Rigid body modeling analysis system based on agricultural transmission shaft processing

Abstract

The invention relates to the technical field of rigid body modeling analysis, and discloses an agricultural transmission shaft processing-based rigid body modeling analysis system, which is characterized in that a transmission shaft is divided into M sub-segments along the central axis based on segmentation constraints, and the structural data of each sub-segment is acquired, including the outer diameter, the inner diameter, the length and the material density; constructing a coordinate system by taking the central axis as an X axis and taking two transverse directions perpendicular to the central axis as a Y axis and a Z axis based on a central original point of the central axis of the transmission shaft; combining the X-axis rotational inertia, the Y-axis rotational inertia and the Z-axis rotational inertia to form a rotational inertia tensor of the transmission shaft; inputting the rotational inertia tensor and an external load torque sequence and a load torque sequence of the transmission shaft under the target working condition into a pre-trained dynamic neural network model to obtain a predicted angular velocity sequence and a predicted rotation angle sequence; and performing verification processing on the maximum shear stress of the transmission shaft based on the predicted angular velocity sequence and the predicted rotation angle sequence.

Description

Technical Field

[0001] The present invention relates to the technical field of rigid body modeling and analysis, and more particularly to a rigid body modeling and analysis system based on agricultural transmission shaft processing. Background Art

[0002] The core of the rigid body modeling and analysis system based on transmission shaft processing lies in the numerical integration method such as the Ronkutta method (RK4) to iteratively solve the dynamic response of angular velocity and rotation angle step by time.

[0003] However, the existing technology has the following problems: 1. The traditional Longkutta method requires iterative calculations for each time step. The amount of calculation increases linearly with the number of time steps, resulting in a long offline modeling time and difficulty in quickly responding to parameter adjustments.

[0004] 2. In online monitoring or real-time control scenarios, step-by-step integration cannot meet the high-frequency dynamic response output requirements, limiting the rapid assessment of the real-time operating status of the drive shaft.

[0005] 3. When the moment of inertia changes, the entire process needs to be recalculated, which lacks the ability to efficiently generalize dynamic parameters. This is especially inefficient in large-scale batch analysis or multi-condition simulation. Summary of the Invention

[0006] The present invention provides a rigid body modeling and analysis system based on agricultural transmission shaft processing, which solves the technical problems raised in the background technology.

[0007] The present invention provides a rigid body modeling and analysis system based on agricultural transmission shaft processing, comprising: A data acquisition module is used to divide the transmission shaft into M sub-segments along the central axis based on the segmentation constraint and obtain the structural data of each sub-segment, including: outer diameter, inner diameter, length and material density; The segmentation constraints include: the outer diameter, inner diameter and material density of each sub-segment are equal everywhere; A first processing module is configured to construct a coordinate system based on the central origin of the central axis of the transmission shaft, with the central axis as the X axis and two transverse directions perpendicular to the central axis as the Y axis and the Z axis; The second processing module is used to calculate the volume, mass and center of mass of the m-th sub-segment based on the structural data of each sub-segment, 1≤m≤M, and m is a positive integer; According to the mass and mass center of the M sub-segments, determine the mass and mass center of the transmission shaft in turn; According to the structural data, mass and center of mass of the M sub-segments and the center of mass of the transmission shaft, the X-axis moment of inertia, the Y-axis moment of inertia and the Z-axis moment of inertia are determined respectively; Combine the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia to form the moment of inertia tensor of the transmission shaft; The third processing module is used to input the moment of inertia tensor, the external load torque sequence and the load torque sequence of the transmission shaft under the target working condition into the pre-trained dynamic neural network model to obtain the predicted angular velocity sequence and the predicted rotation angle sequence; The modeling and analysis module is used to verify the maximum shear stress of the transmission shaft based on the predicted angular velocity sequence and the predicted rotation angle sequence.

[0008] Furthermore, based on the structural data of each sub-segment, the volume, mass and center of mass of the m-th sub-segment are calculated respectively, including: Determine the volume of the mth subsegment for: ,in, represents the outer diameter of the mth sub-segment, represents the inner diameter of the mth subsegment, Indicates the length of the mth sub-segment; Determine the quality of the mth subsegment for: , represents the material density of the mth subsegment; Determine the centroid of the mth subsegment for: , Indicates the coordinate of the starting point of the mth sub-segment on the X axis.

