A system and method for calculating and analyzing deformation and stress of gravel cushion layer of road embankment

Abstract

The present invention relates to the field of data analysis technology, and specifically to a system and method for calculating and analyzing the deformation and force of a road embankment gravel cushion. A system for calculating and analyzing the deformation and force of a road embankment gravel cushion comprises: a data acquisition module and a gravel cushion deformation and force analysis module. The present invention processes the gradation characteristics, pavement layer characteristics, embankment fill layer characteristics and vehicle load condition vectors of the gravel cushion into a gravel cushion deformation and force analysis model, and outputs a corresponding deformation and force analysis image set, thereby realizing deformation and force calculation analysis of the gravel cushion. This method no longer requires traditional finite element analysis software and is not limited by the number of particles of traditional finite element analysis software, thereby improving the efficiency of the deformation and force calculation analysis of the gravel cushion. Furthermore, the gradation changes of the graded particle crushing are considered in the deformation and force analysis model of the gravel cushion, thereby improving the accuracy of the deformation and force calculation analysis of the gravel cushion.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a system and method for calculating and analyzing deformation and stress of a road embankment gravel cushion. Background Art

[0002] Gravel cushions for road embankments are crucial in road engineering. However, when performing deformation and stress calculations on gravel cushions, traditional finite element analysis software, especially when simulating graded gravel cushions, needs to model a large number of particles of different sizes in order to accurately reflect the actual particle gradation composition. This leads to a sharp increase in the number of finite element model units and an exponential growth in the calculation scale, exceeding current computing capabilities. This forces traditional methods to simplify the model, such as treating the graded gravel cushion as a homogeneous material or roughly simplifying the particle gradation. This ignores the discreteness of the graded particles and the important influence of the gradation characteristics on the deformation and stress characteristics. As a result, the calculation results deviate significantly from the actual situation, making it difficult to accurately reflect the actual stress and deformation state of the gravel cushion. Summary of the Invention

[0003] The present invention processes the gradation characteristics of the crushed stone cushion layer, the pavement layer characteristics, the embankment fill layer characteristics and the vehicle load condition vector into the crushed stone cushion layer deformation force analysis model, outputs the corresponding deformation force analysis image set, and realizes the deformation force calculation analysis of the crushed stone cushion layer. It no longer needs to use traditional finite element analysis software and is not limited by the number of particles of traditional finite element analysis software, thereby improving the efficiency of the crushed stone cushion layer deformation force calculation analysis; and the gradation change of the graded particle crushing is taken into account in the crushed stone cushion layer deformation force analysis model, thereby improving the accuracy of the crushed stone cushion layer deformation force calculation analysis.

[0004] The present invention provides a method for calculating and analyzing deformation and stress of a road embankment gravel cushion, comprising:

[0005] Obtain the gradation characteristics of the crushed stone cushion layer; obtain the set vehicle load condition vector; obtain the pavement layer characteristics of the pavement layer and the embankment fill layer characteristics of the embankment fill layer;

[0006] Gradation characteristics of the crushed stone cushion, characteristics of the pavement layer, characteristics of the embankment fill layer, and vehicle load condition vectors are combined into crushed stone cushion deformation force analysis data, and the crushed stone cushion deformation force analysis data is sent to a crushed stone cushion deformation force analysis model for processing, and a corresponding deformation force analysis image set is output, which includes a number of stress analysis images and deformation analysis images;

[0007] The deformation and stress analysis model of the gravel cushion layer is established based on the variational autoencoder, including an encoder layer, a sampling layer, a graded particle time series analysis layer, a decoder layer and a deformation and stress analysis image set output layer. The encoder layer is used to encode the deformation and stress analysis data of the gravel cushion layer and output the corresponding mean vector and standard deviation vector; the sampling layer is used to perform sampling operations on the mean vector and standard deviation vector and output the corresponding latent vector; the graded particle time series analysis layer is used to analyze the time series changes of the gradation characteristics under the vehicle load conditions corresponding to the vehicle load condition vector and output the time series processing stress analysis characteristics; the decoder layer is used to perform deconvolution operations on the time series processing stress analysis characteristics and output the corresponding deformation and stress analysis image set; the deformation and stress analysis image set output layer is used to output the deformation and stress analysis image set.

