Tolerance handling method and system for automotive suspension assembly

Abstract

The application provides a tolerance processing method and system for automobile suspension assembly, which obtains a model file of an automobile suspension and analyzes the model file to determine three-dimensional geometric information, a model tree structure and three-dimensional labeling information carrying engineering semantics; determines constraint relationships and measurement references based on the model tree structure and the three-dimensional geometric information, and creates virtual features required according to the constraint relationships and the measurement references if missing parts in the model file are detected; defines positioning relationships, transmission relationships and tolerance directions between parts of the automobile suspension in a preset semantic constraint window based on the three-dimensional labeling information, and generates a suspension multi-body assembly constraint chain; determines measurement objects and reference objects in the suspension multi-body assembly constraint chain according to obtained analysis requirements, performs Monte Carlo simulation calculation in combination with dimensional tolerances and geometric and position tolerances, and outputs a tolerance analysis report. The application reduces human intervention links through automatic identification and semantic processing, and improves tolerance processing efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of vehicle assembly technology, and in particular to a tolerance handling method and system for automobile suspension assembly. Background Technology

[0002] In the automotive manufacturing industry, assembly tolerance analysis is a core step in ensuring the precision of component fit and the overall vehicle quality. Three-dimensional dimensional deviations in key components such as body panels and chassis structural parts directly affect appearance quality, wind noise levels, and long-term reliability. Related technologies typically utilize 3D scanning equipment to acquire point cloud data of body panels or structural parts, and then compare this data with the original design model to calculate and evaluate 3D dimensional deviations.

[0003] However, when studying automotive suspensions, double wishbone suspensions, including toe angle, camber angle, kingpin inclination angle, and caster angle, require a complete re-establishment due to the complexity of the suspension structure and the cumbersome definition of assembly constraints. This is extremely inefficient, especially when changing vehicle models or modifying parameters. Furthermore, traditional tolerance analysis software necessitates creating numerous virtual points, virtual axes, and virtual planes for analysis. The lack of data consistency between different models and versions makes data inheritance difficult, leading to repetitive modeling that significantly reduces efficiency and weakens traceability. Moreover, traditional tolerance analysis software cannot directly recognize 3D models and tolerance information from CATIA, NX, and CREO, requiring data re-entry, which is time-consuming, error-prone, and disrupts the Model-Based Definition (MBD) value chain, making it impossible to directly read MBD models and PMI (Product Manufacturing Information) tolerance information. Therefore, automotive suspension assembly urgently needs a tolerance analysis solution that can directly utilize MBD models and is compatible with traditional virtual features and the entire process. Summary of the Invention

[0004] This invention provides a tolerance processing method and system for automotive suspension assembly, to solve or alleviate the problem of cumbersome tolerance processing in automotive suspension assembly, which leads to low processing efficiency, as described above.

[0005] In a first aspect, the present invention provides a tolerance processing method for automotive suspension assembly, comprising the following steps: acquiring a model file of an automotive suspension and parsing the model file to determine three-dimensional geometric information, a model tree structure, and three-dimensional annotation information carrying engineering semantics, wherein the three-dimensional annotation information includes dimensional tolerances and geometric tolerances of product manufacturing information, and the automotive suspension is assembled from multiple parts; determining constraint relationships and measurement benchmarks based on the model tree structure and the three-dimensional geometric information; if a part is detected to be missing in the model file, creating the required virtual features based on the constraint relationships and the measurement benchmarks, wherein the virtual features include virtual points, virtual holes and shafts, and virtual planes; defining the positioning relationships, transfer relationships, and tolerance directions between the various parts of the automotive suspension in a preset semantic constraint window based on the three-dimensional annotation information, and generating a multi-body assembly constraint chain for the suspension; acquiring analysis requirements, determining the measurement objects and benchmark objects in the multi-body assembly constraint chain based on the analysis requirements, performing Monte Carlo simulation calculations in conjunction with the dimensional tolerances and the geometric tolerances, and outputting a tolerance analysis report, wherein the measurement objects and the benchmark objects are mutually compatible.

[0006] In some embodiments of the first aspect, the method further includes: if the target part is detected to be non-compliant in the tolerance analysis report, the tolerance parameters of the target part are readjusted through a preset semantic constraint window; the tolerance parameters are recalculated, and the dimensional chain of the target part is updated until the tolerance parameters of the target part pass the verification; wherein, if the target part is the toe angle, the position parameters of the shock absorber connection point or the frame adjustment shim parameters are modified through the preset semantic constraint window until the distribution curve of the toe angle converges to the target range.

[0007] In some embodiments of the first aspect, the tolerance parameters are recalculated, including: constructing an undirected graph of the assembly based on the correspondence between each assembly pair and the constraint object in the vehicle suspension, wherein nodes represent feature surfaces of the assembly pair, and edges represent internal sub-relationships or assembly pair relationships of the same target part; calculating the shortest path between nodes in the undirected graph using the Dijkstra algorithm, and generating the shortest assembly dimension chain based on the measured object by searching; generating a Jacobian matrix based on the shortest assembly dimension chain and combining the pose information of each feature surface, and calculating the tolerance sensitivity and tolerance contribution of each dimensional tolerance and geometric tolerance using the Jacobian matrix.

[0008] In some embodiments of the first aspect, the tolerance parameters are recalculated, including: dividing the bushing region in the vehicle suspension where flexible deformation exists into finite element substructures, and establishing a set of local control points in the area where the target part is located in the bushing region; obtaining the maximum offset and displacement direction of each local control point under various load boundary conditions through deformation simulation under stress, and constructing the displacement vector of the local control point; and embedding the displacement vector of the local control point as a dynamic node into the error propagation path of the topology within the multi-body assembly constraint chain of the suspension to perform flexible response and error coupling calculation.

[0009] In some embodiments of the first aspect, creating the required virtual features based on the constraint relationship and the measurement reference includes: determining the constraint relationship and measurement reference of the virtual features in the vehicle suspension, wherein the measurement reference characterizes the geometric dimensions, shape, and positional relationship of the virtual features, and the constraint relationship defines a set of parameterized rules for the relative position, size, and geometric characteristics between the virtual features and existing real geometric features of the parts, and between multiple virtual features; if the virtual features are to be created, virtual points conforming to the constraint relationship and the measurement reference are created by inputting coordinates and direction vectors, or virtual points conforming to the constraint relationship and the measurement reference are created based on model features, wherein the model features are based on the model file and include basic geometric features and engineering structural features; if the virtual features are to be created, virtual holes or virtual planes conforming to the constraint relationship and the measurement reference are created based on model features or existing virtual points; wherein, when it is necessary to complete the definition of virtual features between symmetrical structures and different parts of the vehicle body, the virtual features and their parameters are copied and pasted onto the target model where the virtual features are located to achieve rapid generation of virtual features between symmetrical structures or similar vehicle models.

[0010] In some embodiments of the first aspect, based on the three-dimensional annotation information, the positioning relationship, transfer relationship, and tolerance direction between the various parts of the vehicle suspension are defined in a preset semantic constraint window to generate a suspension multi-body assembly constraint chain, including: opening the integrated preset semantic constraint window, sequentially defining the correspondence between each assembly pair and the constraint object in the vehicle suspension, the correspondence including positioning relationship, transfer relationship, and tolerance direction; sequentially selecting a first point on the upper control arm and a first virtual point on the frame, and a second point on the upper control arm and a second virtual point on the frame to construct an upper control arm constraint; sequentially selecting a third point on the lower control arm and a third virtual point on the frame, and a third virtual point on the lower control arm... The fourth point on the control arm and the fourth virtual point on the frame are used to construct the lower control arm constraint; the upper end point of the shock absorber and the virtual point on the frame, and the lower end point of the shock absorber and the corresponding point of the lower control arm or steering knuckle are selected in sequence to construct the shock absorber constraint; the upper ball joint point of the steering knuckle and the corresponding point of the upper control arm, and the lower ball joint point of the steering knuckle and the corresponding point of the lower control arm are selected in sequence to construct the steering knuckle constraint; at least one of the upper control arm constraint, the lower control arm constraint, the shock absorber constraint and the steering knuckle constraint is used as the constraint object; according to the correspondence between each assembly pair and the constraint object, combined with the preset shock absorber stroke or the extension range and direction vector of the frame adjustment, a multi-body assembly constraint chain of the double wishbone suspension is generated.

