A method, apparatus, device, and medium for generating marine physical information

Abstract

The present application discloses a method, apparatus, device and medium for generating marine physical information, relating to the field of data generation. The method includes: inputting a random noise vector for information generation into a marine physical information generation model to obtain the generated marine physical information; a method for determining the marine physical information generation model, including: using actual marine physical information and a random noise vector for training, training a generative adversarial network with the goal of minimizing a loss function to obtain the marine physical information generation model; the generative adversarial network includes a generator and a discriminator; the generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module and a first neural network connected in sequence; the loss function at least includes: a physical loss; the physical loss is determined by performing physical differentiation calculation on the marine fluid in the marine physical information output by the first neural network. The present application can reduce the calculation cost and improve the reliability of the generated data.

Description

Technical Field

[0001] The present application relates to the field of data generation, and particularly to a method, apparatus, device, and medium for generating ocean physical information. Background Art

[0002] Currently, the generation and simulation of ocean physical information mainly rely on physical modeling and data-driven methods. Physical modeling methods (such as numerical simulation methods based on partial differential equations) can accurately describe ocean dynamics processes, but they have high computational costs and are sensitive to complex and changing boundary conditions. Data-driven methods, whose statistical models are fitted through historical data, can quickly generate results, but when the data is insufficient or the environmental conditions change, their prediction effects significantly decline, so their reliability is poor. Summary of the Invention

[0003] The purpose of the present application is to provide a method, apparatus, device, and medium for generating ocean physical information, which can reduce the computational cost and improve the reliability of the generated data.

[0004] To achieve the above purpose, the present application provides the following solutions.

[0005] In a first aspect, the present application provides a method for generating ocean physical information, including: obtaining a random noise vector for information generation; inputting the random noise vector for information generation into an ocean physical information generation model to obtain the generated ocean physical information.

[0006] Among them, the determination method of the ocean physical information generation model includes: obtaining actual ocean physical information and a random noise vector for training; using the actual ocean physical information and the random noise vector for training to train a generative adversarial network with the goal of minimizing the loss function to obtain an ocean physical information generation model.

[0007] The generative adversarial network includes a generator and a discriminator.

[0008] The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence.

[0009] The adaptive spectrum module is used to process the random noise vector for training by using fast Fourier transform; the attention module is used to perform operations on the processed data by using an attention mechanism; the first neural network is used to perform gradient operations on the data after the operations and output ocean physical information.

[0010] The discriminator is used to discriminate the ocean physical information output by the first neural network according to the actual ocean physical information to obtain a discrimination result.

[0011] The loss function at least includes: a physical loss; the physical loss is determined by performing physical differential calculation on the ocean fluid in the ocean physical information output by the first neural network.

[0012] The trained generator serves as the ocean physical information generation model.

[0013] In a second aspect, the present application provides an ocean physical information generation device, including: a data acquisition module for acquiring a random noise vector for information generation; a model determination module for determining an ocean physical information generation model; and an information generation module for inputting the random noise vector for information generation into the ocean physical information generation model to obtain the generated ocean physical information.

[0014] The model determination module includes: a data acquisition unit for acquiring actual ocean physical information and a random noise vector for training; and a model training unit for training a generative adversarial network with the actual ocean physical information and the random noise vector for training with the goal of minimizing the loss function to obtain an ocean physical information generation model.

[0015] The generative adversarial network includes a generator and a discriminator.

[0016] The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence.

[0017] The adaptive spectrum module is used to process the random noise vector for training by using fast Fourier transform; the attention module is used to perform operations on the processed data by using an attention mechanism; and the first neural network is used to perform gradient operations on the data after the operations and output ocean physical information.

[0018] The discriminator is used to discriminate the ocean physical information output by the first neural network according to the actual ocean physical information to obtain a discrimination result.

[0019] The loss function at least includes: a physical loss; the physical loss is determined by performing physical differential calculation on the fluid in the ocean physical information output by the first neural network.

[0020] The trained generator serves as the ocean physical information generation model.

[0021] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the above-mentioned ocean physical information generation method.

[0022] Fourthly, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned method for generating ocean physical information is implemented.

[0023] According to the specific embodiments provided by the present application, the present application has the following technical effects: The present application provides a method, device, equipment and medium for generating ocean physical information. By using actual ocean physical information and a random noise vector for training, a generative adversarial network is trained with the goal of minimizing the loss function to obtain an ocean physical information generation model. Inputting the random noise vector for information generation into the ocean physical information generation model to obtain the generated ocean physical information can reduce the computational cost; the generator in the generative adversarial network uses a physical information neural network, and a physical loss is introduced into the loss function, which can generate ocean physical information that conforms to the basic physical laws of the actual ocean process. While maintaining physical consistency, high-quality ocean physical information is generated, improving the reliability of the generated data. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 It is an application environment diagram of a method for generating ocean physical information in an embodiment of the present application.

