High-precision construction simulation analysis method and system for digital assembly

Abstract

The invention relates to the technical field of high-precision building simulation analysis for digital assembly, and discloses a high-precision building simulation analysis method and system for digital assembly, which comprises a high-precision building simulation analysis method and a high-precision building simulation analysis system for digital assembly. The key technical bottlenecks that in the prior art, the assembly precision depends on artificial experience, tolerance distribution lacks an optimization basis, errors cannot be controllably accumulated and the like are overcome, precision modeling and data structured expression of the whole assembly process are achieved, the error propagation simulation capacity based on a statistical method is provided, a performance-oriented assembly precision evaluation mechanism is established, and the assembly precision evaluation efficiency is improved. A closed-loop tolerance optimization and parameter feedback control path is constructed, and the predictability and the digital control level of the assembly process are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of high-precision construction simulation analysis technology for digital assembly, specifically to a high-precision construction simulation analysis method and system for digital assembly. Background Technology

[0002] The product structures in the aerospace, rail transportation, shipbuilding, and high-end equipment manufacturing sectors are characterized by large size, complexity, and modularity. Modular construction, as an important means to improve manufacturing efficiency and ensure assembly consistency, has been widely used in engineering practice. However, due to uncertainties in manufacturing precision, assembly errors, and process control among modules, the overall construction precision is difficult to guarantee effectively, thus affecting product performance stability and service reliability.

[0003] In existing technologies, the assembly accuracy control of modular products mainly relies on the following methods: first, reducing error sources by improving the machining accuracy of parts and the rigidity of tooling; second, using post-assembly measurement and correction methods, such as welding repair, grinding, and shim adjustment, to compensate for assembly deviations; and third, using rules of thumb for tolerance design and assembly process planning. However, these methods have significant limitations:

[0004] Simply relying on improved manufacturing precision will significantly increase costs, and it is difficult to achieve globally consistent high-precision manufacturing in large components. Secondly, post-processing corrective measures not only increase rework time but also easily introduce new accumulated errors. Thirdly, empirical tolerance allocation methods lack quantitative basis and cannot be reasonably adapted to different structures, assembly paths, and process parameters, making it difficult to meet the precision collaborative control requirements under variable assembly scenarios.

[0005] In recent years, with the development of digital manufacturing and modeling simulation technologies, error modeling, tolerance analysis, and virtual verification technologies for assembly processes have gradually emerged, possessing the potential to predict error propagation trends and assess assembly quality in advance through simulation. However, most existing methods currently remain at the level of local assembly error analysis, lacking an integrated mechanism for precision modeling, simulation, and optimization covering the entire process, and have not yet formed a digital closed-loop precision control system for the entire assembly chain.

[0006] Therefore, there is an urgent need for a digital construction accuracy analysis method that integrates multi-source heterogeneous data modeling, dimensional chain error propagation analysis, error simulation evaluation based on statistical methods, and tolerance optimization feedback, in order to achieve high-precision construction simulation and controllable assembly accuracy management in modular assembly scenarios, thereby improving the consistency and predictability of system-level assembly quality. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: a high-precision construction simulation analysis method for digital assembly, comprising: collecting multi-source heterogeneous data and constructing a digital precision modeling dataset of the assembly object.

[0008] Based on the digital precision modeling dataset, an assembly dimension chain model is generated to describe the error propagation relationship between components.

[0009] The assembly dimension chain model was simulated and analyzed, and the assembly error propagation simulation was performed based on statistical methods.

[0010] Based on the simulation results, the distribution of deviations at key assembly nodes and the impact of cumulative errors on assembly performance are output.

[0011] By using the influencing indicators as optimization inputs, a tolerance allocation optimization model under the target accuracy constraint is constructed.

[0012] The initial accuracy modeling parameters are updated based on the output of the tolerance allocation optimization model.

[0013] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the acquisition of multi-source heterogeneous data includes the three-dimensional geometric dimensions, tolerance requirements, and historical assembly error information of the components in the assembly object.

[0014] The three-dimensional geometric dimension data of the components in the assembly object are extracted from the assembly process database. Based on the CAD model, the geometric parameters of the key assembly interface and feature contour are extracted. At the same time, the initial dimension-tolerance mapping relationship table is constructed by combining the nominal tolerance range recorded in the design drawings.

[0015] The measurement deviation data collected from similar process routes in the past are retrieved from the production history data platform, and an error data index structure with a one-to-one correspondence is established based on the component number and assembly batch number.

[0016] Measurement deviation data include initial docking error, fixture positioning error, assembly residual deformation value, and process measurement correction amount.

[0017] The system sets conditions to check the consistency of timestamps and the integrity of data. When missing fields or inconsistent items are found, the system triggers data completion logic to interpolate, reconstruct, and remove incomplete records.