[0009] Furthermore, the mass and center of mass of the transmission shaft are determined in sequence according to the mass and center of mass of the M sub-segments, including: The sum of the masses of the M sub-segments is taken as the mass of the transmission shaft; Based on the centroids of the M sub-segments, the centroid of the drive shaft is determined by mass-weighted averaging as follows: ; in, represents the center of mass of the drive shaft, Indicates the mass of the drive shaft.

[0010] Furthermore, based on the structural data, mass, and center of mass of the M sub-segments, as well as the center of mass of the transmission shaft, the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia are determined respectively, including: Determine the X-axis moment of inertia for: ; Determine the Y-axis moment of inertia for: ; Determine the Z-axis moment of inertia and Y-axis moment of inertia equal.

[0011] Furthermore, the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia are combined to form the moment of inertia tensor of the transmission shaft, including: Initialize the moment of inertia tensor ,in, 、 and 、 、 and Both represent the product of moment of inertia; The transmission shaft is a symmetrical rigid body around the X axis, so: the product of inertia is 0; The final moment of inertia tensor is .

[0012] Furthermore, the pre-trained dynamic neural network model includes: Establish a dynamic neural network model annotation dataset as follows: Determine the external loading torque and load torque under typical working conditions, combined with the moment of inertia tensor , the angular velocity and rotation angle are obtained by the fourth-order Runge-Kutta method; The external applied torque, load torque and moment of inertia tensor Splice sample data to form a labeled dataset; The angular velocity and rotation angle obtained by the fourth-order Runge-Kutta method are used as sample labels of the labeled dataset; A dynamic neural network model is obtained by pre-training based on the labeled dataset.

[0013] Furthermore, the hidden layer of the dynamic neural network model includes: The hidden layer of the dynamic neural network model is constructed based on the PINN neural network; Among them, the physical constraints of the PINN neural network are constructed based on the rotation equation as follows: ; in, represents the first derivative of the angular velocity obtained by the fourth-order Runge-Kutta method, represents the external loading torque obtained by the fourth-order Runge-Kutta method, represents the rotation angle obtained by the fourth-order Runge-Kutta method.

[0014] Furthermore, the loss function of the dynamic neural network model is as follows: ; in, represents the loss function value of the dynamic neural network model, represents the predicted value of the dynamic neural network model, Indicates the sample label corresponding to the predicted value, and represent the first loss weight and the second loss weight respectively, Represents the square of the L2 norm.

[0015] Furthermore, based on the predicted angular velocity sequence and the predicted rotation angle sequence, a verification process is performed on the maximum shear stress of the transmission shaft, including: Obtain the maximum value of the internal and external loading torque under typical working conditions as the characteristic torque; Determine the polar moment of inertia of the drive shaft based on the structural parameters of the drive shaft , ,in, represents the average outer diameter of M sub-segments, represents the average inner diameter of M subsegments; Calculates maximum shear stress based on characteristic torque, polar moment of inertia, and mean outer diameter, including: ; in, represents the maximum shear stress of the transmission shaft, represents the characteristic torque; If the maximum shear stress is greater than or equal to the allowable shear stress of the drive shaft, the verification fails; otherwise, the verification passes.

[0016] The beneficial effects of the present invention are: by modeling the drive shaft in sections and calculating the physical parameters of each sub-segment, constructing the moment of inertia tensor, and combining it with the PINN neural network embedded with the physical constraints of the rotation equation, efficient prediction of angular velocity and angle sequence is achieved, which greatly shortens the offline modeling time and meets the online real-time dynamic response requirements. At the same time, the strength and safety of the drive shaft are ensured through maximum shear stress verification, which significantly improves the generalization ability for different working conditions and structural parameters, and provides a rigid body modeling and analysis solution with both high precision and high real-time performance for agricultural drive shaft processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a module diagram of a rigid body modeling and analysis system based on agricultural transmission shaft processing according to the present invention. DETAILED DESCRIPTION

