Service-performance-driven inverse design method for ideal prestress distribution on assembly interface, and related apparatus

Abstract

The present invention belongs to the technical field of mechanical designs. Disclosed are a service-performance-driven inverse design method for ideal prestress distribution on an assembly interface, and a related apparatus. The method comprises: performing assembly interface topography-stress feature mapping on a mating surface, in order to obtain contact pressure values and relative coordinates of each node; on the basis of the contact pressure values and static contact stress, performing inversion of static ideal contact stress distribution, in order to obtain static ideal contact stress distribution; and on the basis of the static ideal contact stress distribution, predicting a topography layout of an assembly interface that is driven by static contact stress distribution. In the present invention, a topography layout of an assembly interface that is driven by static contact stress distribution can be obtained, and machining can be performed on the assembly interface on the basis of the topography layout of the assembly interface, so that the uniformity of the connection performance of the assembly interface in a dynamic service state can be effectively improved, thereby achieving the aim of improving the dynamic connection performance of parts of advanced equipment such as an aero-engine.

Description

An inversion design method and related apparatus for ideal prestress distribution of assembly interface based on service performance traction Technical Field

[0001] This invention belongs to the field of mechanical design technology and relates to a method for improving the performance of the assembly interface of mechanical equipment under service conditions. Specifically, it relates to an inversion design method and related device for the ideal prestress distribution of the assembly interface driven by service performance. Background Technology

[0002] Mechanical connection structures are used to connect mechanical components. The assembly interface between these structures is a crucial carrier for supporting and ensuring the intended function of mechanical equipment. The contact stress distribution at the assembly interface is a key factor affecting contact mechanical properties and even the overall service performance of the machine. For example, uneven contact stress distribution and small effective contact area at the assembly interface are major factors contributing to high-cycle fatigue and excessive vibration in components of high-end equipment such as aero-engines. Compared to optimizing the assembly process of the assembly interface, focusing on the design of the assembly interface can directly affect the assembly connection performance. Currently, surface morphology design of assembly interfaces is increasingly being applied. For instance, world-leading aero-engine manufacturers like Rolls Royce in the UK have significantly improved the overall performance of aero-engines and effectively enhanced their competitiveness in the international market by designing radial textured "micro-spline" morphology on the assembly interface between the high-pressure turbine disk and the rear journal. By simulating the load conditions of mechanical equipment during service, deep learning technology can be used to inversely design the optimal assembly interface morphology layout. With current high-precision machining as a design guarantee, the machining results can help improve the dynamic performance of mechanical equipment, including reducing vibration, increasing reliability, and extending service life. It is evident that starting from the mechanical essence of assembly connection performance and aiming at controlling physical quantity performance indicators such as assembly contact stress, the technical approach is a research direction with great development potential.

[0003] However, due to the high pressure, high and low temperature, strong oxidation, and strong vibration and impact of the assembly interface of large equipment such as aerospace, the stress distribution of the assembly components after assembly is completed will be "redesigned". Therefore, the uniformity of contact stress distribution of the assembly interface pursued by the existing technology in the assembly completion stage is not the direct target form of contact stress distribution controlled by the active design of the morphology layout of the assembly interface. Summary of the Invention

[0004] This invention addresses the problem in existing technologies that cannot determine the stress distribution relationship between the service stage and the assembly completion stage of the assembly interface, and cannot design the assembly interface based on the service conditions of the assembly interface. The purpose is to propose a service performance-driven inversion design method and related device for the ideal prestress distribution of the assembly interface, which inversely evolves the static ideal contact stress distribution of the assembly interface from the uniform dynamic service state contact stress distribution, thus facilitating the processing of the assembly interface.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] An inversion design method for ideal prestress distribution at assembly interfaces driven by service performance includes the following steps:

[0007] The morphology-stress feature mapping of the assembly interface is performed on the mating surface to obtain the contact pressure value and the relative coordinates of each node;

[0008] Based on the contact pressure value and static contact stress, the static ideal contact stress distribution is inverted to obtain the static ideal contact stress distribution;

[0009] Based on the static ideal contact stress distribution, predict the morphological layout driven by the static contact stress distribution of the assembly interface.

