Ploughing depth detection method and system based on ploughing equipment factor data

Abstract

The invention provides a tilling depth detection method based on tilling equipment factor data, and the method comprises the following steps: S1, collecting equipment factor data, collecting corresponding tilling depth data, and taking the collected equipment factor data and tilling depth data as a training set; s2, selecting a regression relation model: selecting the regression relation model, training the regression relation model by using the training set, and determining parameters of the regression relation model by minimizing a loss function (such as a mean square error); s3, regression prediction: using the regression relation model trained in S2 to predict new ploughing equipment data to obtain a corresponding ploughing depth prediction value; s4, neural network prediction; s5, adjusting and optimizing; S5.1, carrying out cross validation; s5.2, adjusting hyper-parameters; and S5.3, integrated adjustment and optimization: performing adjustment and optimization on the neural network model by adopting a mode of combining an integrated learning method and a machine learning algorithm, thereby improving the performance of the neural network model.

Description

Technical Field

[0001] The present disclosure relates to the technical field of agricultural machinery, and in particular to a tillage depth detection method based on tillage equipment factor data and a system thereof. Background Art

[0002] In agricultural production, accurate detection of tillage depth of tillage equipment is crucial to optimizing soil structure and increasing crop yields. Traditional tillage depth detection methods often have problems such as low accuracy and poor adaptability. With the development of technology, using the factor data of tillage equipment itself to detect tillage depth has become a new idea, which can improve the accuracy and reliability of tillage depth detection.

[0003] There is currently no corresponding solution on the market. Summary of the invention

[0004] In response to the above-mentioned technical problems in the related art, the present disclosure proposes a tillage depth detection system based on tillage equipment factor data, which can screen out products within the required power range and capable of constant power control before installation, and can overcome the above-mentioned shortcomings of the prior art.

[0005] To achieve the above technical objectives, the technical solution of the present disclosure is implemented as follows: The first object of the present disclosure is to provide a tillage depth detection method based on tillage equipment factor data, which comprises the following steps: S1: Collect equipment factor data and corresponding tillage depth data, and use the collected equipment factor data and tillage depth data as training sets; S2: selecting a regression relationship model: selecting a regression relationship model, training the regression relationship model using the training set, and determining the parameters of the regression relationship model by minimizing a loss function (such as mean square error); S3 regression prediction: using the regression relationship model trained in S2 to predict the new tillage equipment data, and obtain the corresponding tillage depth prediction value; S4 neural network prediction: S4.1: construct a neural network model, including an input layer, a hidden layer and an output layer; the input layer is used to receive the equipment factor data, and the output layer is used to output a tillage depth prediction value; S4.2: Selecting an activation function, a loss function and an optimization algorithm for the neural network model; S4.3: training the neural network model using the training set, and minimizing the loss function by continuously adjusting the weights and biases of the neural network model; S4.4: Use the neural network model trained in S4.3 to predict the new equipment factor data to obtain a new estimated value of the tillage depth; S5 Tuning: S5.1 Cross-validation: Divide the training set into several subsets of equal size, select one subset as the validation set and the other subsets as the training set in a cross-validation manner, repeatedly train and validate the neural network model, and obtain the performance of the neural network model on different data subsets, so as to evaluate the stability and generalization ability of the neural network; S5.2 Adjust hyperparameters: According to the results of cross-validation in S5.1.1, gradually adjust the hyperparameters of the neural network model, and apply the adjusted hyperparameters with the best performance to the neural network model, thereby improving the performance of the neural network model; the hyperparameters include the number of layers, number of neurons, learning rate, etc. of the neural network model. S5.3 Integrated tuning: The neural network model is tuned by combining an integrated learning method and a machine learning algorithm to improve the performance of the neural network model.

[0006] Preferably, the S5.2 specifically includes the following sub-steps: S5.2.1: According to the result of cross validation according to S5.1.1, select the hyperparameters that need to be adjusted from the neural network model; S5.2.2: For all the hyperparameters that need to be adjusted, set corresponding parameter ranges and step sizes, and then perform grid search in the corresponding parameter ranges; S5.2.3: For all the hyperparameters that need to be adjusted, use the training set to train, and evaluate and record them on the corresponding validation set; S5.2.4: Select the hyperparameters with the best performance on the validation set as the final model parameters.

