Engine assembly bolt tightening multi-dimensional analysis method and system

Abstract

The invention provides an engine assembly bolt tightening multi-dimensional analysis method and system, and relates to the technical field of engine assembly, and the method comprises the steps: receiving key feature parameters subjected to data preprocessing and feature engineering processing; in the construction process of the key characteristic parameters, according to geometric structure parameters of the bolt, the three-dimensional shape of the bolt is equivalent to a regular prism for space volume calculation, and volume characteristic parameters are formed; performing data fusion processing on the volume characteristic parameters and the material mechanical properties of the bolt to construct a key characteristic parameter set for finite element simulation analysis; and performing a finite element analysis process based on the key characteristic parameter set, and obtaining a stress field distribution state of the bolt connection structure under a tightening working condition through numerical simulation calculation. According to the method, the accuracy and pertinence of bolt tightening performance analysis can be improved.

Description

Technical Field

[0001] This invention relates to the field of engine assembly technology, and in particular to a multi-dimensional analysis method and system for engine assembly bolt tightening. Background Technology

[0002] During engine assembly, the quality of bolt tightening directly affects the overall structural stability and operational reliability of the engine. The stress distribution and load-bearing characteristics analysis under bolt tightening conditions are key aspects to ensure assembly quality.

[0003] Existing engine bolt tightening analysis methods mostly rely on finite element simulation technology, but they suffer from the technical deficiency of single-dimensional analysis. For example, they do not deeply integrate bolt geometric features with material mechanical properties, and they do not conduct multi-dimensional grouped statistical analysis according to engine type, functional parts, and bolt specifications.

[0004] The aforementioned deficiencies lead to the following limitations in existing methods: for example, they can only obtain stress data under a single operating condition, and cannot establish the correlation between stress distribution and bolt geometric parameters and material properties, thus making it difficult to accurately reflect the differences in tightening performance of bolts of different models, locations, or specifications. Because of this limitation, the analysis results lack specificity and cannot provide accurate data support for bolt selection optimization and assembly process adjustments in different scenarios. This may lead to stress concentration hazards after bolt assembly, affecting engine operating safety and service life. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a multi-dimensional analysis method for the tightening of engine assembly bolts, which can reduce the risk of stress concentration and improve the safety and service life of engine operation.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] Firstly, a multi-dimensional analysis method for engine assembly bolt tightening, the method comprising:

[0008] Receive key feature parameters after data preprocessing and feature engineering; in the process of constructing key feature parameters, based on the geometric structural parameters of the bolt, use the prism volume calculation method to convert the three-dimensional shape of the bolt into a regular prism for spatial volume calculation, and form volume feature parameters;

[0009] The volumetric characteristic parameters and the material mechanical properties of the bolts are fused together to construct a set of key characteristic parameters for finite element simulation analysis.

[0010] Based on the key feature parameter set, a finite element analysis process is carried out to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation.

[0011] Stress concentration areas are identified based on the stress field distribution to locate critical stress areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius based on the nominal diameter; the equivalent bearing area of ​​the critical stress area is calculated using the reference calculation radius;

[0012] The equivalent bearing area and stress field distribution are fused to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set.

[0013] The statistical analysis process is applied to multidimensional grouped datasets to calculate the statistical characteristics of each group and generate multidimensional analysis results that include model comparison analysis, part difference analysis, and specification impact analysis.

[0014] Furthermore, the system receives key feature parameters after data preprocessing and feature engineering. During the construction of these key feature parameters, based on the bolt's geometric parameters, a prism volume calculation method is used to equate the bolt's three-dimensional shape to a regular prism for spatial volume calculation, forming volume feature parameters, including:

[0015] Obtain the original structural data of the engine bolts, including their geometric dimensions and material properties;

[0016] Based on geometric dimensions, the prism volume calculation method equates the complex shape of the bolt to a columnar geometric body with regular base surfaces;

[0017] Based on the base projection and axial length parameters of the columnar geometry, the spatial volume is calculated to obtain the volume characteristic parameters.