[0011] In some embodiments of the first aspect, the measurement object and reference object in the suspension multi-body assembly constraint chain are determined according to the analysis requirements. Monte Carlo simulation calculations are performed in conjunction with the dimensional tolerances and the geometric tolerances, and a tolerance analysis report is output. This includes: selecting the measurement object and reference object according to the analysis requirements, and setting the calculation direction to create an angle closed loop; wherein, in the angle closed loop setting window, the measurement object is selected as the outer plane feature of the wheel, the reference object is selected as the virtual plane of the frame, and the corresponding calculation direction is set according to different calculation objects to create four angle closed loops: toe angle, camber angle, kingpin inclination angle, and kingpin caster angle; the assembly dimensional chain relationship in the suspension multi-body assembly constraint chain is analyzed, and Monte Carlo simulation calculations are performed in conjunction with the tolerance information of the product manufacturing information, and a tolerance analysis report including the result range, pass rate, optimization suggestions, transfer coefficient, and contribution rate is output.

[0012] In some embodiments of the first aspect, the method further includes: if the vehicle suspension is an adjustable suspension structure, then constraining the inner point of the toe-in or the inner point of the lower control arm with the center point of the eccentric bolt that mounts the waist-shaped hole, and introducing an auxiliary part between the eccentric bolt and the waist-shaped hole as an intermediary of the constraint mechanism, driving the eccentric bolt to move along the waist-shaped hole by rotating it, so as to adjust the wheel camber angle and toe-in angle in the adjustable suspension structure; wherein, a six-sided assembly constraint method is used to assemble and constrain the steering knuckle and brake disc sub-assemblies with the suspension links.

[0013] In some embodiments of the first aspect, at least one of the following is further included: verifying the virtual features through a constructed physical information neural network; constructing a digital twin model of the vehicle suspension, the digital twin model including the geometric tolerances, form and position tolerances, and assembly relationships of all parts; before assembly, inputting the actual measured dimensions of each part into the digital twin model, simulating the tolerance accumulation during the assembly process in real time, and predicting the assembly accuracy and performance; if the prediction results do not meet the preset requirements, generating an optimal shim compensation scheme or adjusting the assembly sequence, wherein the optimal assembly sequence is found and adjusted through a genetic algorithm; establishing a multiphysics coupling model of each part in the vehicle suspension between working conditions, deformation, and tolerance drift, calling the multiphysics coupling model to predict tolerance changes under different working conditions, and outputting dynamic tolerance compensation amounts.

[0014] This invention also provides a tolerance processing system for automotive suspension assembly. The system includes: a file acquisition module for acquiring a model file of the automotive suspension and parsing the model file to determine three-dimensional geometric information, a model tree structure, and three-dimensional annotation information carrying engineering semantics. The three-dimensional annotation information includes dimensional tolerances and geometric tolerances related to product manufacturing information. The automotive suspension is assembled from multiple parts. A feature creation module is also provided, which determines constraint relationships and measurement benchmarks based on the model tree structure and the three-dimensional geometric information. If a missing part is detected in the model file, the system determines the constraint relationships and measurement benchmarks accordingly. The system creates the required virtual features, including virtual points, virtual holes and axes, and virtual planes; a constraint chain module defines the positioning relationships, transfer relationships, and tolerance directions between the various parts of the automotive suspension based on the 3D annotation information in a preset semantic constraint window, generating a multi-body assembly constraint chain for the suspension; a tolerance analysis module is used to obtain analysis requirements, determine the measurement objects and reference objects in the multi-body assembly constraint chain according to the analysis requirements, perform Monte Carlo simulation calculations in conjunction with the dimensional tolerances and the geometric tolerances, and output a tolerance analysis report, wherein the measurement objects and the reference objects are mutually compatible.

[0015] The beneficial effects of this invention are as follows: By directly reading and parsing the model file, i.e., the 3D MBD model, geometric information, 3D annotation information, and model tree structure are extracted, and the tolerances from the design stage are natively inherited; this makes the 3D MBD model the sole carrier of tolerance semantics, avoiding the distortion and inefficiency caused by manual secondary input of tolerance data, realizing the integration of design and tolerance simulation, and improving the accuracy and efficiency of data flow; the model feature-driven modeling logic significantly reduces the number of virtual features; at the same time, through the integrated semantic constraint window, the cumbersome "many-to-one" single constraint is integrated into a "many-to-many" semantic integrated constraint that conforms to the logic of automotive engineering, significantly reducing operation steps, greatly simplifying the definition process of complex suspension assembly constraints, and improving modeling efficiency.

[0016] In summary, from CAD design to tolerance simulation to manufacturing measurement, all data is based on the same 3D MBD model, avoiding semantic loss and inconsistency caused by multi-source data conversion; through automated recognition and semantic processing, a large number of repetitive manual operations are transformed into automatic completion, freeing engineers from tedious modeling work; reducing human intervention links and lowering the probability of human error, while through more accurate tolerance simulation, potential assembly problems can be detected in advance, avoiding rework and waste in the manufacturing stage.

[0017] Furthermore, this invention supports directly modifying the design parameters or tolerance requirements of key components within the same semantic constraint window and triggering calculation verification, updating the dimension chain, and recalculating. This achieves closed-loop control of calculation, optimization, and verification, avoiding repetitive work that requires remodeling or modifying virtual point coordinates, and greatly improving the iteration efficiency of the engineering change phase.

[0018] Furthermore, for adjustable suspension structures, this invention introduces auxiliary parts as intermediaries to simulate the adjustment process of eccentric bolts, and adopts a six-sided assembly constraint method that conforms to actual processes. This solves the problems in related technologies where the adjustment process cannot be simulated and the assembly constraints do not match the actual processes, significantly improving the accuracy of simulation analysis results.

[0019] Furthermore, this invention constructs an undirected graph of the assembly based on graph theory and uses the Dijkstra algorithm to automatically search for the shortest assembly dimension chain. Combined with the Jacobian matrix to calculate tolerance sensitivity and contribution, it realizes automatic search and high-precision tolerance analysis of dimension chains in complex assembly structures, and can accurately identify key positioning points. At the same time, by dividing the flexible deformation region into finite element substructures and establishing a local control point set, the flexible deformation response is explicitly modeled as a displacement vector and embedded in the dimension chain topology. This eliminates the defect of ignoring flexible deformation in related technologies, which leads to error propagation distortion, and significantly improves the analysis accuracy of flexible response and error coupling under complex assembly paths. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0021] In the attached diagram: Figure 1 This is a flowchart illustrating a tolerance handling method for automotive suspension assembly provided in one embodiment of the present invention. Figure 2This is a schematic diagram of an optional structure of a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of the virtual point structure in the MBD model of a double wishbone suspension provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an optional interface for 3D annotation of a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 5 This is a schematic diagram of an optional interface for virtual points in a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 6 This is a schematic diagram of an optional interface for the virtual hole shaft in the MBD model of a double wishbone suspension provided in one embodiment of the present invention; Figure 7 An optional interface diagram is created for virtual points in the MBD model of a double wishbone suspension provided in one embodiment of the present invention; Figure 8 An optional interface diagram is created for virtual points in the MBD model of a double wishbone suspension provided in one embodiment of the present invention; Figure 9 A schematic diagram of an optional interface for copying virtual point data in a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 10 This is a schematic diagram of an optional interface for angle closure in a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 11 This is a schematic diagram of an optional interface for suspension constraints in a double wishbone suspension MBD model provided in one embodiment of the present invention; Figure 12 This is a schematic diagram of an optional interface for optimization calculation in the MBD model of a double wishbone suspension provided in one embodiment of the present invention; Figure 13 This is a schematic diagram of another optional interface for optimization calculation in the MBD model of a double wishbone suspension provided in one embodiment of the present invention; Figure 14 This is a schematic diagram of the hardware structure of a tolerance processing system for automobile suspension assembly provided in one embodiment of the present invention. Detailed Implementation

[0022] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0023] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0024] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0025] In related technologies, the front-to-back direction of a vehicle is considered the first direction, i.e., the X-axis; the left-to-right direction is considered the second direction, i.e., the Y-axis; and the up-and-down direction is considered the third direction, i.e., the Z-axis. In this invention, the use of the same concept or term remains unique and consistent throughout the entire text.

