Method and device for predicting wear performance of soil entering part of agricultural machine

Abstract

The application discloses a kind of agricultural machinery earth-penetrating component wear performance prediction method and device, it is related to deep learning technical field, obtains the component basic information and working condition information of the agricultural machinery earth-penetrating component to be predicted;The component basic information and working condition information of the agricultural machinery earth-penetrating component to be predicted are input into the pre-trained agricultural machinery earth-penetrating component wear performance prediction model, and the predicted wear of the agricultural machinery earth-penetrating component to be predicted and predicted wear appearance are obtained.Such, the prediction efficiency of agricultural machinery earth-penetrating component wear performance is improved.

Description

Technical Field

[0001] This application relates to the field of deep learning technology, and in particular to a method and device for predicting the wear performance of agricultural machinery soil-entry components. Background Technology

[0002] As the core actuators in the tillage process, the wear resistance of the materials used in agricultural machinery directly determines the operating efficiency, tillage quality, and maintenance costs of the machinery.

[0003] In existing technologies, dedicated wear performance testing benches are typically used to simulate working conditions for agricultural machinery components that enter the soil, and the wear performance is evaluated by manually collecting and analyzing the test data. However, this method is highly dependent on manual operation, has low data processing efficiency, is highly subjective, and has a long testing cycle and requires significant investment in manpower and equipment, resulting in high overall testing costs and making it difficult to meet the needs of rapid iteration and efficient R&D in modern agricultural equipment. Summary of the Invention

[0004] The purpose of this application is to provide a method and device for predicting the wear performance of agricultural machinery soil-entry components, which can improve the efficiency of predicting the wear performance of agricultural machinery soil-entry components.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for predicting the wear performance of agricultural machinery soil-entry components, including: Obtain basic component information and operating condition information of the agricultural machinery soil-entry parts to be predicted; The basic information and working condition information of the agricultural machinery soil-entry component to be predicted are input into the pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted.

[0006] In an optional embodiment, the method for predicting the wear performance of agricultural machinery soil-entry components further includes: Obtain basic information about the soil-entry components of the sample agricultural machinery, operating condition information, and the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery; The wear performance prediction model of agricultural machinery soil entry components was trained using the basic component information, working condition information, actual wear amount and actual wear appearance of the sample agricultural machinery soil entry components.

[0007] In an optional embodiment, the method for predicting the wear performance of agricultural machinery soil-entry components further includes: The working process of the soil-entry component of the sample agricultural machinery was simulated to obtain simulation data; A model for predicting the wear performance of agricultural machinery soil-entry components was trained using simulation data.

[0008] In an optional embodiment, before training the wear performance prediction model for agricultural machinery soil-entry components using simulation data, the method for predicting the wear performance of agricultural machinery soil-entry components further includes: Based on the basic information of the sample agricultural machinery's soil-entry components, operating condition information, and the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry components, the simulation data is filtered to obtain the filtered simulation data. A model for predicting the wear performance of agricultural machinery soil-entry components was trained using simulation data, including: A model for predicting the wear performance of agricultural machinery soil-entry components was trained using the selected simulation data.

[0009] In one optional embodiment, the wear performance prediction model for agricultural machinery soil-entry components includes LightGBM and PINN. The model is trained using basic component information, operating condition information, and the actual wear amount and appearance of the sample agricultural machinery soil-entry components. The training includes: LightGBM was trained using the basic information and working condition information of the soil-entry components of sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of sample agricultural machinery. PINN was trained using the basic information and working condition information of the soil-entry components of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery.

[0010] In one optional embodiment, PINN is trained using the basic component information, operating condition information, and actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry component, including: The basic information and operating condition information of the sample agricultural machinery's soil-entry components, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry components, are used for iterative training until the termination condition is met. Each iteration of training includes: Input the basic information and working condition information of the soil-entry components of the sample agricultural machinery into PINN, and output the predicted wear amount and predicted wear appearance of the soil-entry components of the sample agricultural machinery. The target loss is calculated based on the predicted wear amount and predicted wear appearance of the soil-entry parts of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry parts of the sample agricultural machinery. The parameters of PINN are adjusted based on the target loss.