[0008] As a preferred aspect, the crushed stone cushion deformation and force analysis data is sent to the crushed stone cushion deformation and force analysis model for processing, and the corresponding deformation and force analysis image set is output, which specifically includes the following steps:

[0009] The deformation and stress analysis data of the crushed stone cushion layer are sent to the encoder layer for full connection processing, and the corresponding mean vector E and standard deviation vector Q are output;

[0010] In the sampling layer, the latent vector Z is calculated by the following formula: Z = E + Qε, where ε is the random noise sampled from the standard normal distribution;

[0011] In the graded particle time series analysis layer, including N LSTM units, the potential vector Z is processed by N LSTM units in sequence, and the time series processing force analysis features are output;

[0012] The decoder layer includes several parallel channels, which send the time series stress analysis features to each channel for deconvolution operation, and the outputs of all channels are combined into a deformation stress analysis image set;

[0013] The deformation and force analysis image set output layer is used to output the deformation and force analysis image set.

[0014] As a preferred aspect, training the crushed stone cushion deformation stress analysis model includes the following steps:

[0015] Acquire several crushed stone cushion deformation force analysis training samples, which include crushed stone cushion deformation force analysis data, and label the crushed stone cushion deformation force analysis training samples through a deformation force analysis image set sequence, which includes several deformation force analysis image sets that change with time. All labeled crushed stone cushion deformation force analysis training samples are combined into a crushed stone cushion deformation force analysis training set, and a crushed stone cushion deformation force analysis model is trained through the crushed stone cushion deformation force analysis training set. The loss value is calculated, and it is determined whether the loss value is within a preset range. If the loss value is within the preset range, the trained crushed stone cushion deformation force analysis model is output; otherwise, the crushed stone cushion deformation force analysis model is continued to be trained through the crushed stone cushion deformation force analysis training set.

[0016] As a preferred aspect, it also includes simulating and optimizing the hyper parameters in the crushed stone cushion deformation stress analysis model, which specifically includes the following steps:

[0017] Constructing several hyperparameter sets of the crushed stone cushion deformation and stress analysis model, wherein the hyperparameter sets of the crushed stone cushion deformation and stress analysis model include each initialized hyperparameter of the crushed stone cushion deformation and stress analysis model; forming all the hyperparameter sets of the crushed stone cushion deformation and stress analysis model into a population set, and setting a maximum number of iterations;

[0018] Calculate the fitness value corresponding to each hyperparameter set of the crushed stone cushion deformation and stress analysis model;

[0019] Iteratively update the population set through the group optimization algorithm;

[0020] Until the number of iterations reaches the maximum number of iterations, the hyperparameter set of the crushed stone cushion deformation force analysis model corresponding to the maximum fitness value is output to set the hyperparameters in the crushed stone cushion deformation force analysis model, and the simulation optimization of the hyperparameters in the crushed stone cushion deformation force analysis model is completed.

[0021] As an optimal aspect, the fitness value corresponding to the hyperparameter set of the crushed stone cushion deformation force analysis model is calculated. The specific calculation method is as follows: the hyperparameters in the crushed stone cushion deformation force analysis model are set through the hyperparameter set of the crushed stone cushion deformation force analysis model, and then the crushed stone cushion deformation force analysis model is trained through the crushed stone cushion deformation force analysis training set, and the accuracy of the crushed stone cushion deformation force analysis model after training is used as the fitness value corresponding to the hyperparameter set of the crushed stone cushion deformation force analysis model.

[0022] As a preferred aspect, the swarm optimization algorithm adopts the sparrow search algorithm.

[0023] The present invention also provides a road embankment crushed stone cushion deformation force calculation and analysis system, comprising:

[0024] The data acquisition module is used to obtain the gradation characteristics of the crushed stone cushion layer; obtain the set vehicle load condition vector; obtain the pavement layer characteristics of the road surface layer and the embankment fill layer characteristics of the embankment fill layer;

[0025] The crushed stone cushion deformation and stress analysis module is used to combine the gradation characteristics of the crushed stone cushion, the characteristics of the pavement layer, the characteristics of the embankment fill layer, and the vehicle load condition vector to form the crushed stone cushion deformation and stress analysis data, and then send the crushed stone cushion deformation and stress analysis data to the crushed stone cushion deformation and stress analysis model for processing, and output the corresponding deformation and stress analysis image set, which includes a number of stress analysis images and deformation analysis images;

