutation operations, if the parameters exceed the limits, they are constrained to the legal range by truncation or bounce projection. The initial population is generated by sampling with a uniform distribution.

[0049] The fitness function primarily transforms the minimization problem into a fitness maximization problem, and prevents division by zero by adding 1 to the denominator, thereby minimizing the model error. The fitness function is defined as follows: ; in, Represents an individual The regularization error value. The smaller this error term, the higher the individual fitness.

[0050] The selection operator is used to determine the parent individuals from the existing population that will participate in parameter transmission. This invention employs a roulette wheel selection mechanism for probability calculation to ensure that reasonable parameter combinations have a higher probability of being preserved and passed on. The selection probability is defined as:

[0051] in, Indicates the first The probability of an individual being selected; For population size. Individuals with higher fitness have a higher probability of selection, thus being more likely to pass on their parameters to the next generation.

[0052] Crossover operations are used for information recombination, generating new child parameter combinations from two parent parameter combinations. This invention employs... The (Blend Crossover) operator is expressed as follows:

[0053] in, and The two parent individuals are in the first generation. The values ​​that can be taken on each parameter; This is the expansion factor (typically 0.3). This represents a uniformly distributed random sampling over a given interval; For the generated offspring in the first The new values ​​for each parameter.

[0054] By introducing random perturbations among different parameter combinations through mutation operations, the diversity of parameter combinations is maintained, effectively preventing the agent from getting trapped in local optima. This invention employs a mutation strategy based on a Gaussian distribution, the mathematical form of which is:

[0055] in, This indicates that the mean is 0 and the variance is 0. Gaussian noise; For the first The intensity of variation of each parameter.

[0056] It should be noted that other search algorithms, such as ant colony optimization, can also be used for the inner layer algorithm, but these will not be discussed in detail here.

[0057] In one or more embodiments, the outer layer algorithm employs a reinforcement learning algorithm, which uses the key statistical features of the inner layer algorithm in real time to form an environmental state and output actions to adjust the core hyperparameters of the inner layer algorithm.

[0058] The genetic algorithm acts as the lower-level searcher, and its behavior strategy is dynamically controlled by the upper-level reinforcement learning agent. The reinforcement learning employs the Soft Actor-Critic (SAC) algorithm, which monitors key statistical features of the genetic algorithm in real time (such as optimal error). Average error Error variance Intergenerational improvement range (etc.), forming an environmental state. and output the action. To adjust the core hyperparameters of GA.

[0059] The environment state of a reinforcement learning algorithm consists of the optimal error, average error, error variance, and intergenerational improvement magnitude; the output action consists of the population size, crossover rate, mutation rate, and mutation intensity.

[0060]

[0061]

[0062] in, For population size, Crossover rate, The variability rate The value represents the intensity of variation.

[0063] When reinforcement learning algorithms get stuck in local convergence or experience a decline in diversity, the mutation rate or mutation intensity is increased to promote exploration. Simultaneously, the population size or crossover rate is increased to expand the search space or promote recombination and information exchange among individuals, thereby enhancing population diversity. When the error continues to decrease and the optimization direction stabilizes, parameters such as population size, crossover rate, mutation rate, and mutation intensity are reduced to improve search efficiency. Through this reinforcement learning-driven adaptive control mechanism, the genetic algorithm can automatically adjust its strategy based on feedback from the search process, achieving a dynamic evolutionary process from global exploration to local refinement, thus achieving higher accuracy and stability in the optimization of SWAN wave model parameters.

[0064] Reinforcement learning and genetic algorithms work together through a nested mechanism of "outer layer regulation - inner layer search": Initialize the parameter population and reinforcement learning state; Genetic algorithms perform several generations of iterations; Summarize error characteristics to form a state vector ; reinforcement learning agents are based on Output Action Update GA hyperparameters; The genetic algorithm then proceeds to the next stage of evolution based on this information. Continue iterating until the requirements are met or the iteration limit is reached.

[0065] After training is complete, the reinforcement learning policy network is integrated with the SWAN wave model to achieve automated calibration and continuous optimization of wave model parameters.