[0011] In one optional embodiment, the target loss is calculated based on the predicted wear and predicted wear appearance of the soil-entry components of the sample agricultural machinery, as well as the actual wear and actual wear appearance of the soil-entry components of the sample agricultural machinery, using the following formula: ; ; in, Represents the total loss function. This represents the mean square error between the predicted wear amount and the actual wear amount of the soil-entry components of the sample agricultural machinery. This represents the mean square error between the predicted wear appearance and the actual wear appearance of the soil-entry components of the sample agricultural machinery. Indicates the balance hyperparameters, V represents physical loss, P represents the actual wear amount, L represents the normal load of the agricultural machinery's soil-entry component in the abrasive grains, H represents the material hardness of the agricultural machinery's soil-entry component, and k represents the dimensionless wear coefficient.

[0012] In one optional embodiment, the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted are input into a pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted, including: Input the basic information and working condition information of the agricultural machinery soil-entry components to be predicted into LightGBM to obtain the first prediction result; Input the basic information and working condition information of the agricultural machinery soil-entry components to be predicted into PINN to obtain the second prediction result; The target prediction result is obtained by dynamically weighted linearly fusing the first and second prediction results. The target prediction result includes the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry parts to be predicted.

[0013] In one optional embodiment, the basic component information includes at least: the substrate material, coating material, coating process, and component type of the agricultural machinery soil-entry component; the working condition information includes at least: the torque value, rotational speed, temperature, force, depth, and abrasive particle size of the agricultural machinery soil-entry component in the abrasive particles.

[0014] Secondly, this application provides a device for predicting the wear performance of agricultural machinery soil-entry components, comprising: The acquisition module is used to acquire the basic information and operating condition information of the agricultural machinery soil-entry components to be predicted; The prediction module is used to input the basic information and working condition information of the agricultural machinery soil-entry component to be predicted into the pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component.

[0015] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for predicting the wear performance of agricultural machinery soil-entry components as described above.

[0016] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting the wear performance of agricultural machinery soil-entry components as described above.

[0017] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method for predicting the wear performance of agricultural machinery soil-entry components as described above.

[0018] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and apparatus for predicting the wear performance of agricultural machinery soil-entry components. The method involves acquiring the basic information and operating condition information of the components to be predicted; inputting this information into a pre-trained wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the components. Compared to existing technologies, this method improves the efficiency of predicting the wear performance of agricultural machinery soil-entry components. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an application environment diagram of a method for predicting the wear performance of agricultural machinery soil-entry components according to an embodiment of this application; Figure 2 A flowchart illustrating a method for predicting the wear performance of agricultural machinery soil-entry components, provided in an embodiment of this application; Figure 3 This is a schematic diagram of the functional modules of a device for predicting the wear performance of agricultural machinery soil-entry components, provided in an embodiment of this application. Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] The method and apparatus for predicting the wear performance of agricultural machinery soil-entry components provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the basic component information and operating condition information of the agricultural machinery soil-entry component to be predicted to server 104. After receiving the basic component information and operating condition information, server 104 uses a trained agricultural machinery soil-entry component wear performance prediction model to predict the wear performance of the agricultural machinery soil-entry component. Server 104 can then feed back the prediction results to terminal 102. In addition, in some embodiments, the method for predicting the wear performance of agricultural machinery soil-entry components can also be implemented by the server 104 or the terminal 102 alone. For example, the terminal 102 can predict the wear performance of agricultural machinery soil-entry components, or the server 104 can obtain the basic information and operating condition information of the agricultural machinery soil-entry components to be predicted from the data storage system, and input the obtained basic information and operating condition information of the agricultural machinery soil-entry components to be predicted into the trained agricultural machinery soil-entry component wear performance prediction model to predict the wear performance of the agricultural machinery soil-entry components.

[0024] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0025] In one exemplary embodiment, such as Figure 2 As shown, a method for predicting the wear performance of agricultural machinery soil-entry components is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 202. Wherein: Step S201: Obtain the basic information and working condition information of the agricultural machinery soil-entry components to be predicted.

[0026] In one alternative embodiment, the agricultural machinery soil-entry component can be a plowshare, rotary tiller, or disc harrow, etc., and this application does not limit this to any particular type.