[0026] In tolerance analysis of automotive chassis suspension systems, four-wheel alignment parameters such as toe angle, camber angle, kingpin inclination angle, and caster angle are key indicators determining the vehicle's handling performance, safety, and tire life. To ensure these parameters meet design regulations and vehicle performance requirements after assembly, the industry generally relies on tolerance analysis software to simulate, evaluate, and optimize assembly deviations during the R&D and manufacturing stages. However, in practical applications, problems exist such as inconsistent data sources, severe duplication of modeling, inability to directly read MBD models and PMI tolerance information, cumbersome definition of assembly constraints for complex suspension structures, weak parameter optimization capabilities, and distortion caused by ignoring flexible deformation, making it difficult to efficiently achieve a closed loop of design, assembly, and optimization within a single system.

[0027] To resolve the above issues, please refer to [link / reference]. Figure 1 A flowchart illustrating a tolerance handling method for automotive suspension assembly provided in one embodiment of the present invention includes the following steps: Step S110: Obtain the model file of the car suspension and parse the model file to determine the three-dimensional geometric information, model tree structure and three-dimensional annotation information carrying engineering semantics. The three-dimensional annotation information includes the dimensional tolerances and geometric tolerances of the product manufacturing information. The car suspension is assembled from multiple parts. For example, the tolerance processing system directly identifies and imports the original 3D model files (i.e., model files) from mainstream computer-aided design (CAD) software, such as CATIA format files .CATPart and .CATProduct. In contrast, traditional data exchange typically uses intermediate format conversions such as STEP (Standard for the Exchange of Product Model Data), which easily leads to feature loss and assembly hierarchy confusion. In this embodiment, the original 3D model file is read directly, avoiding feature and model loss caused by intermediate format conversion and ensuring data integrity. Figure 2 The diagram shown illustrates an optional structural schematic of a double wishbone suspension MBD model provided in one embodiment of the present invention. After importing the MBD model based on the model definition, the 3D geometric information, the 3D annotation information of the Product Manufacturing Information (PMI), and the model tree structure are parsed and extracted. The 3D annotation information directly inherits the dimensional and geometric tolerances annotated during the design phase, eliminating the need to redefine tolerances during the analysis phase. This native direct reading and parsing mechanism achieves 100% native inheritance of tolerance data, avoiding the distortion and inefficiency caused by manual secondary entry of tolerance data, ensuring the complete preservation of design semantics during the tolerance simulation phase, and realizing the integration of design and tolerance simulation.

[0028] Step S120: Determine constraint relationships and measurement benchmarks based on the model tree structure and three-dimensional geometric information. If a part is detected to be missing in the model file, create the required virtual features according to the constraint relationships and measurement benchmarks. Virtual features include virtual points, virtual holes and shafts, and virtual planes. In one embodiment, creating the required virtual features based on constraint relationships and measurement benchmarks includes: The constraints and measurement references of virtual features in automotive suspension are determined. The measurement references characterize the geometric dimensions, shape, and positional relationships of the virtual features. The constraints define a set of parameterized rules governing the relative positions, dimensions, and geometric properties between virtual features and existing real geometric features of parts, as well as among multiple virtual features. If a virtual feature is created, virtual points conforming to the constraints and measurement references are created by inputting coordinates and direction vectors, or virtual points conforming to the constraints and measurement references are created based on model features. Model features are based on the identification of model files and include basic geometric features and engineering structural features. If a virtual feature is created, virtual holes, shafts, or virtual planes conforming to the constraints and measurement references are created based on model features or existing virtual points. When defining virtual features between symmetrical structures and different parts of the vehicle body, the virtual features and their parameters are copied and pasted onto the target model where the virtual feature is located, enabling rapid generation of virtual features between symmetrical structures or similar vehicle models.

[0029] For example, please refer to Figure 3 , Figure 4 Employing a model feature-driven modeling logic, the system can directly create constraint relationships and measurement benchmarks based on geometric features, establishing virtual features only when necessary. This significantly reduces the number of virtual features that need to be generated, ensuring data semantics remain consistent with geometric features and reducing the probability of human intervention and errors. For virtual features that must be created, the tolerance processing system provides a creation and reuse mechanism. Specifically, in the virtual feature creation window, virtual points on the chassis can be created by directly inputting precise position coordinates and direction vectors, or virtual points can be automatically generated based on model features. For the creation of virtual holes, shafts, or virtual planes, creation is based on model features or existing virtual points, by specifying positioning parameters. Furthermore, the tolerance processing system provides a data copying function in the data copying window. When defining virtual features between the double wishbone structure and different parts of the vehicle body, users only need to select the source model to directly copy its virtual features and parameters, and paste them onto the target model containing the virtual features with one click. This symmetrical feature copying and reuse mechanism enables rapid transfer of features between symmetrical structures or similar vehicle models, shortening the creation time of symmetrical features and improving modeling efficiency and traceability.

[0030] In this embodiment, a creation and reuse mechanism is provided for virtual features that must be created; specifically, such as... Figure 5As shown, in the quick virtual feature creation window, you can create virtual points on the chassis by directly inputting precise position coordinates, such as: X = -201.997, Y = -419.567, Z = 566.917; direction vectors: I = -0.99405, J = -0.07927, K = 0.07554. Alternatively, virtual points can be automatically generated based on model features (such as hole centers, axis intersections, sphere centers, etc.). For the creation of virtual hole shafts or virtual planes, they are created by specifying positioning parameters (such as centerline creation method, object, datum, direction vector, inner or outer diameter, length, etc.) based on model features or existing virtual points. For example, as... Figure 6 As shown, in the virtual hole shaft creation window, select the hole shaft type, set its positioning parameters and errors, including: inner diameter or outer diameter, tolerance range, and length, to generate a high-precision virtual hole shaft. Additionally, please refer to... Figures 7 to 9 The data copy window provides a data copy function. When it is necessary to complete the virtual feature definition between the double wishbone structure and different parts of the vehicle body, the user only needs to select the source model (such as the left suspension) to directly copy its virtual features and parameters, and paste them onto the target model (such as the right symmetrical part) with one click. This mechanism for copying and reusing symmetrical features enables the rapid transfer of features between symmetrical structures or similar vehicle models, greatly shortening the creation time of symmetrical features and improving modeling efficiency and traceability.

[0031] Step S130: Based on the three-dimensional annotation information, define the positioning relationship, transfer relationship and tolerance direction between the various parts of the automotive suspension in the preset semantic constraint window, and generate the suspension multi-body assembly constraint chain. In some embodiments, step S130 specifically includes the following methods: Open the integrated preset semantic constraint window and define the correspondence between each assembly pair and constraint object in the automotive suspension in sequence. The correspondence includes positioning relationship, transfer relationship and tolerance direction. The upper control arm constraint is constructed by sequentially selecting a first point on the upper control arm and a first virtual point on the frame, and a second point on the upper control arm and a second virtual point on the frame; the lower control arm constraint is constructed by sequentially selecting a third point on the lower control arm and a third virtual point on the frame, and a fourth point on the lower control arm and a fourth virtual point on the frame; the shock absorber constraint is constructed by sequentially selecting an upper end point on the shock absorber and a virtual point on the frame, and a lower end point on the shock absorber and a corresponding point on the lower control arm or steering knuckle; the steering knuckle constraint is constructed by sequentially selecting an upper ball joint point on the steering knuckle and a corresponding point on the upper control arm, and a lower ball joint point on the steering knuckle and a corresponding point on the lower control arm; at least one of the upper control arm constraint, lower control arm constraint, shock absorber constraint, and steering knuckle constraint is used as the constraint object. Based on the correspondence between each assembly pair and the constraint object, and combined with the preset shock absorber stroke or the extension range and direction vector of the frame adjustment, a multi-body assembly constraint chain for the double wishbone suspension is generated.