[0010] A further improvement of this invention lies in performing a feature mapping of the morphology-stress of the mating interface on the mating surface to obtain the contact pressure value and the relative coordinates of each node, including the following steps:

[0011] (1.1) Divide the assembly interface into finite element meshes, construct the finite element model for contact analysis of the assembly interface, and apply boundary constraints and load conditions;

[0012] (1.2) Perform transient finite element contact analysis, select a period of n sampling, and calculate the contact pressure value and average contact pressure of the nodes at the sampling point on the assembly interface;

[0013] (1.3) Calculate the variance of contact stress based on the average contact pressure;

[0014] (1.4) Determine the contact stress variance δ. If δ≤ε, where ε is the threshold, or if the number of optimization iteration steps k satisfies k≤N, where N is the total number of nodes in the assembly interface finite element calculation process, then the iteration terminates and the relative coordinates of each node are obtained.

[0015] A further improvement of the present invention is that if δ>ε or k>N, then step (1.5) is executed;

[0016] (1.5) Set up the area for active adjustment of morphology, use the relative coordinates of nodes as the optimization design variables in theoretical analysis, set the magnification factor and reduction factor, and in the new optimization iteration step, the optimization method and update method of the active adjustment of the relative coordinate values ​​of nodes are as follows: after the update, delete the current external load conditions, calculate the contact stress of each node under the current morphology layout under the condition of no external load based on the contact pressure value of the reference contact node, and determine the relative coordinate values ​​of nodes based on the contact pressure value of the reference contact node and the contact stress of the node.

[0017] (1.6) Update the finite element model of the assembly interface contact analysis described in step (1.1) based on the relative coordinate values ​​of the nodes obtained in step (1.5).

[0018] A further improvement of the present invention is that the average contact pressure Calculated using the following formula:

[0019] In the formula: N is the total number of nodes in the finite element calculation process of the assembly interface. This represents the contact pressure value of the i-th node during the j-th sampling on the assembly interface.

[0020] The contact stress variance δ is calculated using the following formula:

[0021] A further improvement of this invention lies in the fact that, based on the contact pressure value and the static contact stress, a static ideal contact stress distribution is inverted to obtain the static ideal contact stress distribution, including the following steps:

[0022] (2.1) The contact pressure value of the i-th node at the j-th sampling time is taken as the dynamic service state contact stress and the static ideal contact stress σ. i The dataset was preprocessed, and the preprocessed data was divided into training and validation sets.

[0023] (2.2) Using the U-Net model as the basic network architecture, a Multi-Bp-U-Net model is established using a BP neural network;

[0024] (2.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data to obtain the trained Multi-Bp-U-Net model;

[0025] (2.4) The contact stress distribution under dynamic service conditions under uniform conditions is inverted using the trained Multi-Bp-U-Net model to obtain the static ideal contact stress distribution under uniform conditions.

[0026] A further improvement of this invention is that the Multi-Bp-U-Net model includes:

[0027] (2.2.1) Feature extraction module, including input convolutional layer and BpUpDown layer, wherein the input convolutional layer is used to extract multidimensional input data to adapt to the input dimension; the BpUpDown layer is used to extract sequence data features according to the multidimensional input data to adapt to the input dimension;

[0028] (2.2.2) Shrinking path module, including multiple BpUpDown layers, used to downsample the sequence feature information;

[0029] (2.2.3) Extended path module, including multiple BpUpDown layers, for upsampling processing;

[0030] (2.2.4) Jump connection module, used to pass low-level features of the lower layer to the upper layer by copying and splicing.