[0018] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0019] like Figure 1 As shown, a rigid body modeling and analysis system based on agricultural transmission shaft processing includes: A data acquisition module is used to divide the transmission shaft into M sub-segments along the central axis based on the segmentation constraint and obtain the structural data of each sub-segment, including: outer diameter, inner diameter, length and material density; The segmentation constraints include: the outer diameter, inner diameter and material density of each sub-segment are equal everywhere; A first processing module is configured to construct a coordinate system based on the central origin of the central axis of the transmission shaft, with the central axis as the X axis and two transverse directions perpendicular to the central axis as the Y axis and the Z axis; The second processing module is used to calculate the volume, mass and center of mass of the m-th sub-segment based on the structural data of each sub-segment, 1≤m≤M, and m is a positive integer; According to the mass and mass center of the M sub-segments, determine the mass and mass center of the transmission shaft in turn; According to the structural data, mass and center of mass of the M sub-segments and the center of mass of the transmission shaft, the X-axis moment of inertia, the Y-axis moment of inertia and the Z-axis moment of inertia are determined respectively; Combine the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia to form the moment of inertia tensor of the transmission shaft; The third processing module is used to input the moment of inertia tensor, the external load torque sequence and the load torque sequence of the transmission shaft under the target working condition into the pre-trained dynamic neural network model to obtain the predicted angular velocity sequence and the predicted rotation angle sequence; The modeling and analysis module is used to verify the maximum shear stress of the transmission shaft based on the predicted angular velocity sequence and the predicted rotation angle sequence.

[0020] In one embodiment of the present invention, based on the structural data of each sub-segment, the volume, mass, and center of mass of the m-th sub-segment are calculated respectively, including: Determine the volume of the mth subsegment for: ,in, represents the outer diameter of the mth sub-segment, represents the inner diameter of the mth subsegment, Indicates the length of the mth sub-segment; Determine the quality of the mth subsegment for: , represents the material density of the mth subsegment; Determine the centroid of the mth subsegment for: , Indicates the coordinate of the starting point of the mth sub-segment on the X axis.

[0021] It's important to note that by calculating volume, mass, and center of mass in sections, the complex driveshaft is broken down into uniform subsegments, simplifying the calculation while ensuring accurate parameters for each subsegment. This lays the foundation for subsequent moment of inertia, total center of mass, and dynamic analysis, improving overall modeling accuracy. Accurate mass and center of mass data make calculations of moments of inertia and total mass more reliable, thereby ensuring the accuracy of analysis of driveshaft dynamic behavior (such as angular velocity and torque response). Ultimately, this improves the efficiency and reliability of driveshaft dynamic performance evaluation, ensuring its stability and safety in practical applications.

[0022] In one embodiment of the present invention, determining the mass and center of mass of the transmission shaft in sequence according to the mass and center of mass of the M sub-segments includes: The sum of the masses of the M sub-segments is taken as the mass of the transmission shaft; Based on the centroids of the M sub-segments, the centroid of the drive shaft is determined by mass-weighted averaging as follows: ; in, represents the center of mass of the drive shaft, Indicates the mass of the drive shaft.

[0023] It's important to note that the center of mass reflects the central location of an object's mass distribution. The center of mass and mass of each subsegment together determine its contribution to the overall center of mass of the drive shaft. By using mass-weighted averaging, subsegments with greater mass have a greater influence on the overall center of mass, conforming to the laws of physics and accurately locating the center of the drive shaft's mass distribution.

[0024] Total mass is a fundamental parameter describing the inertial characteristics of a drive shaft and is widely used in dynamic equations (such as the application of Newton's second law to rotational and translational motion). Accurate total mass ensures that subsequent analysis conforms to physical reality and avoids analytical deviations caused by incorrect mass parameters.

[0025] The total center of mass determined by the mass-weighted average method accurately reflects the impact of each subsegment's mass distribution on the overall center of mass. When calculating moments of inertia, the center of mass location is crucial for applying the parallel axis theorem. When analyzing drive shaft motion (such as vibration and rotation), an accurate center of mass provides a more realistic description of its dynamic behavior, improving the accuracy and reliability of dynamic analysis (such as angular velocity and torque response).

[0026] In one embodiment of the present invention, the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia are determined based on the structural data, mass, and center of mass of the M sub-segments and the center of mass of the transmission shaft, respectively, including: Determine the X-axis moment of inertia for: ; Determine the Y-axis moment of inertia for: ; Determine the Z-axis moment of inertia and Y-axis moment of inertia equal.

[0027] It should be noted that the moment of inertia is the core parameter that describes the rotational inertia of an object. It is indispensable in dynamic analysis (such as torque, angular velocity relationship, and vibration characteristics). Calculating the three-axis moment of inertia can comprehensively characterize the rotational characteristics of the drive shaft.