[0027] In another optional embodiment, the basic information of the components includes at least: the base material, coating material, coating process and component type of the agricultural machinery soil-entry component; the working condition information includes at least: the torque value, rotational speed value, temperature, force, depth and abrasive particle size of the agricultural machinery soil-entry component in the abrasive particles.

[0028] It should be noted that the torque value of the agricultural machinery's soil-entry component refers to the rotational resistance torque output by the motor or drive shaft when the component rotates within the abrasive grains; the speed value refers to the angular velocity of the component's rotation around its own axis; the temperature refers to the surface temperature of the component when it rotates within the abrasive grains; and the force refers to the reaction force experienced by the component. Depth refers to the depth of the component within the abrasive grains during rotation. The abrasive grains are used to simulate the soil environment.

[0029] In one specific embodiment, the abrasive grains can be pebbles, irregular stones, tree roots, or crop straw, etc., and this application does not limit them.

[0030] In another specific embodiment, the operating condition information can be data obtained by testing the agricultural machinery's soil-entry component using an agricultural machinery soil-entry component abrasive wear performance test bench. The dynamic torque sensor in the test bench measures the rotational speed and torque values ​​of the agricultural machinery's soil-entry component within the abrasive particles. A temperature sensor in the test bench measures the surface temperature of the agricultural machinery's soil-entry component when it rubs against the abrasive particles. A tillage depth measuring instrument in the test bench measures the depth of the agricultural machinery's soil-entry component within the abrasive particles. It should be noted that the agricultural machinery soil-entry component abrasive wear performance test bench is an existing test bench, and this application does not limit its specific structure.

[0031] Step S202: Input the basic information and working condition information of the agricultural machinery soil entry component to be predicted into the pre-trained agricultural machinery soil entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil entry component to be predicted.

[0032] In an optional embodiment, the method for predicting the wear performance of agricultural machinery soil-entry components provided in this application further includes: Obtain basic information about the soil-entry components of the sample agricultural machinery, operating condition information, and the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery; The wear performance prediction model of agricultural machinery soil entry components was trained using the basic component information, working condition information, actual wear amount and actual wear appearance of the sample agricultural machinery soil entry components.

[0033] In one specific embodiment, the soil-entry components of the sample agricultural machinery before and after the test are weighed using a balance and compared to obtain the actual wear amount of the soil-entry components. The balance accuracy is 0.01g. A 3D scanner is used to perform a 3D scan of the soil-entry components before and after the test, and the results are compared to obtain the actual wear appearance of the soil-entry components. The volume accuracy of the 3D scanner is 0.020mm + 0.040mm / m.

[0034] In another optional embodiment, the acquired basic information of the sample agricultural machinery's soil-entry components, operating condition information, and the actual wear amount and appearance of the soil-entry components can be subjected to data filtering and noise reduction processing. Data filtering processing may include at least one of the following methods: missing value handling, outlier handling, data deduplication, or data format and data type unification. Noise reduction processing may include at least one of the following methods: filtering or signal smoothing, etc.

[0035] In another optional embodiment, the wear performance prediction model for agricultural machinery soil-entry components includes LightGBM (Light Gradient Boosting Machine) and PINN (Physics-Informed Neural Networks). The model is trained using the basic component information, operating condition information, and actual wear amount and appearance of the sample agricultural machinery soil-entry components. This includes: training LightGBM using the basic component information, operating condition information, and actual wear amount and appearance of the sample agricultural machinery soil-entry components; and training PINN using the basic component information, operating condition information, and actual wear amount and appearance of the sample agricultural machinery soil-entry components.

[0036] It should be noted that LightGBM employs a histogram-based gradient boosting decision tree framework. During training, the operating condition information is discretized into a fixed number of bins using a histogram algorithm. The model grows a tree structure using a leaf-wise strategy, selecting the leaf node with the largest splitting gain for splitting in each iteration. By fitting the residuals from the previous iteration, it gradually approximates the actual wear amount and appearance. The loss function uses mean squared error, and multiple regression trees are trained using gradient boosting. This model can efficiently capture the nonlinear relationship between operating conditions and wear, and is naturally robust to issues such as feature dimensions and missing values.