[0026] Figure 2 It is a schematic flowchart of a method for generating ocean physical information provided in an embodiment of the present application.

[0027] Figure 3 It is a schematic flowchart of a method for determining an ocean physical information generation model provided in an embodiment of the present application.

[0028] Figure 4 It is a schematic structural diagram of a generative adversarial network provided in an embodiment of the present application.

[0029] Figure 5 It is a schematic structural diagram of a generator in a generative adversarial network provided in an embodiment of the present application.

[0030] Figure 6 It is a schematic diagram of functional modules of an ocean physical information generation device provided in another embodiment of the present application.

[0031] Figure 7 It is a schematic structural diagram of a computer device provided in an embodiment of the present application. Detailed implementation manners

[0032] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0033] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0034] Currently, although the related marine physical information generation methods can meet the generation requirements of marine physical information to a certain extent, there are still some problems. Although the physical modeling method can generate data that conforms to physical laws, it consumes a huge amount of computing resources and is difficult to quickly respond to the actual application requirements. Therefore, there is a problem of high computing cost. The complexity of the marine environment and the difficulty of data collection result in limited available observation data, and the statistical models of existing data-driven methods perform poorly in the case of scarce data. Therefore, there is a problem of scarce data. Although the traditional Generative Adversarial Networks (GAN) can generate realistic data, due to the lack of physical constraints, the generated data often does not conform to the basic physical laws of the actual marine process in terms of physical meaning, which limits its reliability in actual applications. Therefore, there is a problem of lack of physical consistency. In addition, other existing models have limited generalization ability when processing marine data in different geographical locations or time periods, and the generated results do not have universality. Therefore, there is a problem of insufficient generalization ability.

[0035] In view of the above problems, this embodiment provides a new method that can combine the advantages of physical information and data-driven models to generate marine physical information with strong physical consistency and high quality. By introducing Physics-Informed Neural Networks (PINN), not only can physical constraints be introduced during the GAN generation process, but also limited observation data can be effectively utilized, significantly improving the quality and practicality of the generated data. PINN is a machine learning model that combines deep learning and physics knowledge. This embodiment combines PINN and GAN to generate marine physical information, solving many deficiencies in the prior art and providing more reliable data support for marine scientific research and applications.

[0036] The marine physical information generation method provided in the embodiments of the present application can be applied to, for example Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set separately, integrated on the server 104, placed on the cloud or other servers. The terminal 102 can send the random noise vector for information generation to the server 104. After receiving the random noise vector for information generation, for the random noise vector for information generation, the server 104 inputs the random noise vector for information generation into the ocean physical information generation model to obtain the generated ocean physical information. The method for determining the ocean physical information generation model includes: obtaining the actual ocean physical information and the random noise vector for training; using the actual ocean physical information and the random noise vector for training to train the generative adversarial network with the goal of minimizing the loss function to obtain the ocean physical information generation model. The generative adversarial network includes a generator and a discriminator. The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence. The loss function at least includes: a physical loss; the physical loss is determined by performing physical differential calculation on the ocean fluid in the ocean physical information output by the first neural network.

[0037] The server 104 can feedback the obtained generated ocean physical information to the terminal 102. In addition, in some embodiments, the ocean physical information generation method can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly process the random noise vector for information generation, or the server 104 can obtain the random noise vector for information generation from the data storage system and process the random noise vector for information generation.

[0038] Among them, the terminal 102 can be, but is not limited to, various desktop computers, laptop computers, smartphones, tablets, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers, and can also be a cloud server.

[0039] In an exemplary embodiment, as Figure 2 shown, a method for generating ocean physical information is provided. This method is executed by a computer device, and can be specifically executed alone by a computer device such as a terminal or a server, or jointly executed by a terminal and a server. In the embodiments of the present application, taking this method applied to Figure 1 the server 104 in it as an example for illustration, it includes the following steps 201 to step 202.

[0040] Step 201: Obtain a random noise vector for information generation.

[0041] Step 202: Input the random noise vector for information generation into the ocean physical information generation model to obtain the generated ocean physical information. The ocean physical information includes: coastline cyclone index, revetment ratio, wetland ratio, temperature and humidity field, ocean current field, pressure field, salinity, and coastal zone resilience evaluation results, etc.

[0042] As Figure 3 shown, the method for determining the ocean physical information generation model includes the following steps.

[0043] Step 301: Obtain actual ocean physical information and random noise vectors for training.

