Soil layer quantitative layering method, system and equipment based on machine learning algorithm fusion prediction model, and medium

Abstract

The invention provides a soil layer quantitative layering method, system and equipment based on a machine learning algorithm fusion prediction model, and a medium. The method comprises the following steps: S1, collecting and sorting pore pressure static sounding data and soil layer classification information; s2, establishing a basic model by adopting SVM, RF and XGBoost; s3, optimizing hyper-parameters of the basic model by adopting a Bayesian optimization algorithm; s4, establishing a fusion prediction model of the three basic models; s5, training the fusion prediction model; s6, using the trained fusion prediction model to predict the soil layer type of another site; and S7, determining division precision, carrying out layer merging processing on a layering result, and finally obtaining a soil layer quantitative layering result. According to the method, the quantitative relation between the pore pressure static sounding data and the soil layer category can be rapidly established, rapid and accurate classification is realized, the uncertainty of soil layer classification can be quantified, and meanwhile, the subjectivity of a traditional soil layer classification method and the limitation of a single algorithm are avoided.

Description

Technical Field

[0001] The present invention relates to the field of geotechnical engineering, and in particular to a soil layer quantitative stratification method, system, equipment and medium based on a machine learning algorithm fusion prediction model. Background Art

[0002] Compared with other in-situ testing methods, the static penetration test is generally considered to be fast, continuous and economical, and has become a representative method for determining soil stratification information in geotechnical engineering. At present, surveyors mainly classify soil layers based on experience and related formulas in the processing of static penetration data, and lack research on the interpretation theory of static penetration data, resulting in low efficiency and accuracy of classification, which affects the actual engineering survey work and subsequent design calculations.

[0003] Chinese patent CN103898890A discloses a soil stratification method based on BP neural network for double bridge static penetration data. After continuous preprocessing of double bridge static penetration data, a soil stratification BP neural network prediction model is established. This method interprets static penetration data based on BP neural network, and has a certain guiding role in the application of static penetration data in soil stratification. However, its data preprocessing method is too simple, and the prediction accuracy is almost not improved compared with the original data set. In addition, the BP neural network structure used is simple, and the data mining depth is shallow. In the output of the prediction results, the soil type labels output by this method are too absolute, the information display is insufficient, and it cannot provide sufficient guidance to engineering designers.

[0004] A domestic paper with DOI 10.13349 / j.cnki.jdxbn.20170527.001 proposed a soil interface identification method for pore pressure static penetration based on probabilistic neural network, which is effective in identifying soil interlayers. However, there are still problems such as the input parameters cannot fully reflect the characteristics of static penetration data and the model output results are not intuitive.

[0005] Therefore, there is an urgent need for a reliable, high-precision, and output-friendly soil quantitative stratification method to provide an effective basis for various calculations in geotechnical engineering investigation and design. Summary of the invention

[0006] The first object of the present invention is to use the existing pore pressure static penetration data and the corresponding soil layer classification information to provide a soil layer quantitative stratification method based on a machine learning algorithm fusion prediction model.

[0007] To this end, the above-mentioned purpose of the present invention is achieved through the following technical solutions:

[0008] A soil layer quantitative stratification method based on a machine learning algorithm fusion prediction model includes the following steps:

[0009] S1. Collection and collation of pore pressure static penetration data and soil layer classification information;

[0010] Collect and organize geotechnical engineering investigation reports, geotechnical test results, and organize penetration data along the depth of the piezocereography holes and soil layer classification information;

[0011] S2, use machine learning algorithms SVM, RF, and XGBoost to build a basic model;

[0012] The machine learning algorithms SVM, RF and XGBoost are used to establish basic prediction models with pore pressure static penetration test data as input and soil layer types as output. The output results are the probability of soil layers corresponding to each soil type.

[0013] S3, using Bayesian optimization algorithm to optimize the basic model hyperparameters;

[0014] The Bayesian optimization algorithm is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models used for quantitative stratification of soil layers by pore pressure penetration testing.

[0015] S4, establish a fusion prediction model of three basic models in a stacking manner;

[0016] A linear meta-model is established to combine the output results of the three basic models in a stacking manner to form a machine learning algorithm fusion prediction model;

[0017] S5. Training the machine learning algorithm fusion prediction model;

[0018] The collated pore pressure static penetration test data and soil layer classification information are input into the machine learning algorithm fusion prediction model to train the model;

[0019] S6. Use the trained machine learning algorithm to fuse the prediction model to predict the soil type of another site;

[0020] Use the trained machine learning algorithm to integrate the prediction model to predict the soil type of another site, output the probability of each soil type corresponding to the soil layer, and take the soil type corresponding to the maximum probability value as the predicted soil type of the soil layer;

[0021] S7, determining the division accuracy, merging the stratification results, and finally obtaining the quantitative stratification results of the soil layers;

[0022] The accuracy is divided according to the engineering needs, the prediction results are merged, and finally the quantitative stratification results of the soil layers are obtained.