[0037] In one specific embodiment, iterative training is performed using the basic component information and operating condition information of the sample agricultural machinery's soil-entry component, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry component, until the termination condition is met. Each iteration of training includes: inputting the basic component information and operating condition information of the sample agricultural machinery's soil-entry component into LightGBM, and outputting the predicted wear amount and predicted wear appearance of the sample agricultural machinery's soil-entry component; calculating the target loss based on the predicted wear amount and predicted wear appearance of the sample agricultural machinery's soil-entry component, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry component; and adjusting the parameters of LightGBM based on the target loss.

[0038] In another specific embodiment, the PINN model is constructed as a multi-layer fully connected feedforward neural network. The PINN is trained using the basic component information, operating condition information, and actual wear and tear measurements of the sample agricultural machinery's soil-entry components. This training includes: iterative training using the basic component information, operating condition information, and actual wear and tear measurements of the sample agricultural machinery's soil-entry components until a termination condition is met. Each iteration includes: inputting the basic component information and operating condition information of the sample agricultural machinery's soil-entry components into the PINN, and outputting the predicted wear and tear measurements of the sample agricultural machinery's soil-entry components; calculating the target loss based on the predicted wear and tear measurements, as well as the actual wear and tear measurements of the sample agricultural machinery's soil-entry components; and adjusting the parameters of the PINN based on the target loss.

[0039] In a more specific embodiment, based on the predicted wear and predicted wear appearance of the soil-entry components of the sample agricultural machinery, and the actual wear and actual wear appearance of the soil-entry components of the sample agricultural machinery, the target loss is calculated using the following formula: ; ; in, Represents the total loss function. This represents the mean square error between the predicted wear amount and the actual wear amount of the soil-entry components of the sample agricultural machinery. This represents the mean square error between the predicted wear appearance and the actual wear appearance of the soil-entry components of the sample agricultural machinery. Indicates the balance hyperparameters, V represents physical loss, P represents the actual wear amount, L represents the normal load of the agricultural machinery's soil-entry component in the abrasive grains, H represents the material hardness of the agricultural machinery's soil-entry component, and k represents the dimensionless wear coefficient.

[0040] It should be noted that during training, k can be optimized as a learnable parameter through backpropagation. It can be flexibly adjusted according to task objectives, data characteristics, and training dynamics.

[0041] In another, more specific embodiment, the termination condition can be at least one of the following conditions: The validation set error is less than the preset convergence threshold, which can be set according to the actual situation. The decrease in validation set error after a preset number of consecutive iterations is less than a preset magnitude threshold. The preset number of iterations and the preset magnitude threshold can be set according to the actual situation. The validation set error did not decrease after a preset number of iterations. The preset number of iterations can be set according to the actual situation. The target loss value corresponding to the training set is less than the preset loss threshold, which can be set according to the actual situation; The total number of iterations reaches the preset maximum number of iterations, which can be set according to the actual situation.

[0042] In another optional embodiment, the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted are input into a pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted. This includes: inputting the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted into LightGBM to obtain a first prediction result; inputting the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted into PINN to obtain a second prediction result; and using dynamic weighted linear fusion of the first prediction result and the second prediction result to obtain a target prediction result, wherein the target prediction result includes the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted.

[0043] In one specific embodiment, if the data density is calculated using kernel density estimation to determine the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted, the higher the data density, the greater the weight of LightGBM, and vice versa, the greater the weight of PINN, with the sum of the weights always being 1. This mechanism ensures that the fitting accuracy of LightGBM is fully utilized in areas with sufficient data, while relying on the physical constraint capability of PINN in areas with sparse data or extrapolation.

[0044] In another optional embodiment, considering that the data sample is small when training the agricultural machinery soil entry component wear performance prediction model using only data collected from a test bench for abrasive wear performance of agricultural machinery soil entry components, the training method for the agricultural machinery soil entry component wear performance prediction model provided in this application embodiment further includes: simulating the working process of the sample agricultural machinery soil entry components to obtain simulation data; and using the simulation data to train the agricultural machinery soil entry component wear performance prediction model.