[0028] Through segmented calculation and accumulation, the contribution of each sub-segment to the overall rotational inertia of the drive shaft is accurately reflected, so that the rotational inertia of the X, Y, and Z axes accurately matches the actual structure and mass distribution.

[0029] The moment of inertia helps to deeply analyze the dynamic response of the drive shaft under different working conditions (such as angular velocity changes and rotational stability under torque), optimize the drive shaft design (such as reducing weight and improving strength), and avoid design defects caused by deviations in rotational characteristic analysis.

[0030] In one embodiment of the present invention, the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia are combined to form the moment of inertia tensor of the transmission shaft, including: Initialize the moment of inertia tensor ,in, 、 and 、 、 and Both represent the product of moment of inertia; The transmission shaft is a symmetrical rigid body around the X axis, so: the product of inertia is 0; The final moment of inertia tensor is .

[0031] It's important to note that the moment of inertia tensor is a core tool for describing the rotational inertia of a rigid body, comprehensively reflecting the body's rotational characteristics along different axes. Initializing the moment of inertia tensor uses a standard form, including the moment of inertia and the product of inertia, providing a universal representation of the rotational inertia of a rigid body.

[0032] As a symmetrical rigid body around the X-axis, the drive shaft's mass distribution is symmetrical about the X-axis, and its product of inertia (reflecting the coupling effect of mass distribution on different axes) is necessarily 0. By simplifying the tensor, we can more accurately match the physical properties of the drive shaft and avoid redundant calculations.

[0033] By eliminating the product of inertia, the physical characteristics of the drive shaft's symmetry around the X-axis are accurately reflected, avoiding model errors caused by retaining redundant products of inertia. This improves the accuracy of dynamic analysis (such as angular velocity and torque response) and makes the results more consistent with actual working conditions.

[0034] In one embodiment of the present invention, the pre-trained dynamic neural network model includes: Establish a dynamic neural network model annotation dataset as follows: Determine the external loading torque and load torque under typical working conditions, combined with the moment of inertia tensor , the angular velocity and rotation angle are obtained by the fourth-order Runge-Kutta method; The external applied torque, load torque and moment of inertia tensor Splice sample data to form a labeled dataset; The angular velocity and rotation angle obtained by the fourth-order Runge-Kutta method are used as sample labels of the labeled dataset; A dynamic neural network model is obtained by pre-training based on the labeled dataset.

[0035] It should be noted that the fourth-order Runge-Kutta method is a high-precision numerical method for solving dynamic differential equations. By combining the external applied torque T, load torque, and moment of inertia tensor under typical operating conditions, this method can accurately calculate angular velocity and angle, providing sample labels for annotating the dataset.

[0036] The external loading torque, load torque and moment of inertia tensor are spliced ​​into sample data, corresponding to the angular velocity and angle sample labels, which clarifies the input-output relationship of the neural network model, enabling the model to learn the mapping rules between these physical quantities, thereby having the ability to predict the dynamic response of the drive shaft.

[0037] The pre-trained dynamic neural network model can quickly predict the angular velocity and angle based on the input torque and moment of inertia tensor, avoiding the time-consuming iterative calculation of each time step using the fourth-order Runge-Kutta method, greatly improving the efficiency of dynamic analysis.

[0038] Neural networks have powerful nonlinear fitting capabilities and can capture the complex nonlinear relationship between external loading torque, load torque, moment of inertia, angular velocity, and angle. Compared with traditional analytical or numerical methods, they are more flexible and can adapt to various working conditions and changes in drive shaft structure.

[0039] In one embodiment of the present invention, the hidden layer of the dynamic neural network model includes: The hidden layer of the dynamic neural network model is constructed based on the PINN neural network; Among them, the physical constraints of the PINN neural network are constructed based on the rotation equation as follows: ; in, represents the first derivative of the angular velocity obtained by the fourth-order Runge-Kutta method, represents the external loading torque obtained by the fourth-order Runge-Kutta method, represents the rotation angle obtained by the fourth-order Runge-Kutta method.

[0040] The equation of rotation is the core physical law that describes the dynamic behavior of a driveshaft. Building a hidden layer based on a physical information neural network (PINN) and embedding this equation as a physical constraint ensures that the model's predictions conform to physical laws, avoiding the potential physical irrationalities (such as incorrect predictions due to noisy or insufficient data) that can arise when traditional neural networks rely solely on data fitting.