[0031] A further improvement of this invention lies in predicting the morphological layout driven by the static contact stress distribution of the assembly interface based on the static ideal contact stress distribution, including the following steps:

[0032] (3.1) The static ideal contact stress and relative coordinate values ​​are used as the dataset for preprocessing, and the preprocessed data is divided into training set and validation set;

[0033] (3.2) Using the U-Net model as the basic network architecture, a Multi-Bp-U-Net model is established by using the fully connected layer of the BP neural network;

[0034] (3.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data described in step (3.1) to obtain the trained Multi-Bp-U-Net model;

[0035] (3.4) Input the static ideal contact stress distribution into the trained Multi-Bp-U-Net model to predict the topography layout and obtain the topography layout under the static ideal contact stress distribution state.

[0036] An inversion design system for ideal prestress distribution at assembly interfaces driven by service performance, comprising:

[0037] The feature mapping module is used to perform feature mapping of the assembly interface morphology-stress on the mating surface to obtain the contact pressure value and the relative coordinates of each node;

[0038] The static ideal contact stress distribution inversion module is used to invert the static ideal contact stress distribution based on the contact pressure value and the static contact stress, and obtain the static ideal contact stress distribution.

[0039] The topography prediction module is used to predict the topography of the assembly interface driven by the static contact stress distribution based on the static ideal contact stress distribution.

[0040] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the inversion design method for the ideal prestress distribution of the assembly interface driven by service performance.

[0041] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the inversion design method for ideal prestress distribution of assembly interface driven by service performance.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] In this invention, the feature mapping step of the assembly interface morphology-stress is used to actively design and regulate the contact stress state of the assembly interface. This allows for the achievement of a local optimum solution for the contact stress distribution under dynamic service conditions, and outputs a sequence dataset containing rich feature information between morphology and stress. This invention derives the static ideal contact stress distribution of the assembly interface from the homogenized contact stress distribution under dynamic service conditions, and proposes a processing guidance scheme driven by the assembly interface morphology layout, thereby improving the dynamic connection performance of mechanical equipment components under service conditions. This invention can obtain the assembly interface morphology layout, and processing on the assembly interface according to this layout can effectively improve the uniformity of the connection performance of the assembly interface under dynamic service conditions, achieving the goal of improving the dynamic connection performance of high-end equipment components such as aero-engines.

[0044] Furthermore, deep learning technology is used to build a Multi-Bp-U-Net model to explore the intrinsic relationship between the dynamic service state contact stress and the static ideal contact stress output by the finite element software. This can achieve the global optimal solution of the contact stress distribution under dynamic service state and predict the morphological layout under this solution state, providing corresponding processing guidance. Moreover, the training and prediction efficiency of deep learning technology is much higher than that of traditional finite element calculation. Attached Figure Description

[0045] Figure 1 is a schematic diagram of a bolt connection;

[0046] Figure 2 is a two-dimensional cross-sectional view of the finite element model of the bolt structure;

[0047] Figure 3 is a diagram showing the average value of the contact stress distribution under dynamic service conditions at the assembly interface.

[0048] Figure 4 is a structural diagram of the Multi-Bp-U-Net model;

[0049] Figure 5 is a distribution diagram of static ideal contact stress;

[0050] Figure 6 is a comparison of the predicted morphology layout driven by the static contact stress distribution at the assembly interface with the initial morphology layout.

[0051] Figure 7 is a comparison of the average values ​​of the contact stress distribution under dynamic service conditions before and after the inversion.

[0052] Figure 8 is a flowchart of the inversion design method for the ideal prestress distribution of the assembly interface driven by service performance.

[0053] Figure 9 is a schematic diagram of the inversion design system for the ideal prestress distribution of the assembly interface driven by service performance.