[0044] Step 302: Use the actual ocean physical information and random noise vectors for training to train the generative adversarial network with the goal of minimizing the loss function to obtain the ocean physical information generation model.

[0045] The generative adversarial network includes a generator and a discriminator. The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence. The trained generator serves as the ocean physical information generation model.

[0046] The adaptive spectrum module is used to process the random noise vector for training using the Fast Fourier Transform (FFT); the attention module is used to perform operations on the processed data using the attention mechanism; the first neural network is used to perform gradient operations on the processed data and output ocean physical information.

[0047] The discriminator is used to discriminate the ocean physical information output by the first neural network based on the actual ocean physical information to obtain a discrimination result.

[0048] The loss function at least includes: physical loss; the physical loss is determined by performing physical differential calculations on the ocean fluid in the ocean physical information output by the first neural network.

[0049] In another exemplary embodiment of the present application, the loss function includes: generator loss and discriminator loss; the generator loss includes: generator itself loss, cross-entropy loss, and physical loss.

[0050] Step 302 specifically includes the following steps.

[0051] (1) Preprocess and encode the actual ocean physical information to obtain the encoded data.

[0052] Specifically: ① Process the missing values and outliers in the actual ocean physical information to obtain the preprocessed data. ② Perform one-hot encoding on the discrete variables (such as the coastal resilience evaluation results) in the preprocessed data to obtain the first encoded data. One-hot encoding can convert the discrete variable into an independent binary vector to adapt to the input of the generative adversarial network. ③ Normalize or standardize the continuous variables in the preprocessed data to obtain the processed continuous data. The purpose of normalizing or standardizing the continuous data in this step is to convert the data into a normal distribution with a mean of 0 and a standard deviation of 1 or scale the data to [0, 1]. Among them, to address the problem of imbalanced class distribution, standardization is performed separately for each class. ④ Perform binary conversion encoding on the processed continuous data to obtain the second encoded data. This step converts the categorical variable into an independent binary vector through encoding. If the ocean physical information categorical variable has n different classes, the variable will be converted into a binary vector of length n, where only one position is 1 and the rest are 0. ⑤ Determine the first encoded data and the second encoded data as the encoded data.

[0053] Among them, the process of dealing with missing values and outliers is as follows: Use the box plot method to remove outliers or truncate outliers; for the actual ocean physical information with missing values, use mean filling, median filling, or filling through regression, etc. Processing missing values helps to avoid performance degradation of the generative adversarial network during training due to data missing.

[0054] Among them, the normalization process is implemented using Mode-specific Normalization. Specifically: First, for each ocean physical information, count the frequencies of each category. Assume that a certain categorical variable has multiple categories, count the frequencies of each category. For each continuous variable, calculate its mean and standard deviation.

[0055] The process of standardization is: Use the formula to implement, where is the mean, is the variance. During the data generation stage, the generator will output a standardized continuous variable . At this time, it is necessary to inverse standardize back to the scale of the original data, , is the original data.

[0056] (2) For the a-th iteration in the training process, input the random noise vector used for training in the a-th iteration into the physics-informed neural network, and the first neural network in the physics-informed neural network outputs the ocean physics information of the a-th iteration; where a > 1.

[0057] (3) Input the encoded data and the ocean physics information of the a-th iteration into the discriminator, and the discriminator outputs the discrimination result of the a-th iteration.

[0058] (4) Determine the generator's own loss of the a-th iteration according to the ocean physics information of the a-th iteration output by the first neural network, determine the cross-entropy loss of the a-th iteration according to the ocean physics information of the a-th iteration output by the first neural network and the encoded data, and perform physical differential calculation on the fluid in the ocean physics information of the a-th iteration output by the first neural network to determine the physical loss of the a-th iteration.

[0059] (5) Determine the generator loss of the a-th iteration according to the generator's own loss of the a-th iteration, the cross-entropy loss of the a-th iteration, and the physical loss of the a-th iteration; determine the discriminator loss of the a-th iteration according to the discrimination result of the a-th iteration.

[0060] Specifically, the calculation formula for the generator loss is as follows.

[0061] 。

[0062] Where, represents the generator loss; represents the generator's own loss; represents the cross-entropy loss, which is used to measure the difference between the ocean physics information generated by the generator and the real data distribution in certain feature dimensions; represents the physical loss; represents the i-th ocean physics information generated; m represents the number of ocean physics information generated; represents the discrimination result of the discriminator for the i-th ocean physics information generated.

[0063] The physical loss is determined by performing physical differential calculation on the ocean fluid in the ocean physics information output by the first neural network. The physical differential equation for calculation is the Navier-Stokes differential equation, and its expression is as follows.