[0032] For example, please refer to Figure 11 This is a schematic diagram of an optional interface for suspension constraints in a double wishbone suspension MBD model provided in an embodiment of the present invention. For the structurally complex double wishbone suspension system, a highly integrated semantic assembly constraint mechanism is introduced. The integrated preset semantic constraint window, i.e., the double wishbone auxiliary constraint setting window, is opened, and the correspondence between each kinematic pair and the constraint object is defined sequentially. Specific steps include: First, performing upper control arm constraints: in the window, sequentially select the first point on the upper control arm and the first virtual point on the frame, and the second point on the upper control arm and the second virtual point on the frame; second, performing lower control arm constraints: sequentially select the third point on the lower control arm and the third virtual point on the frame, and the fourth point on the lower control arm and the fourth virtual point on the frame; next, performing shock absorber constraints: sequentially select the upper end point of the shock absorber and the virtual point on the frame, and the lower end point of the shock absorber and the corresponding point on the lower control arm or steering knuckle; then, performing steering knuckle constraints: sequentially select the upper ball joint point of the steering knuckle and the corresponding point on the upper control arm, and the lower ball joint point of the steering knuckle and the corresponding point on the lower control arm. Finally, the range and direction vector of the shock absorber travel or frame adjustment are set. Based on the above configuration, the complete double wishbone suspension multi-body assembly constraint chain is parsed and generated in the background. In contrast, traditional software requires the creation of multiple individual constraints when defining such complex assemblies, often requiring more than eight constraints and about one hundred steps to establish a complete assembly relationship. In this embodiment, a highly integrated assembly constraint mechanism is introduced, which can be completed in only about twenty-five steps, significantly reducing the number of steps and greatly simplifying the process of defining assembly constraints for complex suspensions.

[0033] Step S140: Obtain analysis requirements, determine the measurement objects and reference objects in the suspension multi-body assembly constraint chain based on the analysis requirements, perform Monte Carlo simulation calculations in combination with dimensional tolerances and geometric tolerances, and output a tolerance analysis report. The measurement objects and reference objects are mutually compatible.

[0034] In some embodiments, step S140 specifically includes the following methods: Select the measurement object and reference object according to the analysis requirements, and set the calculation direction to create an angle closed loop; in the angle closed loop setting window, select the outer plane feature of the wheel as the measurement object, select the virtual plane of the frame as the reference object, and set the corresponding calculation direction according to different calculation objects to create four angle closed loops: toe angle, camber angle, kingpin inclination angle and kingpin caster angle. The assembly dimension chain relationship in the multi-body assembly constraint chain of the suspension is analyzed, and Monte Carlo simulation calculation is performed in combination with the tolerance information of product manufacturing information. The output includes a tolerance analysis report including the result range, pass rate, optimization suggestions, transfer coefficient and contribution rate.

[0035] For example, the closed-loop measurement object is set and simulation calculation is performed in the angle closed-loop settings window. First, the measurement object and reference object are selected according to the analysis requirements. Specifically, the outer plane feature of the wheel is selected as the measurement object, and the virtual plane of the frame is selected as the reference object. Depending on the calculation object, the corresponding calculation direction is set, thereby creating four angle closed loops in the tolerance processing system, which are used to calculate the toe angle, camber angle, kingpin inclination angle, and caster angle, respectively. Next, based on the set measurement object and integrated constraints, the assembly dimension chain relationship is automatically resolved, and combined with the imported product manufacturing information (PMI) 3D annotation information, at least 10,000 Monte Carlo simulation calculations are performed in the simulation calculation results interface. After the calculation is completed, intuitive statistical charts and key indicators are output, including result range, pass rate, optimization suggestions, transfer coefficient, and contribution rate. Through the rapid definition of multi-directional angle closed loops and high-precision Monte Carlo simulation, high-precision prediction of four-wheel alignment parameters is achieved, enabling accurate identification of key deviation sources and providing data support for subsequent closed-loop optimization design.

[0036] Specifically, please refer to Figure 10 This is a schematic diagram of an optional interface for angle closed-loop calculation in a double wishbone suspension MBD model provided in one embodiment of the present invention. The closed-loop measurement object is set and simulation calculation is performed in the angle closed-loop setting window. First, the measurement object and reference object are selected according to the analysis requirements. Specifically, the outer plane feature of the wheel is selected as the measurement object, and the virtual plane of the vehicle frame is selected as the reference object. Depending on the calculation object, the corresponding calculation direction is set, thereby creating four angle closed loops in the tolerance processing system, which are used to calculate the toe angle, camber angle, kingpin inclination angle, and caster angle, respectively. Next, based on the set measurement object and integrated constraints, the assembly dimension chain relationship is automatically parsed, and combined with the 3D annotation information of the imported product manufacturing information, at least ten thousand Monte Carlo simulation calculations are performed in the simulation calculation result interface.

[0037] After calculation, intuitive statistical charts and key indicators are output, including result range, pass rate, optimization suggestions, transfer coefficient, and contribution rate. For example, the target range for camber angle is set to [-0.4, 0.4]. Through at least 10,000 Monte Carlo simulation calculations, the output calculation distribution curve shows that the pass rate for camber angle calculation reaches 99.7%, meeting the design requirements. However, the calculation result for toe angle shows that it exceeds the preset target requirement, and the pass rate does not meet the standard, providing optimization suggestions. In this embodiment, through the rapid definition of multi-directional angle closed loop and high-precision Monte Carlo simulation, high-precision prediction of four-wheel alignment parameters is achieved, which can accurately identify key deviation sources and provide data support for closed-loop optimization design.

[0038] Optionally, in some embodiments, it further includes: If the target part is found to be non-compliant in the tolerance analysis report, the tolerance parameters of the target part will be readjusted through the preset semantic constraint window. Recalculate the tolerance parameters and update the dimension chain of the target part until the tolerance parameters of the target part pass the verification. If the target part is the toe angle, the position parameters of the shock absorber connection point or the parameters of the frame adjustment shims are modified through the preset semantic constraint window until the distribution curve of the toe angle converges to the target range.

[0039] For example, see details Figure 12 This is a schematic diagram of an optional interface for optimization calculation in a double wishbone suspension MBD model provided in one embodiment of the present invention; for example, see [link to related documentation]. Figure 13 This is a schematic diagram of another optional interface for optimization calculation in the MBD model of a double wishbone suspension provided in one embodiment of the present invention. For example, when the simulation calculation result interface shows that the calculated toe angle exceeds the target requirements and the pass rate fails to meet the standard, through closed-loop optimization design, engineers do not need to remodel or modify complex virtual point coordinates. Instead, they can directly modify the shock absorber connection point position parameters or frame adjustment shim parameters in the integrated double wishbone auxiliary constraint setting window. After modification, a one-click calculation verification is performed in the window. In response to this operation, the dimensional chain is automatically updated in the background and a one-click verification calculation is performed again. After recalculation, the toe angle distribution curve completely converges to the target range, the pass rate reaches the preset requirements, and the design requirements are met. In this embodiment, by directly modifying parameters and performing one-click verification in the same double wishbone auxiliary constraint setting window, closed-loop control of calculation, optimization, and verification is achieved within a single system. This rapid closed-loop iteration mechanism for design, assembly, and optimization greatly reduces repetitive work in the engineering change phase and realizes a digital R&D process that is assembled in one go.

[0040] Optionally, in some embodiments, the tolerance parameters are recalculated, including: An undirected graph of the assembly is constructed based on the correspondence between each assembly pair and the constraint object in the automobile suspension. The nodes represent the feature surfaces of the assembly pair, and the edges represent the internal sub-relationships or assembly pair relationships of the assembly pair of the same target part. Dijkstra's algorithm is used to calculate the shortest path between nodes in an undirected graph, and the shortest assembly dimension chain based on the measured object is generated by search. Based on the shortest assembly dimension chain, a Jacobian matrix is ​​generated by combining the pose information of each feature surface, and the tolerance sensitivity and tolerance contribution of each dimensional tolerance and geometric tolerance are calculated by using the Jacobian matrix.