[0066] In this embodiment, the genetic algorithm acts as the lower-level optimizer, responsible for performing a global search in a complex parameter space. The reinforcement learning algorithm acts as the upper-level intelligent regulator, adjusting the core parameters of the genetic algorithm (such as crossover rate, mutation rate, and population size) based on its performance in each generation of the search. This ensures high exploratory activity in the early stages and enhanced convergence in later stages, thereby achieving efficient approximation of the global optimum. This scheme integrates the adaptive learning capability of reinforcement learning with the global search capability of the genetic algorithm, constructing a wave model parameter optimization system with intelligent adjustment, closed-loop feedback, and continuous evolution characteristics. It is widely applicable to high-precision parameter optimization and automatic calibration of ocean wave models.

[0067] It should be noted that the outer layer algorithm can also be implemented using other existing deep learning models. The deep learning model only needs to be trained to learn the correlation between the key statistical features of the inner layer algorithm and the core hyperparameters of the inner layer algorithm.

[0068] like Figure 2 As shown, the system for optimizing wave parameters based on internal and external double-layer iterative optimization provided in this embodiment of the invention can be implemented in software. The system for optimizing wave parameters based on internal and external double-layer iterative optimization includes the following software modules: wave model construction module 201 and parameter vector optimization module 202.

[0069] The functions of each software module in the system based on internal and external two-layer iterative optimization of wave parameters are described below: The wave model construction module 201 is used to construct a wave model based on the wave pattern with the spectral energy balance equation as the core. The wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus constitute the parameter vector to be optimized. The parameter vector optimization module 202 is used to find the optimal parameter vector using an inner and outer double-layer iterative optimization method to minimize the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is characterized by a regularized error function, which is a normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer two-layer iterative optimization method, the inner algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer algorithm adjusts the core hyperparameters of the inner algorithm by using the key statistical features of the inner algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

[0070] It should be noted that each module in the system based on internal and external double-layer iterative optimization of wave parameters in the embodiments of the present invention corresponds one-to-one with each step in the method based on internal and external double-layer iterative optimization of wave parameters in the above embodiments, and their specific implementation processes are the same, so they will not be repeated here.

[0071] The structure of the electronic device according to an embodiment of the present invention will be described in detail below. Figure 3 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present invention. It can be understood that... Figure 3 The diagram shows only an exemplary structure of the electronic device, not the entire structure. Some or all of the structures shown may be implemented as needed.

[0072] The electronic device provided in this embodiment of the invention includes: at least one processor 301, a memory 302, a user interface 303, and at least one network interface 304. The various components in the system based on internal and external dual-layer iterative optimization of wave parameters are coupled together through a bus system 305. It can be understood that the bus system 305 is used to realize the connection and communication between these components. In addition to a data bus, the bus system 305 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 305.

[0073] The user interface 303 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0074] It is understood that memory 302 can be volatile memory or non-volatile memory, or both. In this embodiment of the invention, memory 302 is capable of storing data to support the operation of the terminal. Examples of this data include any computer programs used to operate on the terminal, such as operating systems and applications. The operating system includes various system programs, such as framework layers, core library layers, driver layers, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications.

[0075] In some embodiments, the system for optimizing wave parameters based on internal and external double-layer iterative optimization provided in this invention can be implemented using a combination of hardware and software. As an example, the system for optimizing wave parameters based on internal and external double-layer iterative optimization provided in this invention can be a processor in the form of a hardware decoding processor, programmed to execute the method for optimizing wave parameters based on internal and external double-layer iterative optimization provided in this invention. For example, the processor in the form of a hardware decoding processor can employ one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0076] As an example, processor 301 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., wherein the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0077] As an example of the hardware implementation of the system for optimizing wave parameters based on internal and external double-layer iterative optimization provided in the embodiments of the present invention, the device provided in the embodiments of the present invention can be directly executed by a processor 301 in the form of a hardware decoding processor. For example, it can be executed by one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components to implement the method for optimizing wave parameters based on internal and external double-layer iterative optimization provided in the embodiments of the present invention.