[0064] 。

[0065] represents the ocean fluid velocity field in the generated ocean physics information; t represents time; represents taking the partial derivative, represents the gradient operator, ( , ), where (x, y, z) represents the spatial coordinates of a point in the three-dimensional space where the physical quantity is defined; represents the kinematic viscosity of the ocean fluid in the generated ocean physical information; represents the density of the ocean fluid in the generated ocean physical information; represents the ocean fluid pressure field in the generated ocean physical information; represents the external force corresponding to the generated ocean physical information.

[0066] According to the above Navier - Stokes differential equation, the generated ocean physical information should conform to the physical laws therein. Therefore, the physical loss is used as part of the loss function as a physical constraint. The calculation formula of the physical loss is as follows.

[0067] .

[0068] Among them, represents the ocean fluid velocity field in the i - th generated ocean physical information; represents the ocean fluid pressure field in the i - th generated ocean physical information; represents the external force corresponding to the i - th generated ocean physical information.

[0069] The calculation formula of the discriminator loss is as follows.

[0070] .

[0071] represents the discriminator loss; represents the actual i - th ocean physical information; represents the discrimination result of the discriminator for the actual i - th ocean physical information.

[0072] (6) If the difference between the generator loss of the a - th iteration and the generator loss of the previous iteration is within the first set difference range, and the difference between the discriminator loss of the a - th iteration and the discriminator loss of the previous iteration is within the second set difference range, then it is determined that the loss function of the a - th iteration is the smallest, and the generator after the a - th iteration is used as the ocean physical information generation model. Otherwise, after updating the iteration number, the next iteration is performed. In this embodiment, the training of the model is achieved through forward propagation and backward update.

[0073] In another exemplary embodiment of the present application, the generator and the discriminator are mainly introduced.

[0074] The generator in this embodiment is a physics-informed neural network; the physics-informed neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence. The adaptive spectrum module is connected to the first neural network with a residual connection. The purpose of the residual connection is to prevent the degradation of the deep neural network. By introducing a skip connection, the problem of gradient vanishing is alleviated, enabling deeper networks to be trained more effectively and improving the training efficiency, convergence speed, generalization ability, and overall performance of the model. The first neural network includes: a first input layer, a hidden layer, and a first output layer; the hidden layer includes 16 layers of hidden neurons, and a residual connection is made between the last layer of hidden neurons and the middle layer of hidden neurons for improving the training of the model. The discriminator is a second neural network, and the second neural network includes: a second input layer, a hidden layer, and a second output layer. The structure of the second input layer is the same as that of the first input layer, and the structure of the second output layer is different from that of the first output layer.

[0075] Set the initial complex weight ( and ) in the adaptive spectrum module as a random floating-point tensor. The shape of this random floating-point tensor is (dim, 2), where dim represents the attribute of obtaining the tensor dimension, and the shape of this random floating-point tensor is a two-dimensional tensor. Multiply the initial complex weight by 0.02 for scaling and further initialize it using a truncated normal distribution with a standard deviation set to 0.02. Initialize the threshold parameter as a random floating-point tensor with a shape of (dim, 1), and the shape of this random floating-point tensor is a one-dimensional tensor. The initial complex weight satisfies .

[0076] The adaptive spectrum module receives an input tensor with a shape of (B, N, C), and this input tensor is a random noise vector, where B represents the batch size, N represents the length of the sequence, and C represents the number of features.

[0077] Use FFT to transform the random noise vector (which is a time-domain signal) into a frequency-domain signal, and the calculation formula is as follows.

[0078] .

[0079] is the frequency-domain signal; is the time-domain signal; represents the imaginary unit; represents the index in the frequency domain; represents the index in the time domain; when performing the frequency-domain conversion, the length N of the sequence is the signal length in the FFT.

[0080] Next, perform frequency-domain weighting. Frequency-domain weighting is to multiply the frequency-domain signal obtained by FFT with the complex weight, and the calculation formula is as follows.

[0081] 。

[0082] Among them, represents the frequency-domain weighting result; is and constructed weight sequence for frequency-domain weighting; and correspond to the weights of the real and imaginary parts respectively.

[0083] Next, energy calculation is performed: 。Among them, represents energy.

[0084] Next, normalization of energy is performed: 。Among them, represents the normalized energy, is the median energy, is a very small constant to avoid division by zero. The normalized energy and the threshold parameter are compared and subtracted to obtain a difference tensor. The difference tensor is converted into a floating-point tensor, and then the floating-point tensor is subjected to a sigmoid operation to be converted into a soft mask between (0~1) , and the calculation formula is: , where c represents the floating-point tensor converted from the difference tensor.