[0041] For example, to achieve automatic search and high-precision tolerance analysis of dimensional chains in complex suspension assembly structures, an analysis method based on graph theory and the Jacobian screw model is adopted when updating and recalculating the dimensional chains. First, an undirected graph of the assembly is constructed based on the correspondence between each kinematic pair and the constraint object of the suspension. In this undirected graph, nodes represent feature faces of the assembly pairs, and edges represent internal sub-relationships or assembly pair relationships within the same part. For example, the assembly undirected graph is expressed as an adjacency matrix; internal sub-relationships or assembly pair relationships are marked as "1" in the corresponding row and column, while those without relationships are marked as "0". Next, the Dijkstra algorithm is used to calculate the shortest path between nodes in the undirected graph. Starting from the node representing the measurement object in the assembly undirected graph, the minimum weight path to the reference node is automatically searched and calculated, thereby automatically generating the shortest assembly dimensional chain based on the measurement object. Subsequently, based on the shortest assembly dimensional chain, a Jacobian matrix is ​​generated by combining the pose information of each feature facet. Specifically, for each characteristic deviation screw along the deviation propagation path, the homogeneous coordinate transformation matrix of the characteristic and the homogeneous coordinate transformation matrix of the target functional feature are used to calculate the Jacobian matrix corresponding to the deviation screw, which is used to express the deviation screw generated by the unit deviation screw of the characteristic in the target functional feature. Finally, the tolerance sensitivity and tolerance contribution of each dimensional tolerance and geometric tolerance are calculated through the Jacobian matrix. Through this automatic analysis mechanism based on graph theory and the Jacobian matrix, the key positioning points that have the greatest impact on the final assembly accuracy can be accurately identified, providing quantitative guidance for the accurate allocation and optimization design of tolerances, and solving the shortcomings of traditional manual analysis of dimensional chains, which are characterized by low efficiency and high error rate.

[0042] Optionally, in some embodiments, the tolerance parameters are recalculated, including: Finite element substructures are used to divide the bushing region in the automotive suspension where flexible deformation occurs, and a local control point set is established for the area where the target part is located in the bushing region. By performing deformation simulation under stress, the maximum offset and displacement direction of each local control point under various load boundary conditions are obtained, and the displacement vector of the local control point is constructed. The displacement vectors of local control points are used as dynamic nodes and embedded in the error propagation path of the topology within the suspension multi-body assembly constraint chain to perform flexible response and error coupling calculations.

[0043] For example, to eliminate the error propagation distortion caused by neglecting flexible deformation in traditional tolerance analysis, a flexible deformation compensation mechanism is introduced when updating the dimensional chain and recalculating. First, the bushing region in the suspension system with flexible deformation is divided into finite element substructures, and a set of local control points is established for key structural regions. Each control point represents a degree-of-freedom control node of the flexible bushing region in space. Next, through static or dynamic deformation simulation under stress, the maximum offset and displacement direction of each control point under various load boundary conditions are obtained, and the nodal control point displacement vector is constructed. vec{d}_{ij} = [d_{ijx}, d_{ijy}, d_{ijz}] ^T Where vec{d}_{ij} is the displacement vector of the j-th control point in the i-th flexible bushing region in the three-dimensional rectangular coordinate system, d_{ijx}, d_{ijy} and d_{ijz} represent the displacement components of the control point along the X-axis, Y-axis and Z-axis, respectively, and ^T is the transpose.

[0044] Principal component decomposition is performed on the displacement vector of each node control point, retaining the main offset direction and response amplitude as a representation of the geometric deviation of the flexible bushing at the dimensional chain nodes. The node control point displacement vectors are treated as dynamic nodes and explicitly embedded into the error propagation path of the dimensional chain topology, coupled and superimposed with the rigid dimensional chain error for calculation. By explicitly modeling the flexible deformation response and embedding it into the dimensional chain, the accuracy of the coupled analysis of flexible response and error is improved under complex assembly paths, greatly enhancing the prediction accuracy and reliability of the tolerance analysis model under actual working conditions.

[0045] Optionally, in some embodiments, it further includes: If the car suspension is an adjustable suspension structure, then the inner point of the toe bar or the inner point of the lower control arm is constrained by the center point of the eccentric bolt that mounts the waist-shaped hole. An auxiliary part is introduced between the eccentric bolt and the waist-shaped hole as an intermediary of the constraint mechanism. By rotating the eccentric bolt, it is driven to move along the waist-shaped hole to adjust the wheel camber angle and toe angle in the adjustable suspension structure. Among them, the six-sided assembly constraint method is used to assemble and constrain the steering knuckle and brake disc sub-assemblies with the suspension links.

[0046] For example, a six-sided assembly constraint method is used to constrain the assembly of the steering knuckle and brake disc sub-assemblies with the suspension links. The selection of positioning point directions includes: selecting three control arm mounting points on the steering knuckle as Y_1, Y_2, and Y_3 direction positioning points; selecting the midpoints of two control arm mounting points as X_1 and X_2 direction positioning points; and selecting the midpoints of the front upper control arm and rear upper control arm mounting points as Z_1 direction positioning points.

[0047] In this embodiment, to adapt to the adjustable suspension structure and improve the consistency between the simulation model and the actual assembly process, the highly integrated semantic assembly constraints also introduce eccentric bolt constraint pairs and a six-sided assembly constraint method. For the adjustable suspension structure, constraints are established between the inner point of the toe-in or the inner point of the lower control arm and the center point of the eccentric bolt mounting the waist-shaped hole. To dynamically simulate the adjustment process of the eccentric bolt in the virtual prototype, an auxiliary part is introduced as an intermediary for the constraint mechanism between the eccentric bolt itself and the waist-shaped hole. This auxiliary part includes a central axis and a straight line L_1 and a plane P_1 parallel to the adjustment direction of the waist-shaped hole, with the straight line L_1 passing through the plane P_1. The auxiliary part and the waist-shaped hole of the subframe are constrained by a rhombus shape, the central axis of the auxiliary part and the eccentric bolt are constrained by a revolute joint, and the center point of the eccentric bolt head and the center line of the eccentric bolt U-shaped track on the subframe mounting surface are constrained by a point-on-line constraint. Through the synergistic effect of the multi-body constraint pair, rotating the eccentric bolt drives the eccentric bolt itself to move along the waist-shaped hole, thereby dynamically adjusting the camber and toe angles of the wheel in the simulation. This effectively avoids manufacturing and process precision errors within a range of ±4.5 mm. Simultaneously, a six-sided assembly constraint method is used to assemble and constrain the steering knuckle and brake disc assembly with the suspension linkage. The selection process for the positioning point directions is as follows: three control arm mounting points on the steering knuckle are selected as Y_1, Y_2, and Y_3 direction positioning points to constrain the steering knuckle's translational and rotational degrees of freedom in the lateral direction; the midpoints of two control arm mounting points are selected as X_1 and X_2 direction positioning points to constrain the steering knuckle's longitudinal degrees of freedom; the midpoints of the front upper control arm and rear upper control arm mounting points are selected as Z_1 direction positioning points, with the direction set to (0, 0, 1) to constrain the steering knuckle's axial degrees of freedom. By precisely defining the six assembly positioning points mentioned above, the complete constraint of the steering knuckle and brake disc sub-assemblies with the suspension linkage is achieved, making the assembly constraint method of the simulation model highly consistent with the actual assembly process, and significantly improving the accuracy of the simulation analysis results.

[0048] Optionally, in some embodiments, at least one of the following is also included: Virtual features are verified by constructing a physical information neural network; A digital twin model of the vehicle suspension is constructed, which includes the geometric tolerances, form and position tolerances, and assembly relationships of all parts. Before assembly, the actual measured dimensions of each part are input into the digital twin model to simulate the tolerance accumulation during the assembly process in real time and predict the assembly accuracy and performance. If the prediction results do not meet the preset requirements, the optimal shim compensation scheme is generated or the assembly sequence is adjusted. The optimal assembly sequence is found and adjusted using a genetic algorithm. Establish a multiphysics coupling model for each component in the vehicle suspension under working conditions, deformation, and tolerance drift. Use the multiphysics coupling model to predict tolerance changes under different working conditions and output dynamic tolerance compensation.

[0049] For example, geometric features, tolerance features, and assembly constraint features involved in suspension digital twin modeling are extracted to construct a virtual feature dataset; a physical information neural network is constructed, based on a fully connected neural network, and the suspension mechanical equilibrium equation and geometric constraint equation are used as physical constraints embedded in the loss function, which includes two terms: data fitting loss and physical residual loss; the virtual feature sample data and physical equation residuals are jointly trained, and the network parameters are iteratively optimized until the physical residual loss converges to a preset threshold; the virtual features to be verified are input into the trained physical information neural network, and the feature compliance score is output. If the score is lower than the threshold, it is judged as abnormal and a correction suggestion is output.