[0018] They are uniformly converted into a standardized format and stored as a digital precision modeling dataset.

[0019] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the construction of the digital precision modeling dataset includes: establishing an attribute label set based on multi-source heterogeneous data, and constructing a data matrix structure indexed by component number.

[0020] The tag set includes component identification number, spatial positioning reference system, dimensional variable group, tolerance tolerance group, historical deviation distribution, deformation response function and assembly sequence number.

[0021] For the assembly relationships between components, an assembly constraint dual table is constructed based on the assembly contact surface type and tolerance fit type to record the coupling information of geometric fit relationship and assembly sequence between two components.

[0022] Establish a three-dimensional relationship diagram of components, connections, and fixtures to express the error propagation path between components, and mark the intensity of the influence transmission through the accuracy sensitivity matrix of the connection nodes.

[0023] When a component has multiple error sources, a path weight adjustment factor is introduced, and the path weight is determined based on the least squares fitting result of the historical residuals.

[0024] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the assembly dimension chain model includes: constructing a directed graph of error propagation based on the structural connection and tolerance fit characteristics between components, where each node represents a component dimension variable, each edge represents a path for the propagation of dimension differences and errors, and the weight of the edge represents the error sensitivity.

[0025] The assembly dimension chain model introduces a node type marking mechanism to distinguish between rigid nodes, flexible nodes, and redundant nodes, and configures different error calculation methods for each node type.

[0026] Set boundary conditions when constructing the transmission path.

[0027] Boundary conditions include initial reference point locking, constraint condition propagation termination rules, and process tolerance upper limits.

[0028] When there are multiple parallel assembly structures in the component assembly process, the assembly dimension chain model adopts a combined chain strategy to construct a set of parallel transmission paths and sets path synthesis rules to statistically fuse multi-path errors.

[0029] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the simulation analysis includes: discretizing the error propagation path in the assembly dimension chain model, setting the number of simulation sampling rounds and the error disturbance range, and simulating the possible assembly error state space by adding random disturbance samples to each path parameter.

[0030] The Monte Carlo simulation method is adopted, and the input disturbance is probabilistically sampled by setting the error distribution type. In each iteration, the cumulative error output of the final assembly result is calculated.

[0031] For highly coupled path nodes, a local sensitivity analysis function is introduced. By calculating the partial derivative of the target assembly dimension with respect to the error of a certain intermediate node, the error amplification link is identified, and the transmission path with high influence weight is recorded.

[0032] Set a standard for judging the stability of the simulation, and set the result variance change to be less than a threshold. If the result variance change is less than the threshold, the simulation process will be terminated early.

[0033] If the variance change of the result is less than the threshold, automatically increase the number of sampling rounds to improve accuracy and confidence.

[0034] Output the critical path distribution histogram, error superposition curve, and extreme deviation distribution point set.

[0035] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the assembly error propagation simulation includes: establishing a local differential perturbation function for the error critical path nodes identified in the assembly error propagation simulation, calculating the response function value of the overall assembly accuracy to the node deviation using the chain rule, and generating an error sensitivity matrix.

[0036] A spatial distribution error mapping map is constructed to map nodal deviations to the outer contour of components or assembly contact surfaces, and an error heat map is generated through an interpolation algorithm.

[0037] When there are intermediate components in the path that are interfered with by fixed clamps, the direction and magnitude of the transmitted error are adjusted according to the clamp stiffness coefficient. It is then determined whether the additional error caused by the clamp deformation exceeds the set error. If it does, it is marked as a rigid shear point.

[0038] The dynamic error propagation process is tracked along the entire path, and the curve of error accumulation rate changing with path depth is recorded to identify the error mutation interval and controllable interval.

[0039] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the output of the performance impact index includes: calculating the position offset of the final assembly node based on the simulation output of assembly error propagation, and calculating the Hausdorff distance between the position offset and the design reference shape based on the assembly target contour as the contour deviation index.

[0040] The surface-to-surface contact unevenness index and the bolt hole center drift distance are extracted as factors affecting connection quality.

[0041] An assembly stability function is introduced, defined as the variance variation trend of error distribution in multiple assemblies. Under different working conditions, the error is concentrated in a few paths, indicating the existence of an error aggregation bottleneck, which serves as an index of stability degradation.

[0042] Calculate the assembly reliability score.

[0043] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the tolerance allocation optimization model includes: constructing a multi-objective optimization problem containing constraints and an objective function, wherein the objective function is a weighted combination of minimizing the sum of squared assembly deviations, the total manufacturing cost, and the response delay penalty term.

[0044] The performance impact index of key nodes is used as the input boundary under constraints. A linear programming algorithm is used to solve the problem of uniform tolerance allocation for regular parts. The linear programming algorithm is enabled when the upper and lower tolerance limits are strictly symmetrical and the transmission path is linearly separable.