[0023] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0024] As a preferred technical solution of the present invention: in step S1, the pore pressure static penetration test data includes: cone tip resistance q c , side wall friction f s , pore water pressure u2.

[0025] As a preferred technical solution of the present invention: in step S1, the soil layer classification information includes: silt and silty soil, silty clay, silt soil, and silty sand, and the corresponding soil type label values ​​are assigned 0, 1, 2, and 3 respectively.

[0026] As a preferred technical solution of the present invention: in step S3, the three basic models of SVM, RF, and XGBoost respectively use a set of optimal hyperparameters obtained by Bayesian optimization, as shown below:

[0027]

[0028] As a preferred technical solution of the present invention: in step S4, the specific process of establishing the linear metamodel is: calculating the weighted average of the probabilities of the soil layers corresponding to the soil types in the results of each basic model, and the weights are determined by the 10-fold cross-validation results of each basic model on the basic model training set.

[0029] As a preferred technical solution of the present invention: in step S5, in the data set used to train the model, 80% is randomly divided for training the basic model, and the remaining 20% ​​is used to test the meta-model. The meta-model prediction effect is required to be better than all the basic models before it can be used to predict the soil type of another site.

[0030] As a preferred technical solution of the present invention: in step S7, the specific process of merging layers is: merging the same soil layers whose total thickness is less than the division accuracy into the previous layer.

[0031] The second object of the present invention is to provide a soil quantitative stratification system.

[0032] To this end, the above-mentioned purpose of the present invention is achieved through the following technical solutions:

[0033] A soil layer quantitative stratification system includes the following modules:

[0034] A data collection module, which is used to collect and organize geotechnical engineering investigation reports, geotechnical test results, and to organize penetration data along the depth of the piezocone penetration holes and soil layer classification information;

[0035] A basic model building module, wherein the basic model building module is used to use machine learning algorithms SVM, RF, and XGBoost to respectively establish basic prediction models with the pore pressure static penetration data obtained by the data collection module as input and the soil layer category obtained by the data collection module as output, and the output result is the probability of the soil layer corresponding to each soil category;

[0036] A hyperparameter optimization module, which is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models obtained by the basic model building module using a Bayesian optimization algorithm;

[0037] A fusion prediction model building module, wherein the fusion prediction model is used to build a linear meta-model, and the output results of the three basic models obtained by the basic model building module are combined in a stacking manner to form a machine learning algorithm fusion prediction model;

[0038] A fusion prediction model training module, wherein the fusion prediction model training module is used to input the pore pressure static penetration data and soil layer classification information obtained by the data collection module into the machine learning algorithm fusion prediction model obtained by the fusion prediction model establishment module to train the model;

[0039] A prediction module, wherein the prediction module is used to predict the soil layer type of another site by using the machine learning algorithm fusion prediction model trained by the fusion prediction model training module, output the probability of the soil layer corresponding to each soil type, and take the soil type corresponding to the maximum probability value as the predicted soil type of the soil layer;

[0040] The stratification module is used to determine the division accuracy, and to perform layer-by-layer processing on the stratification results obtained by the prediction module to finally obtain the quantitative stratification results of the soil layers.

[0041] The third object of the present invention is to provide an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus, and wherein:

[0042] A memory, the memory being used to store a computer program;

[0043] A processor is used to execute a computer program stored in a memory to implement the steps of the soil layer quantitative stratification method based on a machine learning algorithm fused with a prediction model as described above.

[0044] Another object of the present invention is to provide a non-transitory readable storage medium, in which a program is stored. When the program is executed by a processor, the steps of the soil layer quantitative stratification method based on the machine learning algorithm fusion prediction model as described above are implemented.

[0045] The present invention provides a soil layer quantitative stratification method, system, device and medium based on a machine learning algorithm fusion prediction model, which has the following beneficial effects:

[0046] 1) The soil layer quantitative stratification method based on the machine learning algorithm fusion prediction model provided by the present invention can take advantage of multiple machine learning algorithms to quickly establish a quantitative relationship between pore pressure static penetration data and soil layer categories, achieve rapid and accurate classification, and quantify the uncertainty of soil layer classification. At the same time, it avoids the subjectivity of traditional soil layer classification methods and the limitations of a single algorithm, and improves the accuracy and reliability of classification. The present invention also avoids the subjectivity and empiricism of traditional soil layer classification methods, and provides a scientific, objective and intelligent solution to the soil layer quantitative stratification problem. The present invention can reduce the drilling work in engineering exploration to a certain extent, save exploration costs, and improve exploration efficiency.