[0050] Using equipment such as coordinate measuring machines and laser scanners, the actual key dimensions, geometric errors, and form and position errors of each part are acquired. The measured dimensional data are imported into the digital twin model, replacing the initial theoretical dimensions of the model and updating the model parameters. Based on the updated model, the assembly process of the parts is simulated according to the preset assembly sequence, and the tolerance transfer and accumulation of each assembly node are calculated to generate a tolerance accumulation chain. Based on the tolerance accumulation results, the assembly accuracy indicators such as key fit clearances and positioning accuracy of the suspension are calculated. Simultaneously, the performance parameters such as suspension stiffness and damping characteristics are correlated to predict the overall performance after assembly.

[0051] Set assembly accuracy thresholds (such as the range of mating clearances) and performance thresholds (such as stiffness indices), compare the predicted results with the thresholds, and determine whether the requirements are met. If the accuracy is not up to standard, extract the clearance deviations of key mating surfaces, and calculate the optimal gasket thickness and quantity for each mating surface based on the deviation values, part material properties, and gasket specification library to generate a compensation scheme. With the goal of achieving optimal assembly accuracy and minimizing cumulative tolerances, construct an assembly sequence optimization mathematical model and define boundary conditions such as part assembly priority and interference constraints. The optimization is achieved through a genetic algorithm, as detailed below: ① Encoding: Each assembly sequence is encoded as a chromosome, with each gene corresponding to a part assembly sequence number; ② Initialize the population and randomly generate multiple initial assembly sequences (initial population); ③ Fitness assessment: Using the cumulative tolerance and assembly accuracy as fitness functions, calculate the fitness value of each chromosome; ④ Genetic manipulation: through selection, crossover, and mutation, a new generation of population is generated, and inferior assembly sequences are eliminated; ⑤ The iteration terminates, and the iteration is repeated until the fitness value converges or the maximum number of iterations is reached, and the optimal assembly order is output. Output the optimal gasket compensation scheme or the optimal assembly sequence to guide the actual assembly operation. (1) Multiphysics parameter definition: Determine the working condition parameters (load, temperature, vibration frequency), deformation parameters (elastic deformation, plastic deformation), tolerance drift parameters (dimensional drift, form and position deviation), and construct the parameter association dataset; Based on the finite element analysis method, this paper integrates the structural mechanical field (load-deformation), temperature field (thermal expansion and contraction), and tolerance field (dimensional drift) to establish a multi-physics coupling model of working condition-deformation-tolerance drift, and defines the coupling relationship and transfer equation between each physical field. The coupled model was trained using real vehicle test data, and the coupling coefficient was adjusted so that the error between the model prediction results and the measured data met the preset requirements. Input parameters for different driving conditions (high speed, bumpy, heavy load, etc.), call the coupled model to predict the deformation and tolerance drift of each part; calculate the dynamic compensation amount based on the tolerance drift, and output the compensation parameters.

[0052] By employing the above methods, a complete technical system encompassing virtual verification, digital twin simulation, assembly optimization, and dynamic compensation is constructed. This system addresses the industry pain points of traditional suspension assembly, such as reliance on manual labor, uncontrollable precision, and poor dynamic adaptability. It enhances the realism of virtual features through physical information neural network verification, achieves pre-assembly precision control using digital twin models, optimizes assembly schemes based on genetic algorithms, and implements dynamic tolerance compensation using multi-physics field models, forming a comprehensive precision assurance mechanism. Through full-process digital modeling and simulation, assembly risks are identified and corrected in advance, assembly schemes are optimized to reduce tolerance accumulation, and tolerance drift caused by operating conditions is dynamically compensated, thereby improving suspension assembly precision, service stability, and service life.

[0053] This invention also provides a high-efficiency tolerance processing system based on MBD for implementing the above method. The system includes: It is used to directly identify and read the native model files (MBD models) of mainstream CAD software, extract and save the geometric features, model tree structure and PMI 3D annotation information of the native model files; It is used to create constraints and measurement benchmarks based on model geometric features and model tree structure, and provides virtual feature quick creation and copy-paste functions; Used to directly read and reuse PMI tolerance information in the model, realizing the integration of design semantics and tolerance simulation; It is used to provide semantically integrated assembly constraint functions, automatically identify component assembly relationships, and simplify constraint operation steps; Used to parse dimensional chain and PMI information, perform tolerance simulation calculations, and output multi-dimensional calculation results; Used to modify component parameters, perform optimized simulation verification, and complete parameter optimization closed loop.

[0054] Through the above methods, technologies such as direct reading of MBD models, intelligent virtual feature creation, and highly integrated assembly constraints are employed to achieve efficient operation of the entire process from model import to parameter optimization. For complex assembly scenarios such as double wishbone suspension systems, the overall efficiency is several times higher than traditional methods. Feature loss caused by intermediate format conversion is avoided, and direct reading and reuse of PMI tolerance information reduces human input deviations. Simulation results are highly consistent with measured data, improving the reliability of tolerance analysis. The number of virtual features and assembly constraint operation steps are significantly reduced, lowering the difficulty for engineers, reducing human intervention, and lowering the probability of errors. The integration of design (MBD model, PMI tolerance) with tolerance simulation and parameter optimization is achieved, making MBD the sole carrier of tolerance semantics. Constraints and parameters form a traceable knowledge loop, adapting to the engineering logic requirements of industries such as automotive.

[0055] Optionally, in this embodiment, for the four closed-loop targets of the double wishbone suspension—toe angle, camber angle, kingpin inclination angle, and caster angle—a tolerance handling method based on MBD for automotive suspension assembly is adopted. The specific steps are as follows: First, the tolerance processing system imports the MBD model, reads the CATIA native data model of the double wishbone suspension, and identifies and extracts the 3D geometric model, three-dimensional annotation information (including the tolerance requirements of each component), and model tree structure from the native data model.

[0056] Second, based on the measurement requirements of the double wishbone suspension system, input coordinates and directions to quickly create the required virtual points; based on the model's own geometric features and the created virtual points, generate auxiliary features such as virtual holes and virtual planes; using the data copy function, select double wishbone structural parts and body parts to quickly complete the copying and pasting of virtual features and parameters, reducing repetitive creation operations.

[0057] Third, set the measurement object in the tolerance processing system. Select the outer plane feature of the wheel as the measurement object and the virtual plane of the frame as the reference object. According to the calculation requirements of the four closed-loop targets (toe angle, camber angle, kingpin inclination angle, and kingpin caster angle), set the corresponding calculation direction and create four independent closed loops for calculating the four positioning angles respectively.

[0058] Fourth, based on the assembly relationship between the double wishbone suspension parts and the body parts, the constraint objects are selected sequentially (e.g., when the upper control arm is constrained to the frame, the upper control arm point and the frame virtual point are selected sequentially, and the upper control arm point and the frame virtual point are selected sequentially). A semantic high-integration constraint mechanism is adopted, and the definition of all assembly constraints of the double wishbone suspension is completed in only about 25 steps. The positioning relationship, tolerance transfer path and tolerance direction of each component are automatically identified.

[0059] Fifth, the assembly dimension chain relationship and PMI tolerance information of the double wishbone suspension are automatically analyzed. The simulation number is set to 10,000 times, and tolerance simulation calculation is performed. After the calculation is completed, the output results are as follows: the calculation results of camber angle, kingpin inclination angle and kingpin caster angle all meet the preset target requirements; the calculation result of toe angle exceeds the target requirements. The result range, pass rate, tolerance transfer coefficient, contribution rate and optimization suggestions of toe angle are output simultaneously.

[0060] Sixth, directly modify the relevant parameters of the shock absorber and frame in the assembly constraint interface. After modification, click "recalculate" and execute the simulation calculation again. The verification results show that the optimized toe angle calculation results meet the preset target requirements, thus completing the tolerance analysis and optimization closed loop of the double wishbone suspension four-wheel alignment parameters.

[0061] Using the above method, the operational efficiency of the four-wheel alignment parameter analysis of the double wishbone suspension is several times higher than that of traditional software. The simulation results are highly consistent with the measured data, effectively solving the problems of low efficiency, poor accuracy, and cumbersome operation in the traditional tolerance analysis process.