[0026] The deformation and stress analysis model of the gravel cushion layer is established based on the variational autoencoder, including an encoder layer, a sampling layer, a graded particle time series analysis layer, a decoder layer and a deformation and stress analysis image set output layer. The encoder layer is used to encode the deformation and stress analysis data of the gravel cushion layer and output the corresponding mean vector and standard deviation vector; the sampling layer is used to perform sampling operations on the mean vector and standard deviation vector and output the corresponding latent vector; the graded particle time series analysis layer is used to analyze the time series changes of the gradation characteristics under the vehicle load conditions corresponding to the vehicle load condition vector and output the time series processing stress analysis characteristics; the decoder layer is used to perform deconvolution operations on the time series processing stress analysis characteristics and output the corresponding deformation and stress analysis image set; the deformation and stress analysis image set output layer is used to output the deformation and stress analysis image set.

[0027] The present invention has the following advantages:

[0028] The present invention processes the gradation characteristics of the crushed stone cushion layer, the pavement layer characteristics, the embankment fill layer characteristics and the vehicle load condition vector into the crushed stone cushion layer deformation force analysis model, outputs the corresponding deformation force analysis image set, and realizes the deformation force calculation analysis of the crushed stone cushion layer. It no longer needs to use traditional finite element analysis software and is not limited by the number of particles of traditional finite element analysis software, thereby improving the efficiency of the crushed stone cushion layer deformation force calculation analysis; and the gradation change of the graded particle crushing is taken into account in the crushed stone cushion layer deformation force analysis model, thereby improving the accuracy of the crushed stone cushion layer deformation force calculation analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the structure of the deformation and force calculation and analysis system for the road embankment gravel cushion layer used in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0031] Example 1, a method for calculating and analyzing deformation and stress of a road embankment gravel cushion layer, comprising:

[0032] Obtain the gradation characteristics of the crushed stone cushion layer. The gradation characteristics here refer to the characteristic information corresponding to the configuration of the crushed stone cushion layer of the road embankment as a gradation model, generally including the crushed stone cushion layer thickness, particle shape, particle size distribution, fractal dimension, porosity and generation method, etc. These gradation characteristics can express the corresponding gradation model; obtain the set vehicle load condition vector. The vehicle load condition vector here represents the load condition of the vehicle when performing deformation force analysis, generally including wheel load pressure and action frequency, etc.; and considering that the road embankment does not only include the crushed stone cushion layer, generally also includes the pavement layer located above the crushed stone cushion layer and the embankment fill layer located below the crushed stone cushion layer, it is also necessary to obtain the pavement layer characteristics of the pavement layer and the embankment fill layer characteristics of the embankment fill layer. The pavement layer characteristics generally include the pavement layer thickness, pavement layer modulus and generation method, etc. The embankment fill layer characteristics generally include the embankment fill layer thickness, embankment fill layer modulus and generation method. Here, the pavement layer modulus and embankment fill layer modulus refer to the corresponding stiffness data and shear strength data, etc.

[0033] Grading characteristics of the crushed stone cushion, characteristics of the pavement layer, characteristics of the embankment fill layer, and vehicle load condition vectors are used to form deformation and stress analysis data of the crushed stone cushion. The deformation and stress analysis data of the crushed stone cushion are then fed into a deformation and stress analysis model for processing, and a corresponding deformation and stress analysis image set is output. The deformation and stress analysis image set includes several stress analysis images and deformation analysis images. The stress analysis images are generally vertical stress images, horizontal stress images, and shear stress images, and the deformation analysis images are generally vertical displacement images and horizontal displacement images. The deformation and stress analysis image set can directly reflect the deformation and stress conditions of the crushed stone cushion, thereby providing assistance for deformation and stress calculation and analysis of the crushed stone cushion of the road embankment, without being limited by the number of particles in traditional finite element analysis software.