[0041] By simultaneously fitting data (such as angular velocity and torque generated by the fourth-order Runge-Kutta method) and satisfying physical constraints (rotation equations), the model can more accurately capture the dynamic characteristics of the drive shaft, reduce prediction errors, and make predictions such as angular velocity and torque response closer to actual physical behavior.

[0042] In one embodiment of the present invention, the loss function of the dynamic neural network model is as follows: ; in, represents the loss function value of the dynamic neural network model, represents the predicted value of the dynamic neural network model, Indicates the sample label corresponding to the predicted value, and represent the first loss weight and the second loss weight respectively, Represents the square of the L2 norm.

[0043] It should be noted that the data fitting term : By minimizing the model prediction value With sample labels The model learns the mapping relationship between external loading torque, load torque, moment of inertia tensor and dynamic response (angular velocity, angle of rotation) by using the information of labeled data to drive model training.

[0044] Physical constraints Based on the rotational equations, the model's predictions are forced to conform to physical laws. Even when the data is incomplete or noisy, physical constraints can guide the model's output to conform to dynamic principles, avoiding the physical irrationality caused by purely data-driven predictions.

[0045] In one embodiment of the present invention, the maximum shear stress of the transmission shaft is verified based on the predicted angular velocity sequence and the predicted rotation angle sequence, including: Obtain the maximum value of the internal and external loading torque under typical working conditions as the characteristic torque; Determine the polar moment of inertia of the drive shaft based on the structural parameters of the drive shaft , ,in, represents the average outer diameter of M sub-segments, represents the average inner diameter of M subsegments; Calculates maximum shear stress based on characteristic torque, polar moment of inertia, and mean outer diameter, including: ; in, represents the maximum shear stress of the transmission shaft, represents the characteristic torque; If the maximum shear stress is greater than or equal to the allowable shear stress of the drive shaft, the verification fails; otherwise, the verification passes.

[0046] It should be noted that the maximum value of the external loading torque in typical working conditions represents the most severe torsional load that the drive shaft bears during operation. Using this as the characteristic torque can be used to specifically verify the strength of the drive shaft under extreme conditions and ensure its safety in actual use.

[0047] The polar moment of inertia is a key parameter that describes the ability of the drive shaft section to resist torsional motion. It is calculated based on the average outer diameter and average inner diameter and reflects the mechanical properties of the overall structure of the drive shaft against torsion.

[0048] According to the torsional shear stress formula in material mechanics, maximum shear stress occurs on the outer surface of the shaft. This formula quantitatively estimates the maximum shear stress of the drive shaft under extreme torque using characteristic torque, polar moment of inertia, and average outer diameter, in accordance with the principles of physical mechanics. The allowable shear stress represents the upper limit of stress for safe operation of the material. Comparing this with the allowable shear stress directly determines whether the drive shaft will experience strength failure under extreme operating conditions, ensuring operational reliability.

[0049] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A rigid body modeling and analysis system based on agricultural transmission shaft processing, characterized in that: include: A data acquisition module is used to divide the transmission shaft into M sub-segments along the central axis based on the segmentation constraint and obtain the structural data of each sub-segment, including: outer diameter, inner diameter, length and material density; The segmentation constraints include: the outer diameter, inner diameter and material density of each sub-segment are equal everywhere; A first processing module is configured to construct a coordinate system based on the central origin of the central axis of the transmission shaft, with the central axis as the X axis and two transverse directions perpendicular to the central axis as the Y axis and the Z axis; The second processing module is used to calculate the volume, mass and center of mass of the m-th sub-segment based on the structural data of each sub-segment, 1≤m≤M, and m is a positive integer; According to the mass and mass center of the M sub-segments, determine the mass and mass center of the transmission shaft in turn; According to the structural data, mass and center of mass of the M sub-segments and the center of mass of the transmission shaft, the X-axis moment of inertia, the Y-axis moment of inertia and the Z-axis moment of inertia are determined respectively; Combine the X-axis moment of inertia, the Y-axis moment of inertia, and the Z-axis moment of inertia to form the moment of inertia tensor of the transmission shaft; The third processing module is used to input the moment of inertia tensor, the external load torque sequence and the load torque sequence of the transmission shaft under the target working condition into the pre-trained dynamic neural network model to obtain the predicted angular velocity sequence and the predicted rotation angle sequence; The modeling and analysis module is used to verify the maximum shear stress of the transmission shaft based on the predicted angular velocity sequence and the predicted rotation angle sequence.