[0062] In some embodiments, please refer to Figure 14 This is a schematic diagram of the hardware structure of a tolerance processing system for automotive suspension assembly provided in an embodiment of the present invention, wherein the tolerance processing system includes: The file acquisition module 210 is used to acquire the model file of the automobile suspension and parse the model file to determine the three-dimensional geometric information, model tree structure and three-dimensional annotation information carrying engineering semantics. The three-dimensional annotation information includes the dimensional tolerances and geometric tolerances of the product manufacturing information. The automobile suspension is formed by assembling multiple parts. The feature creation module 220 determines the constraint relationships and measurement benchmarks based on the model tree structure and three-dimensional geometric information. If a part is detected to be missing in the model file, the required virtual features are created according to the constraint relationships and measurement benchmarks. The virtual features include virtual points, virtual holes and shafts, and virtual planes. The constraint chain module 230 defines the positioning relationship, transfer relationship and tolerance direction between various parts of the automotive suspension based on the 3D annotation information in the preset semantic constraint window, and generates the multi-body assembly constraint chain of the suspension. The tolerance analysis module 240 is used to obtain analysis requirements, determine the measurement objects and reference objects in the suspension multi-body assembly constraint chain based on the analysis requirements, perform Monte Carlo simulation calculations in combination with dimensional tolerances and geometric tolerances, and output a tolerance analysis report. The measurement objects and reference objects are mutually compatible.

[0063] It is understood that the tolerance processing system for automotive suspension assembly provided in the above embodiments and the tolerance processing method for automotive suspension assembly provided in the above embodiments belong to the same concept. The specific execution method of the tolerance processing method for automotive suspension assembly has been described in detail in the above method embodiments and will not be repeated here. In practical applications, the tolerance processing system for automotive suspension assembly provided in the above embodiments can allocate the above functions to different functional modules as needed. That is, the internal structure of the tolerance processing system for automotive suspension assembly is divided into different functional modules, and then all or part of the functions of the corresponding functional modules are implemented through the tolerance processing method for automotive suspension assembly described in the above embodiments. For example, all or part of the functions of the file acquisition module 210 can be implemented through the relevant execution process of step S110; all or part of the functions of the feature creation module 220 can be implemented through the relevant execution process of step S120; all or part of the functions of the constraint chain module 230 can be implemented through the relevant execution process of step S130; and all or part of the functions of the tolerance analysis module 240 can be implemented through the relevant execution process of step S140. For specific implementation processes, please refer to the above embodiments, and no specific limitations are imposed here.

[0064] In summary, the tolerance processing system for automotive suspension assembly proposed in this invention directly reads and parses model files, i.e., 3D MBD models, extracting geometric information, 3D annotation information, and model tree structure, and natively inheriting tolerances from the design stage. This makes the 3D MBD model the sole carrier of tolerance semantics, avoiding the distortion and inefficiency caused by manual secondary input of tolerance data, achieving the integration of design and tolerance simulation, and improving the accuracy and efficiency of data flow. The system employs model feature-driven modeling logic, significantly reducing the number of virtual features. Simultaneously, through an integrated semantic constraint window, the cumbersome "many-to-one" single constraints are integrated into "many-to-many" semantic integrated constraints that conform to automotive engineering logic, significantly reducing operation steps, greatly simplifying the definition process of complex suspension assembly constraints, and improving modeling efficiency.

[0065] In summary, from CAD design to tolerance simulation to manufacturing measurement, all data is based on the same 3D MBD model, avoiding semantic loss and inconsistency caused by multi-source data conversion. Through automated recognition and semantic processing, a large number of repetitive manual operations are transformed into automatic system completion, freeing engineers from tedious modeling work. The reduction of human intervention links lowers the probability of human error, while more accurate tolerance simulation can identify potential assembly problems in advance, avoiding rework and waste in the manufacturing stage.

[0066] In an exemplary embodiment of the present invention, a computer device is also provided. The computer device may include a memory, a processor, and a computer program stored in the memory. The processor can execute the computer program to cause the computer device to perform actions such as... Figure 1 The steps of a tolerance handling method for automotive suspension assembly are shown. This invention provides a computer device comprising: a processor, a memory, a power supply, a display unit, and an input unit.

[0067] The processor is the control center of a computer device. It connects various components via interfaces and circuits, and performs various functions of the computer device by running or executing computer programs / instructions stored in memory, thereby providing overall monitoring of the computer device. In some embodiments, when the processor calls a computer program stored in memory, it can execute actions such as... Figure 1 The steps of the tolerance handling method for automotive suspension assembly are shown. Optionally, the processor may include one or more processing units; preferably, the processor may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. In some embodiments, the processor and memory may be implemented on a single chip; in other embodiments, they may be implemented separately on independent chips.

[0068] The memory mainly consists of a stored program area and a stored data area. The stored program area can store the operating system, various applications, etc.; the stored data area can store instruction data created according to the use of the computer device. In addition, the memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0069] Computer equipment also includes power supplies (such as batteries) that power various components. These power supplies can be connected to the processor logic through a power management system, which can then manage functions such as charging, discharging, and power consumption.

[0070] The display unit can be used to display information input by the user or information provided to the user, and can also be used to display various menus of the computer device. In this embodiment of the invention, it is mainly used to display the display interface of various applications in the computer device, as well as text, images, and other objects displayed in the display interface. The display unit may include a display panel. The display panel may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0071] The input unit can be used to receive information such as numbers or characters input by the user. The input unit may include a touch panel and other input devices. The touch panel can also be referred to as a touch screen, and the touch panel can collect touch operations on or near the user (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel).

[0072] Specifically, the touch panel can detect user touch operations, detect the signals generated by the touch operations, convert these signals into touch point coordinates and send them to the processor, receive commands transmitted by the processor and execute them. Furthermore, various types of touch panel interaction can be implemented, including resistive, capacitive, infrared, and surface acoustic wave inputs. Other input devices include, but are not limited to, one or more of the following: physical keyboard, function keys (such as volume control buttons, power buttons, etc.), trackball, mouse, and joystick.

[0073] Of course, the touch panel can also cover the display panel. When the touch panel detects a touch operation on or near it, it can transmit the information to the processor to determine the type of touch event. Subsequently, the processor provides corresponding visual output on the display panel based on the type of touch event. The touch panel and the display panel are two separate components to implement the input and output functions of the computer device. However, in some embodiments, the touch panel and the display panel can be integrated to implement the input and output functions of the computer device.

[0074] Computer devices may also include one or more sensors, such as pressure sensors, gravity acceleration sensors, proximity sensors, etc. Of course, depending on the specific application scenario, the aforementioned computer devices may also include other components such as cameras.

[0075] In an exemplary embodiment of the present invention, a computer-readable storage medium is also provided, which stores a computer program / instructions. When executed by a processor, the computer program / instructions enable the computer device to perform the functions described in the present invention. Figure 1 The steps of the tolerance handling method for automotive suspension assembly are shown.

[0076] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A tolerance handling method for automobile suspension assembly, characterized in that, The method includes: The model file of the automobile suspension is obtained and parsed to determine the three-dimensional geometric information, model tree structure and three-dimensional annotation information carrying engineering semantics. The three-dimensional annotation information includes the dimensional tolerances and geometric tolerances of the product manufacturing information. The automobile suspension is assembled from multiple parts. Based on the model tree structure and the three-dimensional geometric information, constraint relationships and measurement benchmarks are determined. If a part is detected to be missing in the model file, the required virtual features are created according to the constraint relationships and the measurement benchmarks. The virtual features include virtual points, virtual holes and shafts, and virtual planes. Based on the three-dimensional annotation information, the positioning relationship, transmission relationship and tolerance direction between the various parts of the automotive suspension are defined in the preset semantic constraint window to generate a multi-body assembly constraint chain for the suspension. The analysis requirements are obtained, and the measurement objects and reference objects in the suspension multi-body assembly constraint chain are determined based on the analysis requirements. Monte Carlo simulation calculations are performed in combination with the dimensional tolerances and the geometric tolerances, and a tolerance analysis report is output. The measurement objects and the reference objects are mutually matched.