[0007] Preferably, the S5.3 specifically includes the following sub-steps: S5.3.1 Use ensemble learning methods to combine multiple weak learners into a strong learner, integrate multiple machine learning algorithms, or integrate different parameter settings of the same algorithm to generate an ensemble learning model; S5.3.2 Tuning the ensemble learning model: Tuning the ensemble learning model, adjusting hyperparameters such as the ensemble method and the number of weak learners, to obtain better performance; S5.3.3 Tuning the neural network model: Tuning the neural network model through the tuned ensemble learning model.

[0008] Preferably, the ensemble learning method may be a random forest or a gradient boosting tree.

[0009] Preferably, the activation function includes a sigmoid function, a ReLU function, etc.

[0010] Preferably, the loss function is preferably mean square error, cross entropy, etc.

[0011] Preferably, the optimization algorithm is preferably stochastic gradient descent (SGD), Adam, etc. Preferably, the regression relationship model is preferably a linear regression model. It can be implemented as follows: y=w0+w1x1+w2x2+...+w n x n ; where: y is the tillage depth, x1, x2,…x n is the farmland equipment factor data, w0, w1, w2, ... w n is the corresponding model parameter. The linear regression model can be implemented using existing technology.

[0012] The second object of the present disclosure is to provide a system for the tillage depth detection method, comprising: Farming equipment data collection module: collects the equipment factor data of the farming equipment during operation, including vibration amplitude data, wear degree data, deformation data, and power output data; Data transmission module: transmits the equipment factor data collected by the farmland equipment data collection module to the data processing module; Data processing module: Analyze and process the received equipment factor data, establish a regression relationship model between the equipment factor data and the tillage depth; fuse the equipment factor data of different types, so as to comprehensively consider the influence of various factors on the tillage depth; and optimize the regression relationship model to improve the accuracy and generalization ability of the regression relationship model; A tillage depth calculation module: a module for calculating the tillage depth of the tillage equipment according to the equipment factor data processed by the data processing module, wherein the tillage depth includes the shovel depth under the tillage ground; Result display module: intuitively displays the tillage depth result calculated by the tillage depth calculation module to the user.

[0013] Preferably, the farming equipment data collection module includes: A vibration detection unit is installed at a key part of the tillage equipment to collect the vibration amplitude data of the tillage equipment during operation; the key part is a frame or a connection between working parts, etc.; A wear detection unit monitors the wear degree data of key components of the tillage equipment, including plowshares, rotary tillers, etc. For example, a camera may be used to capture images of the components and the wear conditions may be determined through an image analysis algorithm; or a displacement sensor may be used to monitor the dimensional changes of the components; The deformation detection unit uses a strain gauge or a laser displacement sensor to detect the deformation of the structural components of the tillage equipment; the strain gauge infers the degree of deformation by measuring the strain, and the laser displacement sensor measures the displacement change of the structural component in a non-contact manner; The power parameter detection unit detects the power output parameters of the tillage equipment, wherein the power output parameters include the traction force, engine speed, etc. of the tillage equipment.

[0014] Preferably, the farming equipment data collection module includes: The vibration detection unit is preferably an acceleration sensor; The wear detection unit uses image recognition technology or sensors to monitor the wear degree data of key components of the tillage equipment. For example, a camera can be used to capture an image of the component and the wear condition can be determined by an image analysis algorithm; or a displacement sensor can be used to monitor the dimensional change of the component. The structural component includes a frame. The strain gauge is attached to the key position.

[0015] Preferably, the data processing module includes: A data analysis unit, using a machine learning algorithm to establish the regression relationship model between the equipment factor data and the tillage depth; A data fusion unit is used to fuse the data of equipment factors belonging to different types, so as to comprehensively consider the influence of the vibration amplitude data, wear degree data, deformation data, and power output data on the tillage depth result; The model tuning unit tunes the regression relationship model according to the fusion result of the data fusion unit to improve the accuracy and generalization ability of the regression relationship model.