[0085] Then, the masked frequency-domain signal is calculated according to the soft mask, and the calculation formula is: 。Among them, represents the amplitude in the frequency domain, represents the phase in the frequency domain. The real part and the imaginary part of ; 。

[0086] The masked frequency-domain signal can be converted back to the time-domain signal through the inverse fast Fourier transform (IFFT), and the calculation formula is: , is the converted time-domain signal, that is, the processed data output by the adaptive spectrum module.

[0087] Next, the processed data output by the adaptive spectrum module , , , where: , is the query matrix, is the key matrix, is the value matrix, , and respectively represent , and the corresponding linear transformation matrices, , represents rows and columns of the set of real numbers, represents rows and columns

[0088] Calculate the dot product between all and by matrix multiplication, that is, calculate , is the transpose of the key matrix, and the result of this step is a score matrix, where each element represents the dot product score between an element in a query matrix and an element in a key matrix. The calculation formula is as follows.

[0089] .

[0090] .

[0091] represents the score matrix, represents the calculated attention weights, represents the query vector, represents the first key vector, represents the th key vector, represents the th key vector, represents the indices corresponding to different elements of the key vector, represents the dimension of the key vector, represents the set of elements of the key vector, represents the normalized exponential function.

[0092] For each row of the score matrix (i.e., each (corresponding scores), select the k1 elements with the highest scores. The selection method is that for each row of scores, randomly select an element as the reference value, divide the scores of this row into two parts, one part is less than the reference value, and the other part is greater than or equal to the reference value. According to the position of the reference value, determine whether to recursively search in a certain part. Specifically, if the reference value is exactly the K1-th largest score, the search ends; if the position of the reference value is greater than K1, search for the K1-th largest score in the part less than the reference value; if the position of the reference value is less than K1, search for the (K1 - left)-th largest score in the part greater than or equal to the reference value, where left is the number of positions less than the reference value. Repeat the above steps until the K1-th largest score is found, and the selected K1 values are represented as follows.

[0093] .

[0094] Among them, represents the element located in the -th row and -th column of the processed matrix; represents the element located in the -th row and -th column of the original score matrix ; represents the set of the -th row of the original score matrix with the highest scores.

[0095] According to the above , use the softmax function to convert the scores into normalized attention weights. Since the non-highest scores in are set to very small numbers, their contributions in the softmax function will be close to zero. Then, multiply these attention weights by the original value matrix

[0096] .

[0097] Among them, is the data after the operation output by the attention module.

[0098] Next, input the data after the operation into the first neural network, and the calculation formula of each neuron is as follows.

[0099] .

[0100] Among them, represents the output of the neuron; represents the weight; represents the input of the neuron; represents the bias.

[0101] Perform gradient operations according to the calculation formula of each neuron. The calculation formula is as follows.

[0102] .

[0103] .

[0104] Among them, represents the output of the 1st hidden layer in the first neural network; represents the output of the 2nd hidden layer in the first neural network; represents the output of the 3rd hidden layer in the first neural network; represents the output of the 4th hidden layer in the first neural network; represents the output of the 5th hidden layer in the first neural network; represents the output of the 15th hidden layer in the first neural network; represents the output of the 16th hidden layer in the first neural network.

[0105] Each gradient calculation implements one iteration: ; . represents the weight parameter between the neurons in the first input layer and the neurons in the hidden layer; represents the weight parameter between the neurons in the hidden layer and the neurons in the first output layer; represents the learning rate; represents the partial derivative of the loss function with respect to the weight.

[0106] The structure of the generative adversarial network is as Figure 4 shown, and the structure of the generator in the generative adversarial network is as Figure 5 shown.

[0107] In another exemplary embodiment of the present application, after step 201, the marine physical information generation method further includes: decoding and verifying the generated marine physical information with actual marine physical information, and determining the marine physical information that passes the verification as the finally generated marine physical information.

[0108] Specifically, the process of decoding and verification is as follows: extract the characteristic statistics of the actual ocean physical information to obtain the first characteristic statistics; extract the characteristic statistics of the generated ocean physical information to obtain the second characteristic statistics; determine the fidelity between the first characteristic statistics and the second characteristic statistics. If the fidelity reaches the set degree value, the generated ocean physical information is determined as the verified ocean physical information. Among them, the characteristic statistics include: mean, standard deviation, correlation coefficient, etc.

[0109] This application compares the characteristic statistics of the generated ocean physical information and the actual ocean physical information, and can evaluate the fidelity of the generated ocean physical information in different physical quantities.

[0110] This application combines the physics-informed neural network and the generative adversarial network to realize the generation of ocean physical information. It can generate high-quality ocean physical information while maintaining physical consistency, and effectively solves multiple problems existing in the prior art.