[0047] 2) The soil layer quantitative stratification method based on the machine learning algorithm fusion prediction model provided by the present invention is not limited to soil layer classification, but is also applicable to the prediction and analysis of other geotechnical parameters (such as permeability, undrained shear strength, etc.) with spatial variability and correlation characteristics, and has strong versatility. The present invention can also flexibly select and adjust machine learning algorithms and fusion strategies according to different geotechnical engineering problems and data characteristics, and has strong adaptability and expansibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The present invention provides a flow chart of a soil layer quantitative stratification method based on a machine learning algorithm fused with a prediction model.

[0049] Figure 2 This is a flow chart of the Bayesian optimization algorithm used to optimize the hyperparameters of the basic model of the machine learning algorithm.

[0050] Figure 3 A simplified schematic diagram of the structure of a prediction model fused with a machine learning algorithm.

[0051] Figure 4 This is the pore pressure static penetration curve and soil layer column diagram of JT33 in Lianyungang, Jiangsu (0.5m division accuracy). DETAILED DESCRIPTION

[0052] The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

[0053] A soil layer quantitative stratification method based on a machine learning algorithm fusion prediction model includes the following steps:

[0054] S1. Collection and collation of pore pressure static penetration data and soil layer classification information;

[0055] The geotechnical engineering investigation reports and geotechnical test results of Rudong area in Jiangsu Province were collected and sorted out, and the penetration data of the borehole pressure static penetration test and the soil layer classification information along the depth were statistically sorted out, totaling 800 groups, including 200 groups of silt and silty soil, silty clay, silt, and silt sand, corresponding to soil class label values ​​0, 1, 2, and 3 respectively. The training sample examples are shown in Table 1 (only 80 groups of 800 groups of data are shown);

[0056] Table 1 Examples of training samples of XGBoost model for quantitative stratification of soil layers by pore pressure penetration testing

[0057]

[0058]

[0059] S2, use machine learning algorithms SVM, RF, and XGBoost to build a basic model;

[0060] Using Python language, machine learning algorithms SVM, RF, and XGBoost were used to establish the pore pressure static penetration test data (cone tip resistance q c , side wall friction f s , pore water pressure u2) as input and soil type label value as output. The predict_proba() method is used for SVM and RF model prediction. The objective function of XGBoost model takes multi:softprob. Finally, the output results of the three basic prediction models are the probability of soil layers corresponding to each soil type.

[0061] S3, using Bayesian optimization algorithm to optimize the basic model hyperparameters;

[0062] Using Python language, the Bayesian optimization algorithm is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models used for quantitative stratification of soil layers by pore pressure static penetration testing. The cross-validation evaluation index in the Bayesian optimization parameter adjustment function is accuracy, the number of folds is 10, and the number of Bayesian optimization iterations is 500. The three basic models of SVM, RF, and XGBoost use Bayesian optimization to obtain a set of optimal hyperparameters, as shown in Table 2;

[0063] Table 2

[0064]

[0065] S4, establish a fusion prediction model of three basic models in a stacking manner;

[0066] A linear meta-model is established, and the output results of the three basic models are integrated in a stacking manner to form a machine learning algorithm fusion prediction model. In the data set used to train the model (as shown in Table 1), 80% is randomly divided for training the basic model, and the remaining 20% ​​is used to test the meta-model. The random state of train_test_split is set to 0. The specific operation of the linear meta-model is to calculate the weighted average of the probability of each soil type corresponding to the soil layer in the results of each basic model. The weight is determined by the 10-fold cross-validation results of each basic model on the basic model training set. The evaluation indicator is accuracy, and the 10-fold cross-validation results and model weights are shown in Table 3. An example of the meta-model implementation is shown in Table 4;

[0067] Table 3

[0068]

[0069] Table 4

[0070]

[0071] S5. Train and test the machine learning algorithm fusion prediction model;

[0072] The collated pore pressure static penetration data and soil layer classification information are input into the machine learning algorithm fusion prediction model, and the model is trained and tested. The basic model training set and meta-model test set are shown in Table 1, the basic model hyperparameters are shown in Table 2, the weights of each model are shown in Table 3, and the meta-model implementation example is shown in Table 4. For the meta-model test set, the meta-model prediction effect is required to be better than all basic models before it can be used to predict the soil layer type of another site. The prediction accuracy of each basic model and fusion model on the meta-model test set is shown in Table 5;