[0078] The memory 302 in this embodiment of the invention is used to store various types of data to support the operation of a system based on inner and outer two-layer iterative optimization of wave parameters, or to store data for execution. Figure 1 The program code for the method shown. Examples of this data include: any executable instructions for operating on a system based on inner and outer double-layer iterative optimization of wave parameters, such as executable instructions that can be included in the executable instructions to implement the method for inner and outer double-layer iterative optimization of wave parameters according to embodiments of the present invention.

[0079] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including functions for executing... Figure 1 The program code for the method shown. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by the central processing unit, it performs the various functions defined in the apparatus of this application.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0081] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing wave parameters based on internal and external double-layer iterative optimization, characterized in that, include: Based on the wave pattern with the spectral energy balance equation as the core, a wave model is constructed; the wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus constitute the parameter vector to be optimized. The optimal parameter vector is found by using an internal and external double-layer iterative optimization method to minimize the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is characterized by a regularized error function, which is a normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer two-layer iterative optimization method, the inner algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer algorithm adjusts the core hyperparameters of the inner algorithm by using the key statistical features of the inner algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

2. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 1, characterized in that, The parameter vector to be optimized includes the wind input coefficient, the white crown dissipation coefficient, and the bottom friction coefficient.

3. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 1, characterized in that, Regularization error function for: in, This is the normalized weighted mean square error; For regularization weights; Let j be the parameter to be optimized; The parameter to be optimized is the empirical value; n is the number of parameters to be optimized. Allowable deviation; This represents the number of observation points; No. Data duration for each observation point; The first simulation of the model Each observation point is at The wave height or period value at any given moment; These are the corresponding measured values; The weighting coefficients for the observation points. ; To prevent small constants with a denominator of zero.

4. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 1, characterized in that, The inner algorithm uses a genetic algorithm, treating a set of parameters to be optimized as an individual; the fitness function is set to... : ; in, Represents an individual The regularization error value.

5. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 1, characterized in that, The outer algorithm employs reinforcement learning, which uses the key statistical features of the inner algorithm in real time to form the environmental state and output actions to adjust the core hyperparameters of the inner algorithm.

6. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 5, characterized in that, The environment state of a reinforcement learning algorithm consists of the optimal error, average error, error variance, and intergenerational improvement magnitude; the output action consists of the population size, crossover rate, mutation rate, and mutation intensity.

7. The method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in claim 5, characterized in that, When reinforcement learning algorithms get stuck in local convergence or diversity declines, increase the mutation rate or mutation intensity to promote exploration, while increasing the population size or crossover rate to expand the search space or promote recombination and information exchange between individuals and improve population diversity. When the error continues to decrease and the optimization direction is stable, reduce parameters such as population size, crossover rate, mutation rate and mutation intensity to improve search efficiency.

8. A system for optimizing wave parameters based on internal and external double-layer iterative optimization, characterized in that, The method based on the inner and outer double-layer iterative optimization of wave parameters as described in any one of claims 1-7 includes: The wave model construction module is used to construct wave models based on wave patterns with the spectral energy balance equation as the core. The wave model contains several empirical parameters and adjustment coefficients, which are the parameters to be optimized and thus form the parameter vector to be optimized. The parameter vector optimization module is used to find the optimal parameter vector using an internal and external double-layer iterative optimization method to minimize the error between the output wave data and the observed wave data of the wave model. The error between the output wave data and the observed wave data is characterized by a regularized error function, which is a normalized weighted mean square error with a prior constraint regularization term. In the process of finding the optimal parameter vector using the inner and outer two-layer iterative optimization method, the inner algorithm searches for the optimal parameter vector that minimizes the regularization error function, while the outer algorithm adjusts the core hyperparameters of the inner algorithm by using the key statistical features of the inner algorithm in real time, and continues to iterate until the preset requirements are met or the iteration limit is reached.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the method for optimizing wave parameters based on inner and outer double-layer iterative optimization as described in any one of claims 1-7.

Images