[0034] The crushed stone cushion deformation stress analysis model is established based on the variational autoencoder, including an encoder layer, a sampling layer, a gradation particle time series analysis layer, a decoder layer and a deformation stress analysis image set output layer, wherein the encoder layer is used to encode the crushed stone cushion deformation stress analysis data, and output the corresponding mean vector and standard deviation vector. The crushed stone cushion deformation stress analysis data can be mapped into the latent space through the mean vector and the standard deviation vector, and the Gaussian distribution of the crushed stone cushion deformation stress analysis data mapped into the latent space can be represented by the corresponding mean vector and the standard deviation vector; the sampling layer is used to sample the mean vector and the standard deviation vector, and output the corresponding latent vector. The latent vector here can reflect a sample that conforms to the Gaussian distribution in the latent space, which is used for subsequent deformation stress analysis image set construction; the gradation particle time series analysis layer is used to analyze the gradation characteristics in the vehicle load condition vector The time series changes that occur under the corresponding vehicle load conditions are output, and the time series processing force analysis characteristics are output. Under vehicle load conditions, the gradation model corresponding to the gravel cushion layer may undergo gradation changes due to the crushing of the gradation particles, which in turn causes changes in the deformation force of the gravel cushion layer. Therefore, the gradation changes caused by the crushing of the gradation particles are fitted through the gradation particle time series analysis layer, thereby improving the accuracy of the deformation force analysis of the gravel cushion layer; the decoder layer is used to perform deconvolution operations on the time series processing force analysis characteristics and output the corresponding deformation force analysis image set. It should be noted that there are several parallel channels in the decoder layer, and each channel performs deconvolution operations, but the deconvolution parameters corresponding to each channel change with the task corresponding to the channel. Each channel here corresponds to the generation of a stress analysis image or a deformation analysis image; the deformation force analysis image set output layer is used to output the deformation force analysis image set;

[0035] The present application processes the gradation characteristics of the crushed stone cushion layer, the characteristics of the pavement layer, the characteristics of the embankment fill layer and the vehicle load condition vector into the deformation force analysis model of the crushed stone cushion layer, and outputs the corresponding deformation force analysis image set, thereby realizing deformation force calculation and analysis of the crushed stone cushion layer. The application no longer requires traditional finite element analysis software and is not limited by the number of particles of traditional finite element analysis software, thereby improving the efficiency of deformation force calculation and analysis of the crushed stone cushion layer. In addition, the gradation changes of graded particle crushing are taken into account in the deformation force analysis model of the crushed stone cushion layer, thereby improving the accuracy of deformation force calculation and analysis of the crushed stone cushion layer.

[0036] The crushed stone cushion deformation and stress analysis data is sent to the crushed stone cushion deformation and stress analysis model for processing, and the corresponding deformation and stress analysis image set is output. Specifically, the steps include:

[0037] The crushed stone cushion deformation and stress analysis data is fed into the encoder layer for full-connection processing, and the corresponding mean vector E and standard deviation vector Q are output. It should be noted that the mean vector E and the standard deviation vector Q have consistent dimensions and are consistent with the dimension of the latent space. The dimension of the latent space is set by the developer and is generally 10;

[0038] In the sampling layer, the latent vector Z is calculated by the following formula: Z = E + Qε, where ε is the random noise sampled from the standard normal distribution;

[0039] The graded particle time series analysis layer includes N LSTM units, which process the potential vector Z sequentially through the N LSTM units and output the time series processing force analysis features. It should be added that if the purpose is to observe the time series changes of the deformation force analysis image set, the output of each LSTM unit can also be regarded as a time series processing force analysis feature and perform subsequent processing;

[0040] The decoder layer includes several parallel channels, which send the time series stress analysis features to each channel for deconvolution operation, and the outputs of all channels are combined into a deformation stress analysis image set;

[0041] The deformation and force analysis image set output layer is used to output the deformation and force analysis image set.

[0042] Training the crushed stone cushion deformation stress analysis model includes the following steps:

[0043] Obtain several crushed stone cushion deformation force analysis training samples, the crushed stone cushion deformation force analysis training samples include crushed stone cushion deformation force analysis data, the crushed stone cushion deformation force analysis data here is obtained through actual crushed stone cushion deformation force calculation analysis performed by finite element analysis software, the finite element analysis software here can be PFC2D or FLAC software, and because it is the construction of training samples, the number of graded particles can be increased without considering the calculation efficiency, and the crushed stone cushion deformation force analysis training samples are annotated through the deformation force analysis image set. It should be noted that the deformation force analysis image set here is output by the finite element analysis software, and the dynamic crushing algorithm of the graded particles can be embedded in the finite element analysis software, and the crushing is triggered based on the contact force between the particles and the energy threshold, and the graded fractal dimension after crushing is updated. number, so that the deformation stress analysis image set is expanded into a deformation stress analysis image set sequence. The deformation stress analysis image set sequence includes several deformation stress analysis image sets that change with time. All labeled gravel cushion deformation stress analysis training samples are composed of a gravel cushion deformation stress analysis training set. The gravel cushion deformation stress analysis model is trained through the gravel cushion deformation stress analysis training set. During the training, the output of each LSTM unit is regarded as a time series processing stress analysis feature, and subsequent processing is performed to calculate the loss value and determine whether the loss value is within the preset range. The preset range is determined by the developer. If the loss value is within the preset range, the trained gravel cushion deformation stress analysis model is output; otherwise, the gravel cushion deformation stress analysis model is continued to be trained through the gravel cushion deformation stress analysis training set.