[0016] Preferably, the machine learning algorithm may be regression analysis, neural network, etc.; During the fusion operation, the data fusion unit may perform weighted fusion on the vibration data, wear data, deformation data, and dynamic parameter data; The model tuning unit may use methods such as cross-validation or grid search to adjust the hyperparameters of the regression relationship model, or use techniques such as ensemble learning to improve the performance of the regression relationship model, thereby improving the accuracy and generalization ability of the regression relationship model.

[0017] Preferably, the result display module is a display screen or a mobile phone APP, etc.

[0018] Beneficial effects of the present disclosure: The present disclosure collects data on factors such as vibration, wear, deformation and dynamic parameters of tillage equipment, and uses machine learning algorithms to establish a regression relationship model between these data and tillage depth, thereby achieving accurate detection of tillage depth. The present disclosure improves the accuracy and generalization ability of the model by adding a model tuning process, so that it can better adapt to different tillage equipment and operating conditions. The present disclosure provides an efficient and accurate tillage depth detection method for agricultural production, helping farmers optimize arable land operations and improve agricultural production efficiency and quality.

[0019] The present invention uses farmland equipment factor data to detect tillage depth, thereby improving the accuracy and reliability of tillage depth detection. The present invention uses a machine learning algorithm to establish a regression relationship model between farmland equipment factor data and tillage depth, which can automatically learn complex patterns in the data and adapt to different farmland equipment and operating conditions.

[0020] The system provided by the present disclosure has good scalability and adaptability. New farmland equipment factor data can be added according to actual needs to improve the accuracy of tillage depth detection. The addition of the model tuning process improves the accuracy and generalization ability of the model, so that it can better adapt to different data sets and actual application scenarios. The result display module displays the tillage depth results to the user in an intuitive way, which is convenient for the user to adjust the operating parameters of the farmland equipment in time and improve the efficiency and quality of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 It is a logic block diagram of the system of the tillage depth detection method described in this application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present disclosure.

[0024] like Figure 1 As shown, in order to facilitate the understanding of the above technical solutions of the present disclosure, the above technical solutions of the present disclosure are described in detail below through specific usage methods.

[0025] The first object of the present disclosure is to provide a tillage depth detection method based on tillage equipment factor data, which comprises the following steps: S1: Collect equipment factor data and corresponding tillage depth data, and use the collected equipment factor data and tillage depth data as training sets; S2: selecting a regression relationship model: selecting a regression relationship model, training the regression relationship model using the training set, and determining the parameters of the regression relationship model by minimizing a loss function (such as mean square error); S3 regression prediction: using the regression relationship model trained in S2 to predict the new tillage equipment data, and obtain the corresponding tillage depth prediction value; S4 neural network prediction: S4.1: construct a neural network model, including an input layer, a hidden layer and an output layer; the input layer is used to receive the equipment factor data, and the output layer is used to output a tillage depth prediction value; S4.2: Selecting an activation function, a loss function and an optimization algorithm for the neural network model; S4.3: training the neural network model using the training set, and minimizing the loss function by continuously adjusting the weights and biases of the neural network model; S4.4: Use the neural network model trained in S4.3 to predict the new equipment factor data to obtain a new estimated value of the tillage depth; S5 Tuning: S5.1 Cross-validation: Divide the training set into several subsets of equal size, select one subset as the validation set and the other subsets as the training set in a cross-validation manner, repeatedly train and validate the neural network model, and obtain the performance of the neural network model on different data subsets, so as to evaluate the stability and generalization ability of the neural network; S5.2 Adjust hyperparameters: According to the results of cross-validation in S5.1.1, gradually adjust the hyperparameters of the neural network model, and apply the adjusted hyperparameters with the best performance to the neural network model, thereby improving the performance of the neural network model; the hyperparameters include the number of layers, number of neurons, learning rate, etc. of the neural network model. S5.3 Integrated tuning: The neural network model is tuned by combining an integrated learning method and a machine learning algorithm to improve the performance of the neural network model.