[0045] In one specific embodiment, based on existing discrete element simulation software, boundary parameters are pre-set to simulate the working process of the sample agricultural machinery's soil-entry component in abrasive particles, and simulation data is obtained.

[0046] It should be noted that the boundary parameters include the contact parameters and physical parameters between the sample agricultural machinery's soil-penetrating components and the abrasive particles. The contact parameters include the static friction coefficient, rolling friction coefficient, and coefficient of restitution between abrasive particles and between abrasive particles and the sample agricultural machinery's soil-penetrating components. The physical parameters include the density, Poisson's ratio, shear modulus, or particle size of the abrasive particles. The simulation data includes the basic information of the sample agricultural machinery's soil-penetrating components, simulation operating condition information, and the simulated wear amount and simulated wear appearance of the agricultural machinery's soil-penetrating components.

[0047] In another optional embodiment, the training method for the prediction model of wear performance of agricultural machinery soil-entry components further includes: filtering simulation data based on the basic information of the sample agricultural machinery soil-entry components, working condition information, and the actual wear amount and actual wear appearance of the sample agricultural machinery soil-entry components, to obtain the filtered simulation data. The model for predicting the wear performance of agricultural machinery soil-entry components was trained using simulation data, including: training the model for predicting the wear performance of agricultural machinery soil-entry components using screened simulation data.

[0048] It should be noted that the simulation data is filtered based on the basic information and operating condition information of the sample agricultural machinery's soil-entry components, as well as the actual wear and tear of these components. The resulting filtered simulation data refers to data that deviates significantly from the actual values, based on the basic information, operating condition information, and actual wear and tear of the sample agricultural machinery's soil-entry components. Subsequently, the boundary parameters can be adjusted based on the filtered simulation data.

[0049] In another optional embodiment, the method for predicting the wear performance of agricultural machinery soil-entry components provided in this application further includes: classifying training samples based on operating condition information, and training different types of training samples to create multiple training models for predicting the wear performance of agricultural machinery soil-entry components. This allows for obtaining the most suitable model for predicting the wear performance of agricultural machinery soil-entry components based on the current operating condition information, facilitating subsequent prediction of the wear performance of these components.

[0050] In another optional embodiment, the method for predicting the performance of agricultural machinery soil-entry components provided in this application further includes: The graphical user interface displays the training process of the agricultural machinery soil-entry component wear performance prediction model, and / or the prediction process of using the pre-trained agricultural machinery soil-entry component wear performance prediction model to predict the wear performance of the agricultural machinery soil-entry component to be predicted.

[0051] In one specific embodiment, the digital twin computer presents a three-dimensional model of the test bench, the working process of the components, the setting of environmental conditions parameters, the presentation of simulation results, and the display of predicted trends.

[0052] This application also provides an application scenario in which the above-mentioned method for predicting the wear performance of agricultural machinery soil-entry components is applied. Specifically, the method for predicting the wear performance of agricultural machinery soil-entry components provided in this embodiment can be applied in farming scenarios. The basic information and operating condition information of the agricultural machinery soil-entry component to be predicted are obtained; the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted are input into a pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted, thereby predicting the service life of the agricultural machinery soil-entry component.

[0053] Based on the same inventive concept, this application also provides an agricultural machinery soil-entry component wear performance prediction device for implementing the above-mentioned method for predicting the wear performance of agricultural machinery soil-entry components. The solution provided by this device is similar to the solution described in the above-mentioned method. Therefore, the specific limitations of one or more embodiments of the agricultural machinery soil-entry component wear performance prediction device provided below can be found in the limitations of the agricultural machinery soil-entry component wear performance prediction method above, and will not be repeated here.

[0054] In one exemplary embodiment, such as Figure 3 As shown, a device for predicting the wear performance of agricultural machinery soil-entry components is provided, comprising: The acquisition module 301 is used to acquire the basic information and working condition information of the agricultural machinery soil-entry components to be predicted; The prediction module 302 is used to input the basic information and working condition information of the agricultural machinery soil entry component to be predicted into the pre-trained agricultural machinery soil entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil entry component to be predicted.

[0055] In an optional embodiment, the apparatus further includes a module for performing the following operations: Obtain basic information about the soil-entry components of the sample agricultural machinery, operating condition information, and the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery; The wear performance prediction model of agricultural machinery soil entry components was trained using the basic component information, working condition information, actual wear amount and actual wear appearance of the sample agricultural machinery soil entry components.