[0044] In order to further improve the accuracy of the crushed stone cushion deformation and stress analysis model, the hyperparameters in the crushed stone cushion deformation and stress analysis model are simulated and optimized, including the following steps:

[0045] Construct several hyperparameter sets for the crushed stone cushion deformation stress analysis model. The hyperparameter sets include the hyperparameters of each initialized crushed stone cushion deformation stress analysis model, such as the number of fully connected layers of the encoder, the number N of LSTM units, and the learning rate. All hyperparameter sets for the crushed stone cushion deformation stress analysis model are grouped into a population set, and the maximum number of iterations is set.

[0046] Calculate the fitness value corresponding to each hyperparameter set of the crushed stone cushion deformation force analysis model. The specific calculation method is as follows: set the hyperparameters in the crushed stone cushion deformation force analysis model through the hyperparameter set of the crushed stone cushion deformation force analysis model, and then train the crushed stone cushion deformation force analysis model through the crushed stone cushion deformation force analysis training set. The accuracy of the crushed stone cushion deformation force analysis model after training is used as the fitness value corresponding to the hyperparameter set of the crushed stone cushion deformation force analysis model.

[0047] Iteratively update the population set through the sparrow search algorithm;

[0048] Until the number of iterations reaches the maximum number of iterations, the hyperparameter set of the crushed stone cushion deformation force analysis model corresponding to the maximum fitness value is output to set the hyperparameters in the crushed stone cushion deformation force analysis model, and the simulation optimization of the hyperparameters in the crushed stone cushion deformation force analysis model is completed.

[0049] Example 2, a road embankment crushed stone cushion deformation force calculation and analysis system, see Figure 1 ,include:

[0050] The data acquisition module is used to obtain the gradation characteristics of the crushed stone cushion layer. The gradation characteristics here refer to the characteristic information corresponding to the configuration of the crushed stone cushion layer of the road embankment as a gradation model, generally including the crushed stone cushion layer thickness, particle shape, particle size distribution, fractal dimension, porosity and generation method, etc. These gradation characteristics can express the corresponding gradation model; obtain the set vehicle load condition vector. The vehicle load condition vector here represents the load condition of the vehicle when performing deformation force analysis, generally including wheel load pressure and action frequency, etc.; and considering that the road embankment does not only include the crushed stone cushion layer, generally also includes the pavement layer located above the crushed stone cushion layer and the embankment fill layer located below the crushed stone cushion layer, it is also necessary to obtain the pavement layer characteristics of the pavement layer and the embankment fill layer characteristics of the embankment fill layer. The pavement layer characteristics generally include the pavement layer thickness, pavement layer modulus and generation method, etc. The embankment fill layer characteristics generally include the embankment fill layer thickness, embankment fill layer modulus and generation method, where the pavement layer modulus and embankment fill layer modulus refer to the corresponding stiffness data and shear strength data, etc.

[0051] The crushed stone cushion deformation and stress analysis module is used to combine the gradation characteristics of the crushed stone cushion, the characteristics of the pavement layer, the characteristics of the embankment fill layer and the vehicle load condition vector into the crushed stone cushion deformation and stress analysis data, and send the crushed stone cushion deformation and stress analysis data to the crushed stone cushion deformation and stress analysis model for processing, and output the corresponding deformation and stress analysis image set. The deformation and stress analysis image set includes several stress analysis images and deformation analysis images. The stress analysis images are generally vertical stress images, horizontal stress images and shear stress images, etc. The deformation analysis images are generally vertical displacement images and horizontal displacement images, etc. The deformation and stress analysis image set can directly reflect the deformation and stress conditions of the crushed stone cushion, thereby providing assistance for the deformation and stress calculation and analysis of the road embankment crushed stone cushion, without being limited by the number of particles in traditional finite element analysis software;