[0026] In a specific embodiment, the S5.2 specifically includes the following sub-steps: S5.2.1: According to the result of cross validation according to S5.1.1, select the hyperparameters that need to be adjusted from the neural network model; S5.2.2: For all the hyperparameters that need to be adjusted, set corresponding parameter ranges and step sizes, and then perform grid search in the corresponding parameter ranges; S5.2.3: For all the hyperparameters that need to be adjusted, use the training set to train, and evaluate and record them on the corresponding validation set; S5.2.4: Select the hyperparameters with the best performance on the validation set as the final model parameters.

[0027] In a specific embodiment, the S5.3 specifically includes the following sub-steps: S5.3.1 Use ensemble learning methods to combine multiple weak learners into a strong learner, integrate multiple machine learning algorithms, or integrate different parameter settings of the same algorithm to generate an ensemble learning model; S5.3.2 Tuning the ensemble learning model: Tuning the ensemble learning model, adjusting hyperparameters such as the ensemble method and the number of weak learners, to obtain better performance; S5.3.3 Tuning the neural network model: Tuning the neural network model through the tuned ensemble learning model.

[0028] In a specific embodiment, the ensemble learning method may be a random forest or a gradient boosting tree.

[0029] In a specific embodiment, the activation function includes a sigmoid function, a ReLU function, etc.

[0030] In a specific embodiment, the loss function is preferably mean square error, cross entropy, etc.

[0031] In a specific embodiment, the optimization algorithm is preferably stochastic gradient descent (SGD), Adam, etc. In a specific embodiment, the regression relationship model is preferably a linear regression model. It can be implemented as follows: y=w0+w1x1+w2x2+...+w n x n ; where: y is the tillage depth, x1, x2, ... x n is the farmland equipment factor data, w0, w1, w2, … w n is the corresponding model parameter. The linear regression model can be implemented using existing technology.

[0032] The second object of the present disclosure is to provide a system for the tillage depth detection method, comprising: Farming equipment data collection module: collects the equipment factor data of the farming equipment during operation, including vibration amplitude data, wear degree data, deformation data, and power output data; Data transmission module: transmits the equipment factor data collected by the farmland equipment data collection module to the data processing module; Data processing module: Analyze and process the received equipment factor data, establish a regression relationship model between the equipment factor data and the tillage depth; fuse the equipment factor data of different types, so as to comprehensively consider the influence of various factors on the tillage depth; and optimize the regression relationship model to improve the accuracy and generalization ability of the regression relationship model; A tillage depth calculation module: a module for calculating the tillage depth of the tillage equipment according to the equipment factor data processed by the data processing module, wherein the tillage depth includes the shovel depth under the tillage ground; Result display module: intuitively displays the tillage depth result calculated by the tillage depth calculation module to the user.

[0033] In a specific embodiment, the farming equipment data collection module includes: A vibration detection unit is installed at a key part of the tillage equipment to collect the vibration amplitude data of the tillage equipment during operation; the key part is a frame or a connection between working parts, etc.; A wear detection unit monitors the wear degree data of key components of the tillage equipment, including plowshares, rotary tillers, etc. For example, a camera may be used to capture images of the components and the wear conditions may be determined through an image analysis algorithm; or a displacement sensor may be used to monitor the dimensional changes of the components; The deformation detection unit uses a strain gauge or a laser displacement sensor to detect the deformation of the structural components of the tillage equipment; the strain gauge infers the degree of deformation by measuring the strain, and the laser displacement sensor measures the displacement change of the structural component in a non-contact manner; The power parameter detection unit detects the power output parameters of the tillage equipment, wherein the power output parameters include the traction force, engine speed, etc. of the tillage equipment.

[0034] In a specific embodiment, the farming equipment data collection module includes: The vibration detection unit is preferably an acceleration sensor; The wear detection unit uses image recognition technology or sensors to monitor the wear degree data of key components of the tillage equipment. For example, a camera can be used to capture an image of the component and the wear condition can be determined by an image analysis algorithm; or a displacement sensor can be used to monitor the dimensional change of the component. The structural component includes a frame. The strain gauge is attached to the key position.

[0035] In a specific embodiment, the data processing module includes: A data analysis unit, using a machine learning algorithm to establish the regression relationship model between the equipment factor data and the tillage depth; A data fusion unit is used to fuse the data of equipment factors belonging to different types, so as to comprehensively consider the influence of the vibration amplitude data, wear degree data, deformation data, and power output data on the tillage depth result; The model tuning unit tunes the regression relationship model according to the fusion result of the data fusion unit to improve the accuracy and generalization ability of the regression relationship model.