[0111] Specifically, the ocean physical information generation method of this application can provide high-quality data generation solutions in fields such as ocean data missing, environmental monitoring, and climate prediction, significantly improving the reliability and applicability of data. By introducing PINN and adding the constraints of physical equations during the data generation process, it is ensured that the generated data is not only realistic in statistical characteristics but also consistent in physical laws, significantly improving the reliability and practicality of the generated data.

[0112] The ocean physical information generation method of this application can generate high-quality data in the case of scarce data, train a physically consistent generation model using limited observation data, and effectively make up for the problem of insufficient data. By combining PINN and GAN, while maintaining physical accuracy, it greatly improves the computational efficiency of data generation. The physical constraints of PINN combined with the generation ability of GAN enable the generation of ocean data that conforms to physical laws without the need for complex numerical simulations, thereby reducing the computational cost.

[0113] The ocean physical information generation method of the present application can not only generate diverse data, but also well adapt to different environmental conditions, improving the generalization ability of the model. In this way, the generated data can better reflect the changes in different ocean environments and have a wider applicability. PINN has a built-in physical correction function during the data generation process to ensure that the generated data conforms to the predetermined physical laws at the generation stage. Combining with the statistically realistic data generated by GAN, this method can automatically perform physical verification and correction during the generation process, reducing the need for additional post-processing and improving the accuracy and reliability of the data. The ocean physical information generation method combining PINN and GAN has significant advantages in improving physical consistency, enhancing the ability to process scarce data, optimizing computational efficiency, improving generalization ability, and accurately verifying the generated data, and can provide higher-quality simulated data support for ocean scientific research and applications.

[0114] In practical applications, an implementation process of the above ocean physical information generation method can be described as follows: Obtain partial ocean physical information; input the partial ocean physical information into the generative adversarial network based on the physics-informed neural network built for training; use the generator (i.e., the ocean physical information generation model) in the trained generative adversarial network to sample and generate high-quality ocean physical information that conforms to physical laws.

[0115] Among them, the training process of the generative adversarial network based on the physics-informed neural network includes: First, divide the obtained partial ocean physical information into a training set, a validation set, and a test set. The loss function uses cross-entropy loss and physical loss to ensure that the generated data conforms to physical laws. Use mini-batch stochastic gradient descent. In each training iteration, first train the discriminator, and then train the generator, and iterate repeatedly until the model converges; Monitor the model performance through cross-validation and early stopping strategies to prevent overfitting, and use random sampling and noise perturbation methods to ensure the diversity of data generation.

[0116] The generation of data is to sample a random noise vector from the standard normal distribution, combine it with the conditional variable, and then input it into the generator; the generator receives the random noise vector and the conditional variable and outputs ocean physical information that meets the set conditions.

[0117] Based on the same inventive concept, the embodiment of the present application also provides an ocean physical information generation device for implementing the above-mentioned ocean physical information generation method. The implementation solutions provided by this device to solve problems are similar to those recorded in the above method. Therefore, the specific limitations in one or more embodiments of the ocean physical information generation device provided below can refer to the limitations on the ocean physical information generation method in the above text, and will not be repeated here.

[0118] In an exemplary embodiment, as Figure 6As shown, an ocean physical information generation device is provided, including: a data acquisition module 601 for acquiring a random noise vector for information generation; a model determination module 602 for determining an ocean physical information generation model; and an information generation module 603 for inputting the random noise vector for information generation into the ocean physical information generation model to obtain the generated ocean physical information.

[0119] Among them, the model determination module 602 includes: a data acquisition unit for acquiring actual ocean physical information and a random noise vector for training; and a model training unit for training a generative adversarial network with the actual ocean physical information and the random noise vector for training, with the goal of minimizing the loss function, to obtain an ocean physical information generation model.

[0120] The generative adversarial network includes a generator and a discriminator.

[0121] The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module, and a first neural network connected in sequence.

[0122] The adaptive spectrum module is used to process the random noise vector for training by using fast Fourier transform; the attention module is used to perform operations on the processed data by using an attention mechanism; and the first neural network is used to perform gradient operations on the data after the operations and output ocean physical information.

[0123] The discriminator is used to discriminate the ocean physical information output by the first neural network according to the actual ocean physical information to obtain a discrimination result.

[0124] The loss function at least includes: a physical loss; the physical loss is determined by performing physical differential calculations on the fluid in the ocean physical information output by the first neural network.

[0125] The trained generator serves as the ocean physical information generation model.