[0052] The crushed stone cushion deformation stress analysis model is established based on the variational autoencoder, including an encoder layer, a sampling layer, a gradation particle time series analysis layer, a decoder layer and a deformation stress analysis image set output layer, wherein the encoder layer is used to encode the crushed stone cushion deformation stress analysis data, and output the corresponding mean vector and standard deviation vector. The crushed stone cushion deformation stress analysis data can be mapped into the latent space through the mean vector and the standard deviation vector, and the Gaussian distribution of the crushed stone cushion deformation stress analysis data mapped into the latent space can be represented by the corresponding mean vector and the standard deviation vector; the sampling layer is used to sample the mean vector and the standard deviation vector, and output the corresponding latent vector. The latent vector here can reflect a sample that conforms to the Gaussian distribution in the latent space, which is used for subsequent deformation stress analysis image set construction; the gradation particle time series analysis layer is used to analyze the gradation characteristics in the vehicle load condition vector The time series changes that occur under the corresponding vehicle load conditions are output, and the time series processing force analysis characteristics are output. Under vehicle load conditions, the gradation model corresponding to the gravel cushion layer may undergo gradation changes due to the crushing of the gradation particles, which in turn causes changes in the deformation force of the gravel cushion layer. Therefore, the gradation changes caused by the crushing of the gradation particles are fitted through the gradation particle time series analysis layer, thereby improving the accuracy of the deformation force analysis of the gravel cushion layer; the decoder layer is used to perform deconvolution operations on the time series processing force analysis characteristics and output the corresponding deformation force analysis image set. It should be noted here that there are several parallel channels in the decoder layer, and each channel performs deconvolution operations, but the deconvolution parameters corresponding to each channel change with the task corresponding to the channel. Each channel here corresponds to the generation of a stress analysis image or a deformation analysis image; the deformation force analysis image set output layer is used to output the deformation force analysis image set.

[0053] It should be understood that those skilled in the art may make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims. Any portion of this specification not described in detail herein is prior art known to those skilled in the art.

Claims

1. A method for calculating and analyzing deformation and stress of a road embankment gravel cushion, characterized in that: include: Gradation characteristics of the crushed stone cushion, characteristics of the pavement layer, characteristics of the embankment fill layer, and vehicle load condition vectors are combined into crushed stone cushion deformation force analysis data, and the crushed stone cushion deformation force analysis data is sent to a crushed stone cushion deformation force analysis model for processing, and a corresponding deformation force analysis image set is output, which includes a number of stress analysis images and deformation analysis images; The crushed stone cushion deformation and stress analysis model is established based on a variational autoencoder, and includes an encoder layer, a sampling layer, a gradation particle time series analysis layer, a decoder layer, and a deformation and stress analysis image set output layer. The encoder layer is used to encode the crushed stone cushion deformation and stress analysis data and output the corresponding mean vector and standard deviation vector; the sampling layer is used to sample the mean vector and standard deviation vector and output the corresponding latent vector; the gradation particle time series analysis layer is used to analyze the time series changes of the gradation characteristics under the vehicle load conditions corresponding to the vehicle load condition vector and output the time series processing stress analysis characteristics; the decoder layer is used to perform a deconvolution operation on the time series processing stress analysis characteristics and output the corresponding deformation and stress analysis image set; The deformation and force analysis image set output layer is used to output the deformation and force analysis image set.

2. A method for calculating and analyzing deformation and stress of a road embankment gravel cushion according to claim 1, characterized in that: The crushed stone cushion deformation and stress analysis data is sent to the crushed stone cushion deformation and stress analysis model for processing, and the corresponding deformation and stress analysis image set is output. Specifically, the steps include: The deformation and stress analysis data of the crushed stone cushion layer are sent to the encoder layer for full connection processing, and the corresponding mean vector E and standard deviation vector Q are output; In the sampling layer, the latent vector Z is calculated by the following formula: Z = E + Qε, where ε is the random noise sampled from the standard normal distribution; In the graded particle time series analysis layer, N LSTM units are included. The potential vector Z is processed by N LSTM units in sequence, and the time series processing force analysis features are output; The decoder layer includes several parallel channels, which send the time series stress analysis features to each channel for deconvolution operation, and the outputs of all channels are combined into a deformation stress analysis image set; The deformation and force analysis image set output layer is used to output the deformation and force analysis image set.