[0045] The constraints include the maximum permissible deviation, the upper limit of imbalance, and the lower limit of reliability.

[0046] For nonlinear error propagation paths, a genetic algorithm is introduced to perform adaptive optimization. Through crossover and mutation operations, a better tolerance combination is searched. The fitness function converges continuously for more than 10 generations, and the results are output in advance.

[0047] As a preferred embodiment of the high-precision construction simulation analysis method for digital assembly described in this invention, the output updating of the initial precision modeling parameters includes: reloading the output tolerance allocation matrix as a weight term to the corresponding component tolerance field in the initial digital modeling dataset, and updating the data items that are better than the initial values ​​by fully covering them.

[0048] Refresh the error propagation weights of the corresponding edges in the assembly dimension chain model, recalculate the sensitivity factors for edges with path weight changes, and update the full-link weight graph.

[0049] The re-simulation process is triggered based on the feedback results. When tolerance optimization leads to a significant decrease in the error of key nodes, the feedback is terminated.

[0050] If no significant decrease in error occurs, the system automatically enters a new round of modeling-simulation-optimization iteration process until the system error stabilizes within the preset accuracy window.

[0051] After the update is completed, the new modeling results will be marked as the current version, the old version will be retained for comparison and rollback, and synchronized to the digital assembly.

[0052] A high-precision construction simulation analysis system for digital assembly is characterized by comprising: a multi-source data acquisition and modeling module, an error propagation modeling module, an assembly simulation analysis module, an assembly performance evaluation module, a tolerance optimization control module, and a modeling parameter feedback correction module.

[0053] The multi-source data acquisition and modeling module includes acquiring multi-source heterogeneous data and constructing a digital precision modeling dataset of the assembly object.

[0054] The error propagation modeling module includes generating an assembly dimension chain model based on a digital precision modeling dataset to describe the error propagation relationship between components.

[0055] The assembly simulation analysis module includes performing simulation analysis on the assembly dimension chain model and executing assembly error propagation simulation based on statistical methods.

[0056] The assembly performance evaluation module includes outputting the distribution of deviations at key assembly nodes and the impact of cumulative errors on assembly performance based on simulation results.

[0057] The tolerance optimization control module includes constructing a tolerance allocation optimization model under target accuracy constraints by taking influencing indicators as optimization inputs.

[0058] The modeling parameter feedback correction module includes updating the initial accuracy modeling parameters based on the tolerance allocation and optimized model output.

[0059] The beneficial effects of this invention are as follows: This invention proposes a high-precision construction simulation analysis method and system for digital assembly, which overcomes the key technical bottlenecks in the prior art, such as assembly accuracy relying on manual experience, lack of optimization basis for tolerance allocation, and uncontrollable accumulation of errors. It realizes the accuracy modeling and data structured expression of the entire assembly process, provides error propagation simulation capability based on statistical methods, establishes a performance-oriented assembly accuracy evaluation mechanism, and constructs a closed-loop tolerance optimization and parameter feedback control path, which significantly improves the predictability and digital control level of the assembly process. Attached Figure Description

[0060] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0061] Figure 1 A flowchart of a high-precision construction simulation analysis method and system for digital assembly provided in the first embodiment of the present invention; Detailed Implementation

[0062] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0063] Example 1, referring to Figure 1 As an embodiment of the present invention, a high-precision construction simulation analysis method for digital assembly is provided, comprising:

[0064] S1: Collect multi-source heterogeneous data to construct a digital precision modeling dataset for assembly objects.

[0065] Multi-source heterogeneous data includes the three-dimensional geometric dimensions, tolerance requirements, and historical assembly error information of the components in the assembly object.

[0066] The three-dimensional geometric dimension data of the components in the assembly object are extracted from the assembly process database. Based on the CAD model, the geometric parameters of the key assembly interface and feature contour are extracted. At the same time, the initial dimension-tolerance mapping relationship table is constructed by combining the nominal tolerance range recorded in the design drawings.

[0067] The measurement deviation data collected from similar process routes in the past are retrieved from the production history data platform, and an error data index structure with a one-to-one correspondence is established based on the component number and assembly batch number.

[0068] Measurement deviation data include initial docking error, fixture positioning error, assembly residual deformation value, and process measurement correction amount.

[0069] Set judgment conditions to check the consistency of timestamps and data integrity. When missing fields or inconsistent items are found, trigger data completion logic to interpolate, reconstruct, and remove incomplete records.

[0070] They are uniformly converted into a standardized format and stored as a digital precision modeling dataset.

[0071] Furthermore, based on multi-source heterogeneous data, an attribute label set is established, and a data matrix structure indexed by component number is constructed.