[0126] This application uses the technology combining PINN and GAN to realize the generation of ocean physical information. PINN is used to model the physical laws in the ocean environment, and physical constraints are introduced to ensure the physical consistency of the generated data. On this basis, GAN generates realistic ocean information data through adversarial training. Specifically: A small amount of ocean information data such as coastline cyclone index, revetment ratio, wetland ratio, temperature and humidity field, etc. is obtained, and is input into the GAN based on the physics-informed neural network for training to obtain a trained model, and then the spatio-temporal data of ocean physical information is complemented through sampling. This application can provide high-quality data generation solutions in the fields of ocean data missing, environmental monitoring, climate prediction, etc. Combining the physical constraints of PINN with the generation ability of GAN can better simulate these complex processes, enabling the generated data to capture the complex spatio-temporal dynamic changes, contributing to the realization of an efficient real-time prediction and monitoring model, improving the response speed, and significantly enhancing the reliability and applicability of the data.

[0127] In an exemplary embodiment, a computer device is provided. This computer device can be a server or a terminal, and its internal structure diagram can be as Figure 7 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store random noise vectors for information generation. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it realizes a method for generating ocean physical information.

[0128] Those skilled in the art can understand that Figure 7 the structure shown in

[0129] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program which, when executed by a processor, implements the steps in the above method embodiments.

[0130] In an exemplary embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the steps in the above method embodiments.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0132] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0133] In each of the embodiments provided in this application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on a blockchain, etc., and is not limited thereto. In each of the embodiments provided in this application, the processor may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, etc., and is not limited thereto.

[0134] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0135] Specific examples are used in this article to elaborate on the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A method for generating ocean physical information, characterized in that: The method for generating ocean physical information comprises: Get the random noise vector for information generation; Inputting the random noise vector used for information generation into the ocean physical information generation model to obtain generated ocean physical information; Wherein, the method for determining the ocean physical information generation model includes: Obtain actual ocean physics information and random noise vectors for training; Using actual ocean physics information and random noise vectors for training, the generative adversarial network is trained with the goal of minimizing the loss function to obtain an ocean physics information generation model. The generative adversarial network includes a generator and a discriminator; The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module and a first neural network connected in sequence; the adaptive spectrum module is residually connected to the first neural network; the first neural network includes: a first input layer, a hidden layer and a first output layer; the hidden layer includes 16 layers of hidden neurons, wherein the last layer of hidden neurons is residually connected to the middle layer of hidden neurons; The adaptive spectrum module is used to process the random noise vector used for training by using fast Fourier transform; the attention module is used to operate on the processed data by using the attention mechanism; the first neural network is used to perform gradient operation on the operated data and output ocean physical information; The discriminator is used to discriminate the ocean physical information output by the first neural network according to the actual ocean physical information to obtain a discriminant result; The loss function at least includes: physical loss; the physical loss is determined by performing physical differential calculation on the ocean fluid in the ocean physical information output by the first neural network; The trained generator is used as the ocean physical information generation model; The loss function includes: generator loss and discriminator loss; the generator loss includes: generator loss, cross entropy loss and physical loss; The calculation formula of the generator loss is: ； in, represents the generator loss; Represents the loss of the generator itself; represents the cross entropy loss; Indicates physical loss; represents the i-th ocean physical information generated; m represents the number of ocean physical information generated; represents the identification result of the i-th ocean physical information generated by the discriminator; The calculation formula of the physical loss is: ； in, represents the ocean fluid velocity field in the i-th ocean physical information generated; t represents time; represents the gradient operator; It means to find partial derivative; represents the ocean fluid pressure field in the i-th generated ocean physical information; Represents the kinematic viscosity of ocean fluids in the generated ocean physics information; Represents the ocean fluid density in the generated ocean physics information; represents the external force corresponding to the i-th generated ocean physical information; The discriminator loss is calculated as: ； represents the discriminator loss; Represents the actual i-th ocean physical information; Represents the identification result of the discriminator on the actual i-th ocean physical information.

2. The method for generating ocean physical information according to claim 1, characterized in that: Using actual ocean physics information and random noise vectors for training, the generative adversarial network is trained with the goal of minimizing the loss function to obtain an ocean physics information generation model, which specifically includes: Preprocessing and data encoding the actual ocean physical information to obtain encoded data; For the ath iteration in the training process, the random noise vector used for training in the ath iteration is input into the physical information neural network, and the first neural network in the physical information neural network outputs the ocean physical information of the ath iteration; wherein a>1; Input the encoded data and the ocean physical information of the a-th iteration into the discriminator, and the discriminator outputs the identification result of the a-th iteration; Determine the generator loss of the a-th iteration according to the ocean physical information of the a-th iteration output by the first neural network, determine the cross entropy loss of the a-th iteration according to the ocean physical information of the a-th iteration output by the first neural network and the encoded data, and perform physical differential calculation on the fluid in the ocean physical information of the a-th iteration output by the first neural network to determine the physical loss of the a-th iteration; Determine the generator loss of the a-th iteration according to the generator loss of the a-th iteration, the cross entropy loss of the a-th iteration, and the physical loss of the a-th iteration; Determine the discriminator loss of the a-th iteration according to the identification result of the a-th iteration; If the difference between the generator loss of the a-th iteration and the generator loss of the previous iteration is within the first set difference range, and the difference between the discriminator loss of the a-th iteration and the discriminator loss of the previous iteration is within the second set difference range, then the loss function of the a-th iteration is determined to be the smallest, and the generator after the a-th iteration is used as the ocean physical information generation model; otherwise, the next iteration is performed after updating the number of iterations.