[0018] Furthermore, the volumetric characteristic parameters are fused with the material mechanical properties of the bolts to construct a set of key characteristic parameters for finite element simulation analysis, including:

[0019] Obtain the mechanical property parameters of the bolt material, including the elastic modulus and Poisson's ratio;

[0020] The volumetric characteristic parameters and mechanical performance parameters are parametrically correlated and integrated to form an integrated parameter set;

[0021] Based on the integrated parameter set, a key feature parameter set containing geometric features and material properties is constructed.

[0022] Furthermore, based on the key feature parameter set, a finite element analysis process is performed to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculations, including:

[0023] A finite element mesh model of the bolt connection structure is established based on the key feature parameter set.

[0024] In the finite element mesh model, identify the contact area between the bolt head and the connector;

[0025] Using the principle of sphere volume calculation, the equivalent radius is determined based on the characteristic dimensions of the contact area, and the equivalent contact volume is calculated by multiplying the cube of the equivalent radius by pi.

[0026] The equivalent contact volume is used as the initial boundary condition and imported into the finite element mesh model;

[0027] After importing the finite element mesh model, based on the initial boundary conditions, the load conditions corresponding to the tightening condition are configured for numerical simulation calculation to obtain the numerical simulation results.

[0028] The numerical simulation results are iteratively corrected and convergence verified to obtain the stress field distribution of the bolted connection structure.

[0029] Furthermore, stress concentration areas are identified based on the stress field distribution to locate critical load-bearing areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius using the nominal diameter as the basis; using the reference calculation radius, the equivalent bearing area of ​​the critical load-bearing area is calculated, including:

[0030] Based on the stress field distribution, identify and extract stress concentration regions;

[0031] Based on the preset stress threshold, stress concentration areas are screened to locate key stress areas;

[0032] After the key stress area is determined, the reference calculation radius for calculating the circular area is determined based on the nominal diameter in the bolt specification parameters;

[0033] Based on the baseline calculation radius, the circular area calculation method is adopted, and the initial bearing area is obtained by multiplying the square of the radius by pi.

[0034] By combining polygon approximation calculation method, the initial bearing area is geometrically corrected according to the actual contour of the key stress area to obtain the accurate equivalent bearing area;

[0035] The equivalent bearing area is associated with the corresponding key stress area to generate bearing area data with area identification.

[0036] Furthermore, the equivalent bearing area and stress field distribution are fused to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification, and bolt specification grade to form a multi-dimensional grouped dataset, including:

[0037] Data on the bearing area with regional identifiers is fused with the stress field distribution to establish a mapping relationship between stress values ​​and bearing area, generating a fused dataset.

[0038] Based on the fused dataset, the data is divided according to three dimensions: engine type, functional part classification, and bolt specification grade, to form several data groups;

[0039] Calculate the geometric distribution boundary of each data group in the parameter space;

[0040] Based on the geometric distribution boundary, the grouped data after division is integrated across dimensions to construct a multidimensional grouped data set with a hierarchical structure.

[0041] 7. Furthermore, a statistical analysis process is performed on the multidimensional grouped dataset to calculate the statistical characteristics of each group, generating multidimensional analysis results including model comparison analysis, part difference analysis, and specification influence analysis, including:

[0042] From the multidimensional grouped dataset, extract the stress distribution parameters and equivalent bearing area data for each group;

[0043] Based on stress distribution parameters and equivalent bearing area data, the statistical characteristics of each group are calculated, including the mean stress, standard deviation of stress, and area distribution characteristic value.

[0044] Based on statistical characteristics, a multi-dimensional comparative analysis is conducted, including comparison of stress distribution between machine models, comparison of load-bearing characteristics between parts, and comparison of the performance impact of specification levels, in order to obtain the results of the multi-dimensional comparative analysis.

[0045] Based on the results of multi-dimensional comparative analysis, a comprehensive analysis report is generated, which includes model comparison analysis, part difference analysis, and specification impact analysis.

[0046] Secondly, a multi-dimensional analysis system for engine assembly bolt tightening, the system performing the method described, including:

[0047] The receiving module is used to receive key feature parameters after data preprocessing and feature engineering. In the process of constructing key feature parameters, based on the geometric structural parameters of the bolt, the prism volume calculation method is used to convert the three-dimensional shape of the bolt into a regular prism for spatial volume calculation, thus forming volume feature parameters.

[0048] The module is used to fuse volumetric characteristic parameters with the material mechanical properties of bolts to construct a set of key characteristic parameters for finite element simulation analysis.