[0072] The tag set includes component identification number, spatial positioning reference system, dimensional variable group, tolerance tolerance group, historical deviation distribution, deformation response function and assembly sequence number.

[0073] For the assembly relationships between components, an assembly constraint dual table is constructed based on the assembly contact surface type and tolerance fit type to record the coupling information of geometric fit relationship and assembly sequence between two components.

[0074] Establish a three-dimensional relationship diagram of components, connections, and fixtures to express the error propagation path between components, and mark the intensity of the influence transmission through the accuracy sensitivity matrix of the connection nodes.

[0075] A preferred scheme for constructing a data matrix structure indexed by component numbers is as follows:

[0076]

[0077] Among them, S k Let N represent the cumulative error of the k-th assembly node, N represent the total number of error source nodes in the dimension chain, i represent the index of the error source number, and W represent the cumulative error of the k-th assembly node. ij δ represents the error sensitivity coefficient from component i to j. ij This represents the error disturbance between component i and j, where j represents the error receiving node number.

[0078] When a component has multiple error sources, a path weight adjustment factor is introduced, and the path weight is determined based on the least squares fitting result of the historical residuals.

[0079] S2: Based on the digital precision modeling dataset, generate an assembly dimension chain model to describe the error propagation relationship between components.

[0080] Based on the structural connections and tolerance fits between components, a directed graph of error propagation is constructed. Each node in the graph represents a component size variable, each edge represents a path for the propagation of size differences and errors, and the weight of the edge represents the error sensitivity.

[0081] The assembly dimension chain model introduces a node type marking mechanism to distinguish between rigid nodes, flexible nodes, and redundant nodes, and configures different error calculation methods for each node type.

[0082] Set boundary conditions when constructing the transmission path.

[0083] Boundary conditions include initial reference point locking, constraint condition propagation termination rules, and process tolerance upper limits.

[0084] When there are multiple parallel assembly structures in the component assembly process, the assembly dimension chain model adopts a combined chain strategy to construct a set of parallel transmission paths and sets path synthesis rules to statistically fuse multi-path errors.

[0085] S3: Perform simulation analysis on the assembly dimension chain model and execute assembly error propagation simulation based on statistical methods.

[0086] The error propagation path in the assembly dimension chain model is discretized, the number of simulation sampling rounds and the error disturbance range are set, and random disturbance samples are added to each path parameter to simulate the possible assembly error state space, thereby generating simulation sample input.

[0087] A preferred method for generating simulation sample input is:

[0088]

[0089] Where, δ k σ represents the error disturbance of the k-th component and connection node, μ represents the average error value of historical data, and σ represents the error disturbance of the k-th component and connection node.2 This represents the variance of the error.

[0090] The Monte Carlo simulation method is adopted, and the input disturbance is probabilistically sampled by setting the error distribution type. In each iteration, the cumulative error output of the final assembly result is calculated.

[0091] For highly coupled path nodes, a local sensitivity analysis function is introduced. By calculating the partial derivative of the target assembly dimension with respect to the error of a certain intermediate node, the error amplification link is identified, and the transmission path with high influence weight is recorded.

[0092] A preferred approach for introducing a local sensitivity analysis function is:

[0093]

[0094] Where S represents the overall assembly dimension output index, x i This represents the size variable of the i-th intermediate component. The sensitivity derivative represents the degree to which the output error responds to the intermediate variable.

[0095] Set a standard for judging the stability of the simulation, and set the result variance change to be less than a threshold. If the result variance change is less than the threshold, the simulation process will be terminated early.

[0096] If the variance change of the result is less than the threshold, automatically increase the number of sampling rounds to improve accuracy and confidence.

[0097] Output the critical path distribution histogram, error superposition curve, and extreme deviation distribution point set.

[0098] Furthermore, for the critical error path nodes identified in the assembly error propagation simulation, a local differential perturbation function is established, and the chain rule is used to calculate the response function value of the overall assembly accuracy to the node deviation, thereby generating an error sensitivity matrix.

[0099] A spatial distribution error mapping map is constructed to map nodal deviations to the outer contour of components or assembly contact surfaces, and an error heat map is generated through an interpolation algorithm.

[0100] When there are intermediate components in the path that are interfered with by fixed clamps, the direction and magnitude of the transmitted error are adjusted according to the clamp stiffness coefficient. It is then determined whether the additional error caused by the clamp deformation exceeds the set error. If it does, it is marked as a rigid shear point.

[0101] The dynamic error propagation process is tracked along the entire path, and the curve of error accumulation rate changing with path depth is recorded to identify the error mutation interval and controllable interval.

[0102] S4: Based on the simulation results, output the key assembly node deviation distribution and the impact index of cumulative error on assembly performance.