3. The method for generating ocean physical information according to claim 2, characterized in that: Preprocess and encode the actual ocean physical information to obtain the encoded data, including: Process missing values ​​and outliers on actual ocean physical information to obtain preprocessed data; Performing one-hot encoding on discrete variables in the preprocessed data to obtain first encoded data; Normalizing or standardizing the continuous variables in the preprocessed data to obtain processed continuous data; Performing binary conversion encoding on the processed continuous data to obtain second encoded data; The first encoded data and the second encoded data are determined as encoded data.

4. The method for generating ocean physical information according to claim 1, characterized in that: After inputting the random noise vector used for information generation into the ocean physical information generation model to obtain the generated ocean physical information, the ocean physical information generation method further includes: The generated ocean physical information is decoded and verified using actual ocean physical information, and the verified ocean physical information is determined as the finally generated ocean physical information.

5. The method for generating ocean physical information according to claim 4, characterized in that: The generated ocean physical information is decoded and verified using actual ocean physical information, including: Extracting characteristic statistics of actual ocean physical information to obtain first characteristic statistics; Extracting characteristic statistics of the generated ocean physical information to obtain a second characteristic statistic; The fidelity of the first characteristic statistic and the second characteristic statistic is determined, and if the fidelity reaches a set degree value, the generated ocean physical information is determined as verified ocean physical information.

6. A device for generating ocean physical information, characterized in that: The ocean physical information generating device comprises: A data acquisition module, used for acquiring a random noise vector for information generation; A model determination module, used for determining a model for generating ocean physical information; An information generation module, used for inputting a random noise vector used for information generation into an ocean physical information generation model to obtain generated ocean physical information; The model determination module comprises: A data acquisition unit, used to acquire actual ocean physical information and random noise vectors for training; A model training unit, used to train the generative adversarial network with the goal of minimizing the loss function by using actual ocean physical information and a random noise vector used for training, so as to obtain an ocean physical information generation model; The generative adversarial network includes a generator and a discriminator; The generator is a physical information neural network; the physical information neural network includes: an adaptive spectrum module, an attention module and a first neural network connected in sequence; the adaptive spectrum module is residually connected to the first neural network; the first neural network includes: a first input layer, a hidden layer and a first output layer; the hidden layer includes 16 layers of hidden neurons, wherein the last layer of hidden neurons is residually connected to the middle layer of hidden neurons; The adaptive spectrum module is used to process the random noise vector used for training by using fast Fourier transform; the attention module is used to operate on the processed data by using the attention mechanism; the first neural network is used to perform gradient operation on the operated data and output ocean physical information; The discriminator is used to discriminate the ocean physical information output by the first neural network according to the actual ocean physical information to obtain a discriminant result; The loss function at least includes: physical loss; the physical loss is determined by performing physical differential calculation on the fluid in the ocean physical information output by the first neural network; The trained generator is used as the ocean physical information generation model; The loss function includes: generator loss and discriminator loss; the generator loss includes: generator loss, cross entropy loss and physical loss; The calculation formula of the generator loss is: ； in, represents the generator loss; Represents the loss of the generator itself; represents the cross entropy loss; Indicates physical loss; represents the i-th ocean physical information generated; m represents the number of ocean physical information generated; represents the identification result of the i-th ocean physical information generated by the discriminator; The calculation formula of the physical loss is: ； in, represents the ocean fluid velocity field in the i-th ocean physical information generated; t represents time; represents the gradient operator; It means to find partial derivative; represents the ocean fluid pressure field in the i-th generated ocean physical information; Represents the kinematic viscosity of ocean fluids in the generated ocean physics information; Represents the ocean fluid density in the generated ocean physics information; represents the external force corresponding to the i-th generated ocean physical information; The discriminator loss is calculated as: ； represents the discriminator loss; Represents the actual i-th ocean physical information; Represents the identification result of the discriminator on the actual i-th ocean physical information.

7. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for generating ocean physical information according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for generating ocean physical information according to any one of claims 1 to 5 is implemented.

Images