[0036] In a specific embodiment, the machine learning algorithm may be regression analysis, neural network, etc. During the fusion operation, the data fusion unit may perform weighted fusion on the vibration data, wear data, deformation data, and dynamic parameter data; The model tuning unit may use methods such as cross-validation or grid search to adjust the hyperparameters of the regression relationship model, or use techniques such as ensemble learning to improve the performance of the regression relationship model, thereby improving the accuracy and generalization ability of the regression relationship model.

[0037] In a specific embodiment, the result display module is a display screen or a mobile phone APP, etc.

[0038] Working principle: Regression analysis is a statistical method used to establish the relationship between independent variables and dependent variables. In the present invention, the vibration amplitude, wear degree, deformation and dynamic parameters of the tillage equipment are used as independent variables, and the tillage depth is used as the dependent variable. Through regression analysis, a mathematical model between the two is established. Different regression relationship models such as linear regression and polynomial regression can be used, and the appropriate model can be selected according to the characteristics of the actual data.

[0039] In summary, through the unique design of the present invention, the present invention uses the farmland equipment factor data to detect the tillage depth, thereby improving the accuracy and reliability of the tillage depth detection. The present invention uses a machine learning algorithm to establish a regression relationship model between the farmland equipment factor data and the tillage depth, which can automatically learn the complex patterns in the data and adapt to different farmland equipment and working conditions.

[0040] The system provided by the present disclosure has good scalability and adaptability. New farmland equipment factor data can be added according to actual needs to improve the accuracy of tillage depth detection. The addition of the model tuning process improves the accuracy and generalization ability of the model, so that it can better adapt to different data sets and actual application scenarios. The result display module displays the tillage depth results to the user in an intuitive way, which is convenient for the user to adjust the operating parameters of the farmland equipment in time and improve the efficiency and quality of agricultural production.

[0041] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A tillage depth detection method based on tillage equipment factor data, characterized in that: The following steps are involved: S1: Collect equipment factor data and corresponding tillage depth data, and use the collected equipment factor data and tillage depth data as training sets; S2: selecting a regression relationship model: selecting a regression relationship model, training the regression relationship model using the training set, and determining the parameters of the regression relationship model by minimizing a loss function (such as mean square error); S3 regression prediction: using the regression relationship model trained in S2 to predict the new tillage equipment data, and obtain the corresponding tillage depth prediction value; S4 neural network prediction: S4.1: construct a neural network model, including an input layer, a hidden layer and an output layer; the input layer is used to receive the equipment factor data, and the output layer is used to output a tillage depth prediction value; S4.2: Selecting an activation function, a loss function and an optimization algorithm for the neural network model; S4.3: training the neural network model using the training set, and minimizing the loss function by continuously adjusting the weights and biases of the neural network model; S4.4: Use the neural network model trained in S4.3 to predict the new equipment factor data to obtain a new estimated value of the tillage depth; S5 Tuning: S5.1 Cross-validation: Divide the training set into several subsets of equal size, select one subset as the validation set and the other subsets as the training set in a cross-validation manner, repeatedly train and validate the neural network model, and obtain the performance of the neural network model on different data subsets, so as to evaluate the stability and generalization ability of the neural network; S5.2 Adjusting hyperparameters: According to the results of the cross-validation in S5.1.1, gradually adjusting the hyperparameters of the neural network model, and applying the adjusted hyperparameters with the best performance to the neural network model, thereby improving the performance of the neural network model; the hyperparameters include the number of layers, the number of neurons, and the learning rate of the neural network model; S5.3 Integrated tuning: The neural network model is tuned by combining an integrated learning method and a machine learning algorithm to improve the performance of the neural network model.

2. The tillage depth detection method according to claim 1, characterized in that: The S5.2 specifically includes the following sub-steps: S5.2.1: According to the result of cross validation according to S5.1.1, select the hyperparameters that need to be adjusted from the neural network model; S5.2.2: For all the hyperparameters that need to be adjusted, set corresponding parameter ranges and step sizes, and then perform grid search in the corresponding parameter ranges; S5.2.3: For all the hyperparameters that need to be adjusted, use the training set to train, and evaluate and record them on the corresponding validation set; S5.2.4: Select the hyperparameters with the best performance on the validation set as the final model parameters.