2. The tolerance handling method for automobile suspension assembly according to claim 1, characterized in that, Also includes: If the target part is found to be non-compliant in the tolerance analysis report, the tolerance parameters of the target part are readjusted through the preset semantic constraint window; The tolerance parameters are recalculated, and the dimension chain of the target part is updated until the tolerance parameters of the target part pass the verification. If the target part is the toe angle, the position parameters of the shock absorber connection point or the parameters of the frame adjustment shims are modified through a preset semantic constraint window until the distribution curve of the toe angle converges to the target range.

3. The tolerance handling method for automobile suspension assembly according to claim 2, characterized in that, The tolerance parameters are recalculated, including: An undirected graph of the assembly is constructed based on the correspondence between each assembly pair and the constraint object in the automobile suspension. The nodes represent the feature surfaces of the assembly pair, and the edges represent the internal sub-relationships of the assembly pair or the assembly pair relationships of the same target part. The Dijkstra algorithm is used to calculate the shortest path between nodes in the undirected graph, and the shortest assembly dimension chain based on the measured object is searched and generated. Based on the shortest assembly dimension chain, a Jacobian matrix is ​​generated by combining the pose information of each feature surface, and the tolerance sensitivity and tolerance contribution of each dimensional tolerance and geometric tolerance are calculated through the Jacobian matrix.

4. The tolerance handling method for automobile suspension assembly according to claim 2, characterized in that, The tolerance parameters are recalculated, including: Finite element substructures are performed on the bushing region of the automobile suspension where flexible deformation occurs, and a local control point set is established for the region where the target part is located in the bushing region. By performing deformation simulation under stress, the maximum offset and displacement direction of each local control point under various load boundary conditions are obtained, and the displacement vector of the local control point is constructed. The displacement vector of the local control point is used as a dynamic node and embedded in the error propagation path of the topology within the constraint chain of the suspension multibody assembly to perform flexible response and error coupling calculation.

5. The tolerance handling method for automobile suspension assembly according to any one of claims 1 to 4, characterized in that, Based on the constraints and the measurement benchmark, the required virtual features are created, including: Determine the constraint relationship and measurement reference of the virtual feature in the vehicle suspension, wherein the measurement reference characterizes the geometric dimensions, shape and positional relationship of the virtual feature, and the constraint relationship defines a set of parameterized rules for the relative position, size and geometric characteristics between the virtual feature and the existing real geometric features of the part, and among multiple virtual features; If the virtual feature is created, a virtual point conforming to the constraint relationship and the measurement benchmark is created by inputting coordinates and direction vectors, or a virtual point conforming to the constraint relationship and the measurement benchmark is created based on model features. The model features are based on the model file and include basic geometric features and engineering structural features. If the virtual feature is created, a virtual hole or virtual plane that conforms to the constraint relationship and the measurement datum is created based on the model feature or existing virtual points. When it is necessary to define virtual features between symmetrical structures and different parts of the vehicle body, the virtual features and their parameters can be copied and pasted onto the target model where the virtual features are located, thereby enabling the rapid generation of virtual features between symmetrical structures or similar vehicle models.

6. The tolerance handling method for automobile suspension assembly according to any one of claims 1 to 4, characterized in that, Based on the 3D annotation information, the positioning relationships, transfer relationships, and tolerance directions between the various parts of the automotive suspension are defined in a preset semantic constraint window, generating a multi-body assembly constraint chain for the suspension, including: Open the integrated preset semantic constraint window and define the correspondence between each assembly pair and constraint object in the automotive suspension in sequence. The correspondence includes positioning relationship, transfer relationship and tolerance direction. The upper control arm constraint is constructed by sequentially selecting a first point on the upper control arm and a first virtual point on the vehicle frame, and a second point on the upper control arm and a second virtual point on the vehicle frame; the lower control arm constraint is constructed by sequentially selecting a third point on the lower control arm and a third virtual point on the vehicle frame, and a fourth point on the lower control arm and a fourth virtual point on the vehicle frame; the shock absorber constraint is constructed by sequentially selecting an upper end point on the shock absorber and a virtual point on the vehicle frame, and a corresponding point on the lower control arm or steering knuckle; the steering knuckle constraint is constructed by sequentially selecting an upper ball joint point on the steering knuckle and a corresponding point on the upper control arm, and a corresponding point on the lower ball joint point on the steering knuckle; at least one of the upper control arm constraint, the lower control arm constraint, the shock absorber constraint, and the steering knuckle constraint is used as the constraint object. Based on the correspondence between each assembly pair and the constraint object, and combined with the preset shock absorber stroke or the extension range and direction vector of the frame adjustment, a multi-body assembly constraint chain for the double wishbone suspension is generated.

7. The tolerance handling method for automobile suspension assembly according to any one of claims 1 to 4, characterized in that, Based on the analysis requirements, determine the measurement objects and reference objects in the suspension multi-body assembly constraint chain. Combine the dimensional tolerances and geometric tolerances to perform Monte Carlo simulation calculations, and output a tolerance analysis report, including: Select the measurement object and reference object according to the analysis requirements, and set the calculation direction to create an angle closed loop; wherein, in the angle closed loop setting window, the measurement object is selected as the outer plane feature of the wheel, the reference object is selected as the virtual plane of the frame, and the corresponding calculation direction is set according to different calculation objects to create four angle closed loops: toe angle, camber angle, kingpin inclination angle and kingpin caster angle. The assembly dimension chain relationship in the multi-body assembly constraint chain of the suspension is analyzed, and Monte Carlo simulation calculation is performed in combination with the tolerance information of the product manufacturing information. The output is a tolerance analysis report including the result range, pass rate, optimization suggestions, transfer coefficient and contribution rate.

8. The tolerance handling method for automobile suspension assembly according to any one of claims 1 to 4, characterized in that, Also includes: If the vehicle suspension is an adjustable suspension structure, then the inner point of the toe bar or the inner point of the lower control arm is constrained by the center point of the eccentric bolt that mounts the waist-shaped hole. An auxiliary part is introduced between the eccentric bolt and the waist-shaped hole as an intermediary for the constraint mechanism. By rotating the eccentric bolt, it is driven to move along the waist-shaped hole to adjust the wheel camber angle and toe angle in the adjustable suspension structure. Among them, a six-sided assembly constraint method is used to assemble and constrain the steering knuckle and brake disc sub-assemblies with the suspension links.

9. The tolerance handling method for automobile suspension assembly according to any one of claims 1 to 4, characterized in that, It also includes at least one of the following: The virtual features are verified by constructing a physical information neural network; A digital twin model of the vehicle suspension is constructed, which includes the geometric tolerances, form and position tolerances, and assembly relationships of all parts. Before assembly, the actual measured dimensions of each part are input into the digital twin model to simulate the tolerance accumulation during the assembly process in real time and predict the assembly accuracy and performance. If the prediction results do not meet the preset requirements, an optimal shim compensation scheme is generated or the assembly sequence is adjusted. The optimal assembly sequence is found and adjusted using a genetic algorithm. A multiphysics coupling model is established for each component in the vehicle suspension under working conditions, deformation, and tolerance drift. The multiphysics coupling model is then used to predict tolerance changes under different working conditions, and the dynamic tolerance compensation amount is output.

10. A tolerance handling system for automotive suspension assembly, characterized in that, The system includes: The file acquisition module is used to acquire the model file of the automobile suspension and parse the model file to determine the three-dimensional geometric information, model tree structure and three-dimensional annotation information carrying engineering semantics. The three-dimensional annotation information includes the dimensional tolerances and geometric tolerances of the product manufacturing information. The automobile suspension is assembled from multiple parts. The feature creation module determines constraint relationships and measurement benchmarks based on the model tree structure and the three-dimensional geometric information. If a part is detected to be missing in the model file, the module creates the required virtual features according to the constraint relationships and the measurement benchmarks. The virtual features include virtual points, virtual holes and shafts, and virtual planes. The constraint chain module defines the positioning relationship, transfer relationship and tolerance direction between the various parts of the automotive suspension based on the three-dimensional annotation information in a preset semantic constraint window, and generates a multi-body assembly constraint chain for the suspension. The tolerance analysis module is used to obtain analysis requirements, determine the measurement objects and reference objects in the suspension multi-body assembly constraint chain based on the analysis requirements, perform Monte Carlo simulation calculations in combination with the dimensional tolerances and the geometric tolerances, and output a tolerance analysis report. The measurement objects and the reference objects are mutually compatible.

Images