[0073] Table 5

[0074]

[0075] S6. Use the trained machine learning algorithm to fuse the prediction model to predict the soil type of another site;

[0076] Based on the pore pressure static penetration data of the static penetration hole No. JT33 of the exploration project of the key technology research project on the reinforcement of deep soft soil foundation of heavy-loaded ore storage yards in Lianyungang, Jiangsu Province, the soil layer type is predicted using the trained prediction model. The output result of the prediction model is the probability value of each soil type corresponding to the soil layer. The soil type corresponding to the maximum probability value is taken as the predicted soil type of the soil layer. The total depth of the penetration hole test is 35.0m. The pore pressure static penetration data and prediction results are shown in Table 6.

[0077] Table 6 JT33 pore pressure static exploration data and prediction results of Lianyungang ore storage yard

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] S6, determining the division accuracy, merging the stratification results, and finally obtaining the quantitative stratification results of the soil layers;

[0086] According to the engineering needs, the division accuracy is determined to be 0.5m, and the prediction results are processed by merging layers. Finally, the quantitative stratification results of the soil layer in the JT33 static exploration hole of the Lianyungang ore yard are obtained, as shown in Table 7, as well as the JT33 pore pressure static penetration curve and soil layer column chart (0.5m division accuracy). Figure 3 shown.

[0087] Table 7 Quantitative stratification results of soil layers in JT33 static exploration hole of Lianyungang ore storage yard (0.5m division accuracy)

[0088]

[0089] The present invention also provides a soil layer quantitative stratification system, comprising the following modules:

[0090] A data collection module, which is used to collect and organize geotechnical engineering investigation reports, geotechnical test results, and to organize penetration data along the depth of the piezocone penetration holes and soil layer classification information;

[0091] A basic model building module, wherein the basic model building module is used to use machine learning algorithms SVM, RF, and XGBoost to respectively establish basic prediction models with the pore pressure static penetration data obtained by the data collection module as input and the soil layer category obtained by the data collection module as output, and the output result is the probability of the soil layer corresponding to each soil category;

[0092] A hyperparameter optimization module, which is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models obtained by the basic model building module using a Bayesian optimization algorithm;

[0093] A fusion prediction model building module, wherein the fusion prediction model is used to build a linear meta-model, and the output results of the three basic models obtained by the basic model building module are combined in a stacking manner to form a machine learning algorithm fusion prediction model;

[0094] A fusion prediction model training module, wherein the fusion prediction model training module is used to input the pore pressure static penetration data and soil layer classification information obtained by the data collection module into the machine learning algorithm fusion prediction model obtained by the fusion prediction model establishment module to train the model;

[0095] A prediction module, wherein the prediction module is used to predict the soil layer type of another site by using the machine learning algorithm fusion prediction model trained by the fusion prediction model training module, output the probability of the soil layer corresponding to each soil type, and take the soil type corresponding to the maximum probability value as the predicted soil type of the soil layer;

[0096] The stratification module is used to determine the division accuracy, and to perform layer-by-layer processing on the stratification results obtained by the prediction module to finally obtain the quantitative stratification results of the soil layers.

[0097] The present invention also provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus, and wherein:

[0098] A memory, the memory being used to store a computer program;

[0099] A processor is used to execute a computer program stored in a memory to implement the steps of the soil layer quantitative stratification method based on a machine learning algorithm fused with a prediction model as described above.

[0100] The present invention also provides a non-transitory readable storage medium, in which a program is stored. When the program is executed by a processor, the steps of the soil layer quantitative stratification method based on the machine learning algorithm fusion prediction model as described above are implemented.

[0101] The above-mentioned non-transitory readable storage medium can be any available medium or data storage device that can be accessed by the processor in the electronic device, including but not limited to magnetic storage such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc., optical storage such as CD, DVD, BD, HVD, etc., and semiconductor storage such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state drives (SSD), etc.