[0103] Based on the simulation output of assembly error propagation, the position offset of the final assembly node is calculated, and the Hausdorff distance between the position offset and the design datum shape is calculated as the contour deviation index based on the assembly target contour.

[0104] A preferred method for calculating the position offset is:

[0105]

[0106] Where A represents the assembly point cloud and model outline generated by simulation, B represents the reference outline generated by design drawings and CAD, a represents a single point in A, b represents a single point in B, ||·|| represents the Euclidean distance, sup represents the supremum of the distance between points in the set, inf represents the infremum of the distance between points in the set, and d represents the d-axis distance. H This represents the Hausdorff distance.

[0107] The surface-to-surface contact unevenness index and the bolt hole center drift distance are extracted as factors affecting connection quality.

[0108] An assembly stability function is introduced, defined as the variance variation trend of error distribution in multiple assemblies. Under different working conditions, the error is concentrated in a few paths, indicating the existence of an error aggregation bottleneck, which serves as an index of stability degradation.

[0109] Calculate the assembly reliability score.

[0110] A preferred method for calculating assembly reliability scores is as follows:

[0111]

[0112] Where R represents the assembly reliability index, Δ k T represents the actual deviation value of the k-th node. k This indicates the maximum allowable deviation for that node, and exp represents the exponential function.

[0113] S5: Using the influencing indicators as optimization inputs, construct a tolerance allocation optimization model under the target accuracy constraint.

[0114] Construct a multi-objective optimization problem that includes constraints and an objective function. The objective function is a weighted combination of minimizing the sum of squared assembly deviations, total manufacturing costs, and response delay penalties.

[0115] A preferred approach to constructing a multi-objective optimization problem that includes constraints and an objective function is as follows:

[0116]

[0117] Where F represents the objective function value, α, β, and γ represent the weighting coefficients of the multi-objective optimization, C represents the manufacturing cost index, and penalty represents the penalty term for delay and response failure.

[0118] It should be noted that the objective function value needs to be minimized.

[0119] The performance impact index of key nodes is used as the input boundary under constraints. A linear programming algorithm is used to solve the problem of uniform tolerance allocation for regular parts. The linear programming algorithm is enabled when the upper and lower tolerance limits are strictly symmetrical and the transmission path is linearly separable.

[0120] The constraints include the maximum permissible deviation, the upper limit of imbalance, and the lower limit of reliability.

[0121] For nonlinear error propagation paths, a genetic algorithm is introduced to perform adaptive optimization. Through crossover and mutation operations, a better tolerance combination is searched. The fitness function converges continuously for more than 10 generations, and the results are output in advance.

[0122] S6: Update the initial accuracy modeling parameters based on the tolerance allocation optimization model output.

[0123] The output tolerance assignment matrix is ​​reloaded as a weight term into the corresponding component tolerance field in the initial digital modeling dataset, and the update method is to fully cover the data items with better values ​​than the initial values.

[0124] Refresh the error propagation weights of the corresponding edges in the assembly dimension chain model, recalculate the sensitivity factors for edges with path weight changes, and update the full-link weight graph.

[0125] A preferred approach to updating the error propagation weights of corresponding edges in the assembly dimension chain model is as follows:

[0126]

[0127] in, W represents the error sensitivity coefficient from component i to j after feedback correction. ij This represents the error sensitivity coefficient from component i to j.

[0128] The re-simulation process is triggered based on the feedback results. When tolerance optimization leads to a significant decrease in the error of key nodes, the feedback is terminated.

[0129] If no significant decrease in error occurs, the system automatically enters a new round of modeling-simulation-optimization iteration process until the system error stabilizes within the preset accuracy window.

[0130] After the update is completed, the new modeling results will be marked as the current version, the old version will be retained for comparison and rollback, and synchronized to the digital assembly.

[0131] The above embodiments also include a high-precision construction simulation analysis system for digital assembly, specifically:

[0132] The system includes a multi-source data acquisition and modeling module, an error propagation modeling module, an assembly simulation analysis module, an assembly performance evaluation module, a tolerance optimization and control module, and a modeling parameter feedback correction module.

[0133] The multi-source data acquisition and modeling module includes acquiring multi-source heterogeneous data and constructing a digital precision modeling dataset of the assembly object.

[0134] The error propagation modeling module includes generating an assembly dimension chain model that describes the error propagation relationship between components, based on a digital precision modeling dataset.

[0135] The assembly simulation analysis module includes performing simulation analysis on the assembly dimension chain model and executing assembly error propagation simulation based on statistical methods.

[0136] The assembly performance evaluation module includes outputting the distribution of deviations at key assembly nodes and the impact of cumulative errors on assembly performance based on simulation results.