[0056] In an optional embodiment, the apparatus further includes a module for performing the following operations: The working process of the soil-entry component of the sample agricultural machinery was simulated to obtain simulation data; A model for predicting the wear performance of agricultural machinery soil-entry components was trained using simulation data.

[0057] In an optional embodiment, the apparatus further includes a module for performing the following operations: Based on the basic information of the sample agricultural machinery's soil-entry components, operating condition information, and the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry components, the simulation data is filtered to obtain the filtered simulation data. A model for predicting the wear performance of agricultural machinery soil-entry components was trained using the selected simulation data.

[0058] In one alternative embodiment, the wear performance prediction model for agricultural machinery soil-penetrating components includes LightGBM and PINN, and the device further includes a module for performing the following operations: LightGBM was trained using the basic information and working condition information of the soil-entry components of sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of sample agricultural machinery. PINN was trained using the basic information and working condition information of the soil-entry components of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery.

[0059] In an optional embodiment, the apparatus further includes a module for performing the following operations: The basic information and operating condition information of the sample agricultural machinery's soil-entry components, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry components, are used for iterative training until the termination condition is met. Each iteration of training includes: Input the basic information and working condition information of the soil-entry components of the sample agricultural machinery into PINN, and output the predicted wear amount and predicted wear appearance of the soil-entry components of the sample agricultural machinery. The target loss is calculated based on the predicted wear amount and predicted wear appearance of the soil-entry parts of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry parts of the sample agricultural machinery. The parameters of PINN are adjusted based on the target loss.

[0060] In an optional embodiment, the apparatus further includes a module for performing the following operations: ; ; in, Represents the total loss function. This represents the mean square error between the predicted wear amount and the actual wear amount of the soil-entry components of the sample agricultural machinery. This represents the mean square error between the predicted wear appearance and the actual wear appearance of the soil-entry components of the sample agricultural machinery. Indicates the balance hyperparameters, V represents physical loss, P represents the actual wear amount, L represents the normal load of the agricultural machinery's soil-entry component in the abrasive grains, H represents the material hardness of the agricultural machinery's soil-entry component, and k represents the dimensionless wear coefficient.

[0061] In one alternative embodiment, the prediction module is configured to: Input the basic information and working condition information of the agricultural machinery soil-entry components to be predicted into LightGBM to obtain the first prediction result; Input the basic information and working condition information of the agricultural machinery soil-entry components to be predicted into PINN to obtain the second prediction result; The target prediction result is obtained by dynamically weighted linearly fusing the first and second prediction results. The target prediction result includes the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry parts to be predicted.

[0062] In one optional embodiment, the basic component information includes at least: the substrate material, coating material, coating process, and component type of the agricultural machinery soil-entry component; the working condition information includes at least: the torque value, rotational speed, temperature, force, depth, and abrasive particle size of the agricultural machinery soil-entry component in the abrasive particles.

[0063] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data on the wear performance prediction of agricultural machinery components. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for predicting the wear performance of agricultural machinery components.

[0064] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0065] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0066] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0067] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0068] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0069] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0070] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0071] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting the wear performance of agricultural machinery soil-entry components, characterized in that, The method for predicting the wear performance of agricultural machinery soil-entry components includes: Obtain basic component information and operating condition information of the agricultural machinery soil-entry parts to be predicted; The basic information and working condition information of the agricultural machinery soil-entry component to be predicted are input into the pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted.

2. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 1, characterized in that, The method for predicting the wear performance of agricultural machinery soil-entry components also includes: Obtain the basic information and working condition information of the soil-entry components of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery; The wear performance prediction model of the agricultural machinery soil-entry component is trained using the basic component information, working condition information, actual wear amount and actual wear appearance of the sample agricultural machinery soil-entry component.

3. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 2, characterized in that, The method for predicting the wear performance of agricultural machinery soil-entry components also includes: The working process of the soil-entry component of the sample agricultural machinery was simulated to obtain simulation data; The simulation data was used to train a model for predicting the wear performance of agricultural machinery components that enter the soil.

4. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 3, characterized in that, Before using the simulation data to train the wear performance prediction model for agricultural machinery soil-entry components, the method for predicting the wear performance of agricultural machinery soil-entry components further includes: Based on the basic information of the sample agricultural machinery's soil-entry component, its working condition information, and the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry component, the simulation data is filtered to obtain the filtered simulation data; The method of training a wear performance prediction model for agricultural machinery soil-entry components using the simulation data includes: The selected simulation data was used to train a model for predicting the wear performance of agricultural machinery components that enter the soil.

5. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 1, characterized in that, The wear performance prediction model for agricultural machinery soil-entry components includes LightGBM and PINN. The model is trained using the basic information and operating condition information of the sample agricultural machinery soil-entry components, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery soil-entry components. This includes: The LightGBM was trained using the basic information and working condition information of the sample agricultural machinery soil-entry component, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery soil-entry component; The PINN is trained using the basic component information, working condition information, actual wear amount, and actual wear appearance of the sample agricultural machinery soil-entry component.

6. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 5, characterized in that, The training of the PINN using the basic information and operating condition information of the sample agricultural machinery's soil-entry components, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry components, includes: Iterative training is performed using the basic information and operating condition information of the sample agricultural machinery's soil-entry component, as well as the actual wear amount and actual wear appearance of the sample agricultural machinery's soil-entry component, until the termination condition is met. Each iteration of training includes: Input the basic information and working condition information of the sample agricultural machinery soil-entry component into the PINN, and output the predicted wear amount and predicted wear appearance of the sample agricultural machinery soil-entry component; The target loss is calculated based on the predicted wear amount and predicted wear appearance of the soil-entry components of the sample agricultural machinery, as well as the actual wear amount and actual wear appearance of the soil-entry components of the sample agricultural machinery. The parameters of the PINN are adjusted based on the target loss.

7. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 6, characterized in that, Based on the predicted wear and appearance of the soil-entry components of the sample agricultural machinery, as well as the actual wear and appearance of the soil-entry components, the target loss is calculated using the following formula: ； ； in, Represents the total loss function. This represents the mean square error between the predicted wear amount and the actual wear amount of the soil-entry component of the sample agricultural machinery. The mean square error represents the difference between the predicted wear appearance of the soil-entry component of the sample agricultural machinery and the actual wear appearance of the soil-entry component of the sample agricultural machinery. Indicates the balance hyperparameters, V represents physical loss, P represents the actual wear amount, L represents the normal load of the agricultural machinery soil-entry component in the abrasive grains, H represents the material hardness of the agricultural machinery soil-entry component, and k represents the dimensionless wear coefficient.

8. The training method for predicting the wear performance of agricultural machinery soil-entry components according to claim 5, characterized in that, The step of inputting the basic information and operating condition information of the agricultural machinery soil-entry component to be predicted into a pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component, including: The basic information and working condition information of the agricultural machinery soil-entry component to be predicted are input into the LightGBM to obtain the first prediction result; The basic information and working condition information of the agricultural machinery soil-entry component to be predicted are input into the PINN to obtain the second prediction result; The first prediction result and the second prediction result are fused by dynamic weighted linear fusion to obtain the target prediction result, wherein the target prediction result includes the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted.

9. The method for predicting the wear performance of agricultural machinery soil-entry components according to claim 1, characterized in that, The basic information of the component includes at least the following: the base material, coating material, coating process, and component type of the agricultural machinery soil-entry component; the working condition information includes at least the following: the torque value, rotational speed, temperature, force, depth, and abrasive particle size of the agricultural machinery soil-entry component in the abrasive particles.

10. A device for predicting the wear performance of agricultural machinery soil-entry components, characterized in that, The wear performance prediction device for agricultural machinery soil-entry components includes: The acquisition module is used to acquire the basic information and operating condition information of the agricultural machinery soil-entry components to be predicted; The prediction module is used to input the basic information and working condition information of the agricultural machinery soil-entry component to be predicted into a pre-trained agricultural machinery soil-entry component wear performance prediction model to obtain the predicted wear amount and predicted wear appearance of the agricultural machinery soil-entry component to be predicted.

Images