[0102] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0106] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A soil layer quantitative stratification method based on a machine learning algorithm fusion prediction model, characterized in that: The method comprises the following steps: S1. Collection and collation of pore pressure static penetration data and soil layer classification information; Collect and organize geotechnical engineering investigation reports, geotechnical test results, and organize penetration data along the depth of the piezocereography holes and soil layer classification information; S2, use machine learning algorithms SVM, RF, and XGBoost to build a basic model; The machine learning algorithms SVM, RF and XGBoost are used to establish basic prediction models with pore pressure static penetration test data as input and soil layer types as output. The output results are the probability of soil layers corresponding to each soil type. S3, using Bayesian optimization algorithm to optimize the basic model hyperparameters; The Bayesian optimization algorithm is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models used for quantitative stratification of soil layers by pore pressure penetration testing. S4, establish a fusion prediction model of three basic models in a stacking manner; A linear meta-model is established to combine the output results of the three basic models in a stacking manner to form a machine learning algorithm fusion prediction model; S5. Training the machine learning algorithm fusion prediction model; The collated pore pressure static penetration test data and soil layer classification information are input into the machine learning algorithm fusion prediction model to train the model; S6. Use the trained machine learning algorithm to fuse the prediction model to predict the soil type of another site; Use the trained machine learning algorithm to integrate the prediction model to predict the soil type of another site, output the probability of each soil type corresponding to the soil layer, and take the soil type corresponding to the maximum probability value as the predicted soil type of the soil layer; S7, determining the division accuracy, merging the stratification results, and finally obtaining the quantitative stratification results of the soil layers; The accuracy is divided according to the engineering needs, the prediction results are merged, and finally the quantitative stratification results of the soil layers are obtained.

2. The method according to claim 1, characterized in that: In step S1, the pore pressure static penetration test data includes: cone tip resistance q c , side wall friction f s , pore water pressure u2.

3. The method according to claim 1, characterized in that: In step S1, the soil layer classification information includes: silt and silty soil, silty clay, silty soil, and silty sand, and the corresponding soil type label values ​​are assigned as 0, 1, 2, and 3 respectively.

4. The method according to claim 1, characterized in that: In step S3, the three basic models of SVM, RF, and XGBoost use Bayesian optimization to find a set of optimal hyperparameters, as shown below:

5. The method according to claim 1, characterized in that: In step S4, the specific process of establishing the linear metamodel is: calculating the weighted average of the probability of each soil type corresponding to the soil layer in the results of each basic model, and the weight is determined by the 10-fold cross-validation results of each basic model on the basic model training set.

6. The method according to claim 1, characterized in that: In step S5, 80% of the data set used to train the model is randomly divided for training the basic model, and the remaining 20% ​​is used to test the meta-model. The meta-model must have better prediction results than all the basic models before it can be used to predict the soil type of another site.

7. The method according to claim 1, characterized in that: In step S7, the specific process of merging layers is: merging the same type of soil layers whose total thickness is less than the division accuracy into the previous layer.

8. A soil layer quantitative stratification system, characterized by: The soil layer quantitative stratification system includes the following modules: A data collection module, which is used to collect and organize geotechnical engineering investigation reports, geotechnical test results, and to organize penetration data along the depth of the piezocone penetration holes and soil layer classification information; A basic model building module, wherein the basic model building module is used to use machine learning algorithms SVM, RF, and XGBoost to respectively establish basic prediction models with the pore pressure static penetration data obtained by the data collection module as input and the soil layer category obtained by the data collection module as output, and the output result is the probability of the soil layer corresponding to each soil category; A hyperparameter optimization module, which is used to optimize the hyperparameters of the SVM, RF, and XGBoost basic models obtained by the basic model building module using a Bayesian optimization algorithm; A fusion prediction model building module, wherein the fusion prediction model is used to build a linear meta-model, and the output results of the three basic models obtained by the basic model building module are combined in a stacking manner to form a machine learning algorithm fusion prediction model; A fusion prediction model training module, wherein the fusion prediction model training module is used to input the pore pressure static penetration data and soil layer classification information obtained by the data collection module into the machine learning algorithm fusion prediction model obtained by the fusion prediction model establishment module to train the model; A prediction module, wherein the prediction module is used to predict the soil layer type of another site by using the machine learning algorithm fusion prediction model trained by the fusion prediction model training module, output the probability of the soil layer corresponding to each soil type, and take the soil type corresponding to the maximum probability value as the predicted soil type of the soil layer; The stratification module is used to determine the division accuracy, and to perform layer-by-layer processing on the stratification results obtained by the prediction module to finally obtain the quantitative stratification results of the soil layers.

9. An electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus, characterized in that: A memory, the memory being used to store a computer program; A processor, wherein the processor is used to execute a computer program stored in a memory to implement the steps of a soil layer quantitative stratification method based on a machine learning algorithm fused with a prediction model as described in any one of claims 1 to 7.

10. A non-transitory readable storage medium, characterized in that: The non-transitory readable storage medium stores a program, and when the program is executed by the processor, it implements the steps of the soil layer quantitative stratification method based on the machine learning algorithm fusion prediction model as described in any one of claims 1-7.

Images