[0137] The tolerance optimization control module includes constructing a tolerance allocation optimization model under the target accuracy constraint by taking the influencing indicators as optimization inputs.

[0138] The modeling parameter feedback correction module includes updating the initial accuracy modeling parameters based on the tolerance allocation and optimized model output.

[0139] Example 2 is an embodiment of the present invention, which provides a high-precision construction simulation analysis method and system for digital assembly. In order to verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0140] This experiment, based on the proposed high-precision construction simulation analysis method for digital assembly, was experimentally verified in the actual component assembly preparation and error modeling stages. First, 10 sets of standard modular components were selected, derived from a typical assembly process path of an engineering project. Three-dimensional dimensional data, nominal tolerance ranges, and historical assembly error parameters of the components were collected based on existing design models and historical production records. Component numbers were automatically generated by the system and included in the data matrix of the digital model.

[0141] The three-dimensional dimensions of the components are obtained from the CAD model, and multi-point measurements are performed on key interfaces for confirmation. Nominal tolerances are obtained through dimension chain constraint node annotations in the engineering drawings. Historical assembly error information is extracted from the manufacturing execution system and historical quality inspection platform, including three core indicators: initial docking error, fixture positioning error, and residual assembly deformation. The collected data is further modeled uniformly through numbered batches, and field consistency and timestamp verification mechanisms are set. When missing items occur, data interpolation reconstruction methods are applied to handle them.

[0142] All data was converted to a standard format and input into the digital precision modeling system. A one-dimensional structural row vector was created, indexed by "component number," recording the corresponding attribute labels and error variables. The total deviation value was then calculated for each component, used as input for subsequent assembly dimensional chain error simulation models. To eliminate random error interference, at least three independent error samples were collected for each component in the experiment, and a weighted average was used in the modeling.

[0143] This experiment strictly followed the multi-source heterogeneous data acquisition framework proposed in the invention, and the data format was compatible with the subsequent error propagation model and sensitivity simulation requirements, fully verifying the engineering applicability and simulation input accuracy of the modeling method. The experimental data are shown in Table 1.

[0144] Table 1 Experimental Data

[0145]

[0146] By analyzing the assembly node data of the above 10 typical components, several key phenomena can be observed.

[0147] First, the actual three-dimensional dimensions of the components fluctuate around the nominal 100mm, with the maximum deviation controlled within ±2mm. However, significant differences are observed in the overall error accumulation. Although CMP-1003 and CMP-1009 have similar three-dimensional dimensions, due to differences in fixture errors and residual deformation factors, the total deviation of CMP-1009 reaches 0.73mm, far exceeding the average level. This phenomenon indicates that the traditional method relying on "dimensional and tolerance unit control" is insufficient to accurately depict the formation mechanism of assembly accuracy. In contrast, the multi-source error linkage modeling method proposed in this invention can effectively distinguish the causes of errors and provide precise input sources.

[0148] Secondly, in this experiment, data acquisition for all error dimensions was conducted through system-normalized structural modeling, avoiding the mismatch between historical process routes and the current modeling structure. For example, by constructing a 3D relationship diagram of "component-connection-fixture," a unified index was achieved between different data sources, greatly improving model stability. Traditional methods often rely on manual processing of design drawings and on-site measurement results, which is inefficient and makes it difficult to close the loop and trace data errors.

[0149] Finally, the one-to-one mapping structure between component numbers and measurement batches allows subsequent simulation analysis to trace the path of each error sample, ensuring high reproducibility. In subsequent error propagation simulations and sensitivity analyses, path weights and error distribution functions are supported by the data in this table, achieving high consistency in simulation input and enhancing simulation reliability.

[0150] More importantly, the distribution trend of the total deviation term reveals that deviation control is not only related to the three-dimensional dimensions themselves, but is also significantly affected by fixture stability and deformation response. This is precisely the key point neglected by traditional error modeling techniques. This invention, by constructing a sensitivity matrix and attribute label set, systematically expresses the mechanism by which assembly accuracy is driven by multiple factors. Compared to existing traditional methods that rely on empirical rules and error amortization, the modeling scheme provided by this invention has higher controllability and adaptability.

[0151] Example 3, an embodiment of the present invention, provides a high-precision construction simulation analysis system for digital assembly, including: a multi-source data acquisition and modeling module, an error propagation modeling module, an assembly simulation analysis module, an assembly performance evaluation module, a tolerance optimization control module, and a modeling parameter feedback correction module.

[0152] The multi-source data acquisition and modeling module includes acquiring multi-source heterogeneous data and constructing a digital precision modeling dataset of the assembly object.

[0153] The error propagation modeling module includes generating an assembly dimension chain model that describes the error propagation relationship between components, based on a digital precision modeling dataset.