[0054] Wherein, 1 is the upper assembly, 2 is the lower assembly, 3 is the bolt and nut, 4 is the upper assembly interface, 5 is the design domain, and 6 is the lower assembly interface. Detailed Implementation

[0055] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0056] Starting from the pursuit of uniform contact stress distribution at the assembly interface during the dynamic service phase of the connection structure, this invention "inversely" derives the contact stress distribution pattern at the assembly completion stage of the connection structure (referred to as the "ideal prestress distribution"), which is the direct target form of the contact stress distribution state controlled by the proactive design of the assembly interface. Therefore, this invention inversely evolves the static ideal contact stress distribution of the assembly interface from the uniform contact stress distribution under dynamic service conditions, and proposes a method for predicting the morphology layout using a morphology prediction model driven by the static contact stress distribution of the assembly interface. Here, the contact stress under dynamic service conditions is the contact stress affected by both external loads and assembly loads. The static ideal contact stress distribution is the static contact stress distribution reflecting the most uniform state of contact stress distribution under dynamic service conditions. The static contact stress is the contact stress affected only by assembly loads.

[0057] Referring to Figure 8, the present invention provides a service performance-driven inversion design method for the ideal prestress distribution of an assembly interface, realizing the morphological layout of the assembly interface. This method includes a feature mapping of the assembly interface morphology-stress, inversion of the static ideal contact stress distribution, and morphology prediction steps driven by the static contact stress distribution of the assembly interface, as detailed below:

[0058] 1. Perform feature mapping of the assembly interface morphology and stress on the mating surface to obtain the contact pressure value and the relative coordinates of each node; the steps include:

[0059] (1.1) Divide the assembly interface into finite element meshes, construct the finite element model for contact analysis of the assembly interface, and apply boundary constraints and load conditions;

[0060] (1.2) Perform transient finite element contact analysis, select a period of n samplings, calculate and output the contact pressure value of the i-th node at the j-th sampling on the assembly interface. With average contact pressure

[0061] In the formula: N is the total number of nodes in the finite element calculation process of the assembly interface.

[0062] (1.3) Based on average contact pressure By calculating the contact stress variance δ, which characterizes the uniformity of contact stress at the assembly interface, the optimization design objective is minδ.

[0063] (1.4) Determine the contact pressure value σ of the reference contact node. ref Generally, the point where the contact pressure value is 0 is taken at the assembly completion stage;

[0064] (1.5) Determine the connection performance uniformity optimization design objective function δ. If δ≤ε is satisfied, or the number of optimization iteration steps k is satisfied k≤N, then the iteration terminates, outputs and saves the relative coordinates of each node; if neither of the above two conditions is satisfied, then execute step (1.6).

[0065] (1.6) Set the region for actively adjusting the morphology, i.e., design domain 5. Use the relative coordinates of the nodes as the optimization design variables in theoretical analysis. Set the magnification factor C1 and the reduction factor C2. In the new optimization iteration step (k+1), the optimization and update methods of the actively adjusted relative coordinate values ​​of the nodes are as follows. After updating, delete the current external load conditions and adjust the values ​​according to the contact pressure value σ of the reference contact node. ref Calculate the contact stress σ at each node. i Based on the contact pressure value σ of the reference contact node ref Contact stress σ at the node i Determine the relative coordinates of the nodes, σ i , The output is saved as sequence data for later retrieval.

[0066] In the formula: δ represents the relative coordinates of the i-th contact node in the k-th iteration; k Let K be the variance of the contact stress in the k-th iteration; op For variable coefficient parameters; σ ref The contact pressure value is used as a reference for the contact node;

[0067] (1.7) Based on the relative coordinate values ​​of the nodes obtained in step (1.6), update the finite element model of the assembly interface described in step (1) and perform a new finite element contact analysis.

[0068] II. Based on the contact pressure value and static contact stress, perform the static ideal contact stress distribution inversion to obtain the static ideal contact stress distribution. The steps include:

[0069] (2.1) The contact pressure value of the i-th node in the saved sequence data. As the contact stress under dynamic service conditions and the static ideal contact stress σ i The dataset was preprocessed and then divided into training and validation sets.

[0070] (2.2) Using the U-Net model as the basic network architecture, the fully connected layers of the BP neural network are used to replace it to establish the Multi-Bp-U-Net model.