3. A method for calculating and analyzing deformation and stress of a road embankment crushed stone cushion according to claim 2, characterized in that: Training the crushed stone cushion deformation stress analysis model includes the following steps: Acquire several crushed stone cushion deformation force analysis training samples, which include crushed stone cushion deformation force analysis data, and label the crushed stone cushion deformation force analysis training samples through a deformation force analysis image set sequence, which includes several deformation force analysis image sets that change with time. All labeled crushed stone cushion deformation force analysis training samples are combined into a crushed stone cushion deformation force analysis training set, and a crushed stone cushion deformation force analysis model is trained through the crushed stone cushion deformation force analysis training set. The loss value is calculated, and it is determined whether the loss value is within a preset range. If the loss value is within the preset range, the trained crushed stone cushion deformation force analysis model is output; otherwise, the crushed stone cushion deformation force analysis model is continued to be trained through the crushed stone cushion deformation force analysis training set.

4. A method for calculating and analyzing deformation and stress of a road embankment crushed stone cushion according to claim 3, characterized in that: It also includes simulation optimization of hyperparameters in the crushed stone cushion deformation stress analysis model, which specifically includes the following steps: Constructing several hyperparameter sets of the crushed stone cushion deformation and stress analysis model, wherein the hyperparameter sets of the crushed stone cushion deformation and stress analysis model include each initialized hyperparameter of the crushed stone cushion deformation and stress analysis model; forming all the hyperparameter sets of the crushed stone cushion deformation and stress analysis model into a population set, and setting a maximum number of iterations; Calculate the fitness value corresponding to each hyperparameter set of the crushed stone cushion deformation and stress analysis model; Iteratively update the population set through the group optimization algorithm; Until the number of iterations reaches the maximum number of iterations, the hyperparameter set of the crushed stone cushion deformation force analysis model corresponding to the maximum fitness value is output to set the hyperparameters in the crushed stone cushion deformation force analysis model, and the simulation optimization of the hyperparameters in the crushed stone cushion deformation force analysis model is completed.

5. A method for calculating and analyzing deformation and stress of a road embankment crushed stone cushion according to claim 4, characterized in that: Calculate the fitness value corresponding to the hyperparameter set of the crushed stone cushion deformation force analysis model. The specific calculation method is as follows: set the hyperparameters in the crushed stone cushion deformation force analysis model through the hyperparameter set of the crushed stone cushion deformation force analysis model, and then train the crushed stone cushion deformation force analysis model through the crushed stone cushion deformation force analysis training set. The accuracy of the crushed stone cushion deformation force analysis model after training is used as the fitness value corresponding to the hyperparameter set of the crushed stone cushion deformation force analysis model.

6. A method for calculating and analyzing deformation and stress of a road embankment crushed stone cushion according to claim 5, characterized in that: The swarm optimization algorithm uses the sparrow search algorithm.

7. A road embankment crushed stone cushion deformation force calculation and analysis system, characterized in that: The system applies the method for calculating and analyzing deformation and stress of a road embankment gravel cushion layer according to any one of claims 1 to 6, comprising: The data acquisition module is used to obtain the gradation characteristics of the crushed stone cushion layer; obtain the set vehicle load condition vector; obtain the pavement layer characteristics of the road surface layer and the embankment fill layer characteristics of the embankment fill layer; The crushed stone cushion deformation and stress analysis module is used to combine the gradation characteristics of the crushed stone cushion, the characteristics of the pavement layer, the characteristics of the embankment fill layer, and the vehicle load condition vector to form the crushed stone cushion deformation and stress analysis data, and then send the crushed stone cushion deformation and stress analysis data to the crushed stone cushion deformation and stress analysis model for processing, and output the corresponding deformation and stress analysis image set, which includes a number of stress analysis images and deformation analysis images; The deformation and stress analysis model of the gravel cushion layer is established based on the variational autoencoder, including an encoder layer, a sampling layer, a graded particle time series analysis layer, a decoder layer and a deformation and stress analysis image set output layer. The encoder layer is used to encode the deformation and stress analysis data of the gravel cushion layer and output the corresponding mean vector and standard deviation vector; the sampling layer is used to sample the mean vector and standard deviation vector and output the corresponding latent vector; the graded particle time series analysis layer is used to analyze the time series changes of the gradation characteristics under the vehicle load conditions corresponding to the vehicle load condition vector and output the time series processing stress analysis characteristics; the decoder layer is used to perform deconvolution operations on the time series processing stress analysis characteristics and output the corresponding deformation and stress analysis image set; the deformation and stress analysis image set output layer is used to output the deformation and stress analysis image set.

Images