[0154] The assembly simulation analysis module includes performing simulation analysis on the assembly dimension chain model and executing assembly error propagation simulation based on statistical methods.

[0155] The assembly performance evaluation module includes outputting the distribution of deviations at key assembly nodes and the impact of cumulative errors on assembly performance based on simulation results.

[0156] The tolerance optimization control module includes constructing a tolerance allocation optimization model under the target accuracy constraint by taking the influencing indicators as optimization inputs.

[0157] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0159] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0160] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0161] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A high-precision construction simulation analysis method for digital assembly, characterized in that, include: Collect multi-source heterogeneous data to construct a digital precision modeling dataset of assembly objects; Based on the digital precision modeling dataset, an assembly dimension chain model is generated to describe the error propagation relationship between components; The assembly dimension chain model is simulated and analyzed, and assembly error propagation simulation based on statistical methods is performed. Based on the simulation results, output the key assembly node deviation distribution and the impact of cumulative error on assembly performance. By using the influencing indicators as optimization inputs, a tolerance allocation optimization model under the target accuracy constraint is constructed. The initial accuracy modeling parameters are updated based on the output of the tolerance allocation optimization model.

2. The high-precision construction simulation analysis method for digital assembly as described in claim 1, characterized in that: The collection of multi-source heterogeneous data includes: Multi-source heterogeneous data includes the three-dimensional geometric dimensions, tolerance requirements, and historical assembly error information of the components in the assembly object; The three-dimensional geometric dimension data of the components in the assembly object are extracted from the assembly process database. Based on the CAD model, the geometric parameters of the key assembly interface and feature contour are extracted. At the same time, the initial dimension-tolerance mapping relationship table is constructed by combining the nominal tolerance range recorded in the design drawings. The measurement deviation data collected under similar process routes in the past are retrieved from the production history data platform, and an error data index structure with one-to-one correspondence is established based on the component number and assembly batch number. Measurement deviation data include initial docking error, fixture positioning error, assembly residual deformation value, and process measurement correction amount; Set judgment conditions to check the consistency of timestamps and data integrity. When missing fields or inconsistent items are found, trigger data completion logic to interpolate, reconstruct, and remove incomplete records. They are uniformly converted into a standardized format and stored as a digital precision modeling dataset.

3. The high-precision construction simulation analysis method for digital assembly as described in claim 2, characterized in that: The construction of the digital precision modeling dataset includes: Based on multi-source heterogeneous data, an attribute label set is established, and a data matrix structure indexed by component number is constructed. The tag set includes component identification number, spatial positioning reference system, dimensional variable group, tolerance tolerance group, historical deviation distribution, deformation response function and assembly sequence number; For the assembly relationships between components, an assembly constraint dual table is constructed based on the assembly contact surface type and tolerance fit type to record the geometric fit relationship and assembly sequence coupling information between two components. Establish a three-dimensional relationship diagram of components, connections, and fixtures to express the error propagation path between components, and mark the intensity of influence transmission through the accuracy sensitivity matrix of connection nodes; When a component has multiple error sources, a path weight adjustment factor is introduced, and the path weight is determined based on the least squares fitting result of the historical residuals.

4. The high-precision construction simulation analysis method for digital assembly as described in claim 3, characterized in that: The assembly dimension chain model includes: Based on the structural connection and tolerance fit characteristics between components, a directed graph of error propagation is constructed. Each node in the graph represents a component size variable, each edge represents a path for the propagation of size difference and error, and the weight of the edge represents the error sensitivity. The assembly dimension chain model introduces a node type marking mechanism to distinguish between rigid nodes, flexible nodes, and redundant nodes, and configures different error calculation methods for each node type; Set boundary conditions when constructing the transmission path; Boundary conditions include initial reference point locking, constraint propagation termination rules, and process tolerance limits; When there are multiple parallel assembly structures in the component assembly process, the assembly dimension chain model adopts a combined chain strategy to construct a set of parallel transmission paths and sets path synthesis rules to statistically fuse multi-path errors.

5. The high-precision construction simulation analysis method for digital assembly as described in claim 4, characterized in that: The simulation analysis includes: The error propagation path in the assembly dimension chain model is discretized, the number of simulation sampling rounds and the error disturbance range are set, and the possible assembly error state space is simulated by adding random disturbance samples to each path parameter. The Monte Carlo simulation method is adopted, and the input disturbance is probabilistically sampled by setting the error distribution type. In each iteration, the cumulative error output of the final assembly result is calculated. For highly coupled path nodes, a local sensitivity analysis function is introduced. By calculating the partial derivative of the target assembly dimension with respect to the error of a certain intermediate node, the error amplification link is identified, and the transmission path with high influence weight is recorded. Set a simulation stability judgment criterion, and set the result variance change to be less than a threshold. If the result variance change is less than the threshold, terminate the simulation process early. If the variance change of the result does not meet the threshold, automatically increase the number of sampling rounds to improve accuracy and confidence. Output the critical path distribution histogram, error superposition curve, and extreme deviation distribution point set.