[0071] (2.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data described in step (2.1) to obtain the trained Multi-Bp-U-Net model.

[0072] (2.4) The trained Multi-Bp-U-Net model is used to invert the contact stress distribution under dynamic service conditions in a uniform state. The resulting static contact stress distribution is called the static ideal contact stress distribution under this state.

[0073] The Multi-Bp-U-Net model includes:

[0074] (2.2.1) Feature extraction module, which includes an input convolutional layer and a BpUpDown layer. The input convolutional layer is used to extract multidimensional data from the input to adapt to the input dimension. The BpUpDown layer uses fully connected layers instead of one-dimensional convolution to extract features from sequence data.

[0075] (2.2.2) Shrinking path module, which includes multiple BpUpDown layers to simulate convolution and pooling operations, used to downsample the feature information obtained by the feature extraction module;

[0076] (2.2.3) Extended path module, which includes multiple BpUpDown layers to simulate deconvolution layer operations for upsampling processing;

[0077] (2.2.4) Jump connection module, which transmits low-level features of the lower layer to the upper layer by copying and splicing.

[0078] III. Based on the static ideal contact stress distribution, predict the morphological layout driven by the static contact stress distribution of the assembly interface, including the following steps:

[0079] (3.1) The static ideal contact stress σ described in the saved sequence data i and relative coordinate values The dataset was preprocessed and then divided into training and validation sets.

[0080] (3.2) Using the U-Net model as the basic network architecture, replace it with the fully connected layer of the BP neural network to establish the Multi-Bp-U-Net model.

[0081] (3.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data described in step (3.1) to obtain the trained Multi-Bp-U-Net model.

[0082] (3.4) Using the trained Multi-Bp-U-Net model, the static ideal contact stress distribution is input to predict the morphological layout under this state, and the morphological layout under the static ideal contact stress distribution state is obtained.

[0083] The aforementioned feature mapping step of assembly interface morphology-stress can both homogenize the distribution of contact pressure on the mating surface to achieve optimization, and also increase the contact pressure variance to expand the dataset and improve the model's generalization ability.

[0084] The static ideal contact stress distribution inversion model includes a BpUpDown module. The BpUpDown module can replace one-dimensional convolution operations, providing a more comprehensive understanding of the global variation trend of the assembly interface morphology and stress, and is more suitable for the output structure of the finite element method of this invention. The BpUpDown module uses a multi-path parallel approach to calculate the loss function, which has a greater probability of ensuring that at least one path can perform feature mapping in a more efficient way, thereby improving the generalization ability of the overall model.

[0085] Referring to Figure 1, the topography layout design can be selected as upper assembly interface 1, lower assembly interface 2, or designed simultaneously on upper assembly interface 4 and lower assembly interface 6.

[0086] Taking the most common bolt and nut assembly structure in the engineering field (including upper assembly 1 and lower assembly 2) and a single vibration load as an example, the beneficial effects of the present invention are further illustrated.

[0087] The two-dimensional cross-sectional view of the finite element model of the bolt structure is shown in Figure 2.

[0088] The preload applied to the structure is 24 kN, and the tangential displacement load on the upper assembly interface is u1 = 2sin(2πt), while the lower assembly interface is fixed. The average value of the contact stress distribution under dynamic service conditions at the assembly interface is shown in Figure 3.

[0089] The construction of the Multi-Bp-U-Net model is shown in Figure 4. The average value of the dynamic service contact stress under uniform conditions is input into the Multi-Bp-U-Net model, and the corresponding distribution of static ideal contact stress is shown in Figure 5.

[0090] Figure 6 shows a comparison between the topography layout output by the topography prediction model driven by the static contact stress distribution of the assembly interface and the initial topography layout.