2. A rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 1, characterized in that: Based on the structural data of each sub-segment, the volume, mass and center of mass of the m-th sub-segment are calculated respectively, including: Determine the volume of the mth subsegment for: ,in, represents the outer diameter of the mth sub-segment, represents the inner diameter of the mth subsegment, Indicates the length of the mth sub-segment; Determine the quality of the mth subsegment for: , represents the material density of the mth subsegment; Determine the centroid of the mth subsegment for: , Indicates the coordinate of the starting point of the mth sub-segment on the X axis.

3. A rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 2, characterized in that: According to the mass and center of mass of the M sub-segments, the mass and center of mass of the transmission shaft are determined in sequence, including: The sum of the masses of the M sub-segments is taken as the mass of the transmission shaft; Based on the centroids of the M sub-segments, the centroid of the drive shaft is determined by mass-weighted averaging as follows: ； in, represents the center of mass of the drive shaft, Indicates the mass of the drive shaft.

4. The rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 3 is characterized in that: Based on the structural data, mass, and center of mass of the M sub-segments, as well as the center of mass of the transmission shaft, the X-axis moment of inertia, Y-axis moment of inertia, and Z-axis moment of inertia are determined respectively, including: Determine the X-axis moment of inertia for: ; Determine the Y-axis moment of inertia for: ; Determine the Z-axis moment of inertia and Y-axis moment of inertia equal.

5. The rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 4 is characterized in that: The X-axis moment of inertia, Y-axis moment of inertia, and Z-axis moment of inertia are combined to form the moment of inertia tensor of the transmission shaft, including: Initialize the moment of inertia tensor ,in, 、 and 、 、 and Both represent the product of moment of inertia; The transmission shaft is a symmetrical rigid body around the X axis, so: the product of inertia is 0; The final moment of inertia tensor is .

6. A rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 4, characterized in that: Pre-trained dynamic neural network models, including: Establish a dynamic neural network model annotation dataset as follows: Determine the external loading torque and load torque under typical working conditions, combined with the moment of inertia tensor , the angular velocity and rotation angle are obtained by the fourth-order Runge-Kutta method; The external applied torque, load torque and moment of inertia tensor Splice sample data to form a labeled dataset; The angular velocity and rotation angle obtained by the fourth-order Runge-Kutta method are used as sample labels of the labeled dataset; A dynamic neural network model is obtained by pre-training based on the labeled dataset.

7. The rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 5 is characterized in that: The hidden layer of the dynamic neural network model includes: The hidden layer of the dynamic neural network model is constructed based on the PINN neural network; Among them, the physical constraints of the PINN neural network are constructed based on the rotation equation as follows: ； in, represents the first derivative of the angular velocity obtained by the fourth-order Runge-Kutta method, represents the external loading torque obtained by the fourth-order Runge-Kutta method, represents the rotation angle obtained by the fourth-order Runge-Kutta method.

8. The rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 6 is characterized in that: The loss function of the dynamic neural network model is as follows: ； in, represents the loss function value of the dynamic neural network model, represents the predicted value of the dynamic neural network model, Indicates the sample label corresponding to the predicted value, and represent the first loss weight and the second loss weight respectively, Represents the square of the L2 norm.

9. The rigid body modeling and analysis system based on agricultural transmission shaft processing according to claim 7, characterized in that: The maximum shear stress of the transmission shaft is verified based on the predicted angular velocity sequence and the predicted rotation angle sequence, including: Obtain the maximum value of the internal and external loading torque under typical working conditions as the characteristic torque; Determine the polar moment of inertia of the drive shaft based on the structural parameters of the drive shaft , ,in, represents the average outer diameter of M sub-segments, represents the average inner diameter of M subsegments; Calculates maximum shear stress based on characteristic torque, polar moment of inertia, and mean outer diameter, including: ； in, represents the maximum shear stress of the transmission shaft, represents the characteristic torque; If the maximum shear stress is greater than or equal to the allowable shear stress of the drive shaft, the verification fails; otherwise, the verification passes.

Images