6. The high-precision construction simulation analysis method for digital assembly as described in claim 5, characterized in that: The assembly error propagation simulation includes: For the critical error path nodes identified in the assembly error propagation simulation, a local differential perturbation function is established, and the chain rule is used to calculate the response function value of the overall assembly accuracy to the node deviation, thereby generating an error sensitivity matrix. Construct a spatial distribution error mapping map to map nodal deviations to the outer contour of components or assembly contact surfaces, and generate an error heat map through interpolation algorithms; When there are intermediate components in the path that are interfered with by fixed clamps, the direction and magnitude of the transmitted error are adjusted according to the clamp stiffness coefficient. It is determined whether the additional error caused by the deformation of the clamp exceeds the set error. If it exceeds, it is marked as a rigid shear point. The dynamic error propagation process is tracked along the entire path, and the curve of error accumulation rate changing with path depth is recorded to identify the error mutation interval and controllable interval.

7. The high-precision construction simulation analysis method for digital assembly as described in claim 6, characterized in that: The output indicators affecting assembly performance include: Based on the simulation output of assembly error propagation, the position offset of the final assembly node is calculated, and the Hausdorff distance between the position offset and the design datum shape is calculated as the contour deviation index based on the assembly target contour. Extract the surface-to-surface contact unevenness index and the bolt hole center drift distance as factors affecting connection quality; An assembly stability function is introduced, defined as the variance variation trend of error distribution in multiple assemblies. Under different working conditions, the error is concentrated in a few paths, indicating the existence of an error aggregation bottleneck, which serves as an indicator of stability degradation. Calculate the assembly reliability score.

8. The high-precision construction simulation analysis method for digital assembly as described in claim 7, characterized in that: The tolerance allocation optimization model includes: Construct a multi-objective optimization problem that includes constraints and an objective function. The objective function is a weighted combination of minimizing the sum of squared assembly deviations, total manufacturing costs, and response delay penalties. The performance impact index of key nodes is used as the input boundary under constraints. A linear programming algorithm is used to solve the problem of uniform tolerance allocation of regular parts. The linear programming algorithm is enabled when the upper and lower tolerance limits are strictly symmetrical and the transmission path is linearly separable. The constraints include the maximum permissible deviation, the upper limit of imbalance, and the lower limit of reliability; For nonlinear error propagation paths, a genetic algorithm is introduced to perform adaptive optimization. Through crossover and mutation operations, a better tolerance combination is searched. The fitness function converges continuously for more than 10 generations, and the results are output in advance.

9. The high-precision construction simulation analysis method for digital assembly as described in claim 8, characterized in that: The output update initial accuracy modeling parameters include: The output tolerance assignment matrix is ​​reloaded as a weight term into the corresponding component tolerance field in the initial digital modeling dataset, and the update method is to fully cover the data items with better values ​​than the initial values; Refresh the error propagation weights of the corresponding edges in the assembly dimension chain model, recalculate the sensitivity factors for edges with path weight changes, and update the full-link weight graph. The re-simulation process is triggered based on the feedback results. When tolerance optimization leads to a significant decrease in the error of key nodes, the feedback is terminated. If no significant decrease in error occurs, the system automatically enters a new round of modeling-simulation-optimization iteration process until the system error stabilizes within the preset accuracy window. After the update is completed, the new modeling results will be marked as the current version, the old version will be retained for comparison and rollback, and synchronized to the digital assembly.

10. A high-precision construction simulation analysis system for digital assembly, characterized in that: It includes a multi-source data acquisition and modeling module, an error propagation modeling module, an assembly simulation analysis module, an assembly performance evaluation module, a tolerance optimization and control module, and a modeling parameter feedback correction module; The multi-source data acquisition and modeling module includes acquiring multi-source heterogeneous data and constructing a digital precision modeling dataset of the assembly object. The error propagation modeling module includes generating an assembly dimension chain model that describes the error propagation relationship between components based on a digital precision modeling dataset. The assembly simulation analysis module includes performing simulation analysis on the assembly dimension chain model and executing assembly error propagation simulation based on statistical methods. The assembly performance evaluation module includes outputting the distribution of deviations at key assembly nodes and the impact indicators of cumulative errors on assembly performance based on simulation results. The tolerance optimization control module includes constructing a tolerance allocation optimization model under target accuracy constraints by taking the influencing indicators as optimization inputs. The modeling parameter feedback correction module includes updating the initial accuracy modeling parameters based on the tolerance allocation and optimized model output.

Images