3. The tillage depth detection method according to claim 1, characterized in that: The S5.3 specifically includes the following sub-steps: S5.3.1 Use ensemble learning methods to combine multiple weak learners into a strong learner, integrate multiple machine learning algorithms, or integrate different parameter settings of the same algorithm to generate an ensemble learning model; S5.3.2 Tuning the ensemble learning model: Tuning the ensemble learning model, adjusting the ensemble method and the number of weak learners to obtain better performance; S5.3.3 Tuning the neural network model: Tuning the neural network model through the tuned integrated learning model.

4. The tillage depth detection method according to claim 1, characterized in that: The ensemble learning method is a random forest or a gradient boosted tree; The activation functions include sigmoid function and ReLU function; The loss function is mean square error, cross entropy; The optimization algorithms are stochastic gradient descent (SGD) and Adam; The regression relationship model is a linear regression model.

5. A system used in the tillage depth detection method according to any one of claims 1 to 4, characterized in that: include: Farming equipment data collection module: collects the equipment factor data of the farming equipment during operation, including vibration amplitude data, wear degree data, deformation data, and power output data; Data transmission module: transmits the equipment factor data collected by the farmland equipment data collection module to the data processing module; Data processing module: Analyze and process the received equipment factor data, establish a regression relationship model between the equipment factor data and the tillage depth; fuse the equipment factor data of different types, so as to comprehensively consider the influence of various factors on the tillage depth; and optimize the regression relationship model to improve the accuracy and generalization ability of the regression relationship model; A tillage depth calculation module: a module for calculating the tillage depth of the tillage equipment according to the equipment factor data processed by the data processing module, wherein the tillage depth includes the shovel depth under the tillage ground; Result display module: intuitively displays the tillage depth result calculated by the tillage depth calculation module to the user.

6. The system according to claim 5, characterized in that The farming equipment data acquisition module comprises: A vibration detection unit is installed at a key part of the tillage equipment to collect the vibration amplitude data of the tillage equipment during operation, wherein the key part is a frame or a connection point of a working part; A wear detection unit, which monitors the wear degree data of key components of the tillage equipment, wherein the key components include a plowshare and a rotary tiller; The deformation detection unit uses a strain gauge or a laser displacement sensor to detect the deformation of the structural components of the tillage equipment; the strain gauge infers the degree of deformation by measuring the strain, and the laser displacement sensor measures the displacement change of the structural components in a non-contact manner; The power parameter detection unit detects the power output parameters of the tillage equipment, wherein the power output parameters include the traction force and engine speed of the tillage equipment.

7. The system according to claim 6, characterized in that The farming equipment data acquisition module comprises: The vibration detection unit is an acceleration sensor; The wear detection unit uses image recognition technology or sensors to monitor the wear degree data of key components of the tillage equipment; The structural components include a frame; The strain gauge is pasted on the key position.

8. The system according to claim 5, characterized in that The data processing module comprises: A data analysis unit, using a machine learning algorithm to establish the regression relationship model between the equipment factor data and the tillage depth; A data fusion unit is used to fuse the data of equipment factors belonging to different types, so as to comprehensively consider the influence of the vibration amplitude data, wear degree data, deformation data, and power output data on the tillage depth result; The model tuning unit tunes the regression relationship model according to the fusion result of the data fusion unit to improve the accuracy and generalization ability of the regression relationship model.

9. The system according to claim 8, characterized in that The machine learning algorithm is regression analysis and neural network; The data fusion unit performs a fusion operation to weightedly fuse the vibration data, wear data, deformation data, and dynamic parameter data; The model tuning unit uses cross-validation or grid search to adjust the hyperparameters of the regression relationship model, or uses ensemble learning to improve the performance of the regression relationship model, thereby improving the accuracy and generalization ability of the regression relationship model.

10. The tillage depth detection system according to claim 5, characterized in that: The result display module is a display screen or a mobile phone APP.

Images