[0091] Figure 7 shows a comparison of the average values ​​of the contact stress distribution under dynamic service conditions before and after the inversion. It can be seen that the inversion design method for the ideal prestress distribution of the assembly interface driven by service performance effectively achieves the inversion of the static ideal contact stress distribution of the assembly interface under a homogenized dynamic service condition contact stress distribution, and proposes a processing guidance scheme driven by the morphology layout of the assembly interface. This invention improves the uniformity of contact stress under dynamic service conditions by 66.11%, and reduces the contact stress variance from 81.089 MPa. 2 It dropped to 27.483 MPa 2 The peak and range of pressure at the assembly interface were significantly reduced, and the average maximum contact stress decreased from 76.366 MPa to 23.987 MPa.

[0092] Referring to Figure 9, another embodiment of the present invention provides an inversion design system for ideal prestress distribution at the assembly interface driven by service performance, comprising:

[0093] The feature mapping module is used to perform feature mapping of the assembly interface morphology-stress on the mating surface to obtain the contact pressure value and the relative coordinates of each node;

[0094] The static ideal contact stress distribution inversion module is used to invert the static ideal contact stress distribution based on the contact pressure value and the static contact stress, and obtain the static ideal contact stress distribution.

[0095] The topography prediction module is used to predict the topography of the assembly interface driven by the static contact stress distribution based on the static ideal contact stress distribution.

[0096] Another embodiment of the present invention provides an electronic 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 inversion design method for ideal prestress distribution of assembly interface driven by service performance.

[0097] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program, characterized in that: when the computer program is executed by a processor, it implements the inversion design method for the ideal prestress distribution of the assembly interface driven by service performance.

[0098] This invention derives the static ideal contact stress distribution of the assembly interface from the uniform dynamic service state contact stress distribution, and proposes a processing guidance scheme driven by the morphology layout of the assembly interface, thereby improving the dynamic connection performance of mechanical equipment components under service conditions.

[0099] By employing deep learning technology, a Multi-Bp-U-Net model is constructed to explore the intrinsic relationship between the dynamic service state contact stress and the static ideal contact stress output by finite element software. This aims to achieve the global optimal solution for the contact stress distribution under dynamic service state and predict the morphological layout under this solution state, providing corresponding processing guidance. The prediction efficiency is far higher than that of traditional finite element calculation.

[0100] The above description is only of the preferred embodiment of the present invention and should not be construed as limiting the scope of the claims. The present invention is not limited to the above embodiments, and variations in its specific structure are permitted. All variations made within the scope of the independent claims of the present invention are also within the scope of protection of the present invention.

[0101] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

Claims

1. A method for inverting the ideal prestress distribution of an assembly interface based on service performance, characterized in that, Includes the following steps: The morphology-stress feature mapping of the assembly interface is performed on the mating surface to obtain the contact pressure value and the relative coordinates of each node; Based on the contact pressure value and static contact stress, the static ideal contact stress distribution is inverted to obtain the static ideal contact stress distribution; Based on the static ideal contact stress distribution, predict the morphological layout driven by the static contact stress distribution of the assembly interface.

2. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 1, characterized in that, Perform feature mapping of the morphology-stress of the mating interface on the mating surface to obtain the contact pressure value and the relative coordinates of each node, including the following steps: (1.1) Divide the assembly interface into finite element meshes, construct the finite element model for contact analysis of the assembly interface, and apply boundary constraints and load conditions; (1.2) Perform transient finite element contact analysis, select a period of n sampling, and calculate the contact pressure value and average contact pressure of the nodes at the sampling point on the assembly interface; (1.3) Calculate the variance of contact stress based on the average contact pressure; (1.4) Determine the contact stress variance δ. If δ≤ε, where ε is the threshold, or if the number of optimization iteration steps k≤N, where N is the total number of nodes in the finite element calculation process of the assembly interface, then the iteration terminates and the relative coordinates of each node are obtained.

3. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 2, characterized in that, If δ > ε or k > N, then proceed to step (1.5); (1.5) Set up the area for actively adjusting the morphology, use the relative coordinates of the nodes as the optimization design variables in the theoretical analysis, set the magnification factor and reduction factor, and in the new optimization iteration step, the optimization method and update method of the actively adjusted relative coordinate values ​​of the nodes are as follows: after the update, delete the current external load conditions, calculate the contact stress of each node under the current morphology layout under the condition of no external load based on the contact pressure value of the reference contact node, and determine the relative coordinate values ​​of the nodes based on the contact pressure value of the reference contact node and the contact stress of the nodes. (1.6) Update the finite element model of the assembly interface contact analysis described in step (1.1) based on the relative coordinate values ​​of the nodes obtained in step (1.5).

4. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 1, characterized in that, Average contact pressure Calculated using the following formula: In the formula: N is the total number of nodes in the finite element calculation process of the assembly interface. This represents the contact pressure value of the i-th node during the j-th sampling on the assembly interface. The contact stress variance δ is calculated using the following formula:

5. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 1, characterized in that, Based on the contact pressure value and static contact stress, the static ideal contact stress distribution is inverted to obtain the static ideal contact stress distribution, including the following steps: (2.1) The contact pressure value of the i-th node at the j-th sampling time is taken as the dynamic service state contact stress and the static ideal contact stress σ. i The dataset was preprocessed, and the preprocessed data was divided into training and validation sets. (2.2) Using the U-Net model as the basic network architecture, a Multi-Bp-U-Net model is established using a BP neural network; (2.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data to obtain the trained Multi-Bp-U-Net model; (2.4) The contact stress distribution under dynamic service conditions under uniform conditions is inverted using the trained Multi-Bp-U-Net model to obtain the static ideal contact stress distribution under uniform conditions.

6. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 5, characterized in that, The Multi-Bp-U-Net model includes: (2.2.1) Feature extraction module, including input convolutional layer and BpUpDown layer, wherein the input convolutional layer is used to extract multidimensional input data to adapt to the input dimension; the BpUpDown layer is used to extract sequence data features according to the multidimensional input data to adapt to the input dimension; (2.2.2) Shrinking path module, including multiple BpUpDown layers, used to downsample the sequence feature information; (2.2.3) Extended path module, including multiple BpUpDown layers, for upsampling processing; (2.2.4) Jump connection module, used to pass low-level features of the lower layer to the upper layer by copying and splicing.

7. The inversion design method for ideal prestress distribution of assembly interface based on service performance traction according to claim 1, characterized in that, Based on the static ideal contact stress distribution, the topographic layout driven by the static contact stress distribution of the assembly interface is predicted, including the following steps: (3.1) The static ideal contact stress and relative coordinate values ​​are used as the dataset for preprocessing, and the preprocessed data is divided into training set and validation set; (3.2) Using the U-Net model as the basic network architecture, a Multi-Bp-U-Net model is established by using the fully connected layer of the BP neural network; (3.3) The Multi-Bp-U-Net model is trained using the training set data and the validation set data described in step (3.1) to obtain the trained Multi-Bp-U-Net model; (3.4) Input the static ideal contact stress distribution into the trained Multi-Bp-U-Net model to predict the topography layout and obtain the topography layout under the static ideal contact stress distribution state.

8. A system for inverting the ideal prestress distribution of assembly interfaces based on service performance, characterized in that, include: The feature mapping module is used to perform feature mapping of the assembly interface morphology-stress on the mating surface to obtain the contact pressure value and the relative coordinates of each node; The static ideal contact stress distribution inversion module is used to invert the static ideal contact stress distribution based on the contact pressure value and the static contact stress, and obtain the static ideal contact stress distribution. The topography prediction module is used to predict the topography of the assembly interface driven by the static contact stress distribution based on the static ideal contact stress distribution.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the inversion design method for ideal prestress distribution of assembly interface based on service performance traction as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the inversion design method for ideal prestress distribution of assembly interface driven by service performance as described in any one of claims 1 to 7.

Images