[0049] The analysis module is used to perform finite element analysis based on a set of key feature parameters, and to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation.

[0050] The calculation module is used to identify stress concentration areas based on the stress field distribution to locate key stress areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius based on the nominal diameter; and the equivalent bearing area of ​​the key stress area is calculated using the reference calculation radius.

[0051] The integration module is used to fuse the equivalent bearing area and stress field distribution to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set.

[0052] The execution module is used to perform statistical analysis on multidimensional grouped datasets to calculate the statistical characteristics of each group and generate multidimensional analysis results including model comparison analysis, part difference analysis, and specification impact analysis.

[0053] Thirdly, a computing device including a memory and a processor;

[0054] The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in the first aspect.

[0055] Fourthly, a computer-readable storage medium for storing a computer program for performing the method as described in the first aspect.

[0056] The above-described solution of the present invention has at least the following beneficial effects:

[0057] By employing a technical approach that integrates bolt geometric volume characteristics and material mechanical properties to construct a set of key feature parameters, accurately obtains stress field distribution and equivalent bearing area through finite element analysis combined with multiple algorithms, and performs multi-dimensional grouped statistical analysis according to engine type, functional parts, and bolt specifications, this approach effectively overcomes the technical problems of existing methods that lack specificity due to single-dimensional analysis and lack of deep integration of core attributes. It accurately reflects the differences in bolt tightening performance under different scenarios, providing precise data support for bolt selection optimization and assembly process adjustment, reducing the risk of stress concentration after bolt assembly, and effectively improving engine operating safety and service life. Attached Figure Description

[0058] Figure 1 A schematic diagram of a multi-dimensional analysis method for engine assembly bolt tightening;

[0059] Figure 2 This is a schematic diagram of a multi-dimensional analysis system for engine assembly bolt tightening.

[0060] Figure 3 This is a schematic diagram of a computing device; Detailed Implementation

[0061] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0062] An embodiment of the present invention proposes a multi-dimensional analysis method for engine assembly bolt tightening, the method comprising:

[0063] Step 1: Receive key feature parameters after data preprocessing and feature engineering; during the construction of key feature parameters, based on the geometric structural parameters of the bolt, the prism volume calculation method is used to convert the three-dimensional shape of the bolt into a regular prism for spatial volume calculation, thus forming volume feature parameters;

[0064] Step 2: The volumetric characteristic parameters and the material mechanical properties of the bolts are fused to construct a set of key characteristic parameters for finite element simulation analysis.

[0065] Step 3: Based on the key feature parameter set, perform finite element analysis to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation.

[0066] Step 4: Identify stress concentration areas based on the stress field distribution to locate key stress areas; determine the reference calculation radius based on the nominal diameter dimension in the bolt specification parameters using the circular area calculation method; calculate the equivalent bearing area of ​​the key stress area using the reference calculation radius.

[0067] Step 5: The equivalent bearing area and stress field distribution are fused to construct a fused dataset containing the relationship between stress and area; Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set.

[0068] Step 6: Perform statistical analysis on the multidimensional grouped data set to calculate the statistical characteristics of each group and generate multidimensional analysis results including model comparison analysis, part difference analysis, and specification influence analysis.

[0069] In this embodiment of the invention, by equating the three-dimensional morphology of the bolt to a regular prism to calculate volumetric characteristic parameters and deeply integrating them with material mechanical properties, a comprehensive and accurate set of key characteristic parameters is constructed, providing a solid data foundation for subsequent analysis. The stress field distribution under tightening conditions is accurately obtained using finite element analysis, and the equivalent bearing area of ​​key stress regions is accurately calculated using nominal diameter and polygon approximation methods, achieving precise matching analysis between stress distribution and bearing capacity. Through multi-dimensional data grouping, integration, and statistical analysis based on engine type, functional parts, and bolt specifications, the differences in bolt tightening performance under different scenarios are clearly presented. These designs effectively improve the comprehensiveness and accuracy of bolt tightening analysis, providing more targeted decision-making basis for bolt selection optimization and assembly process adjustment, effectively reducing the risk of stress concentration after bolt assembly, and further ensuring the stability of the engine's overall structure.

[0070] In a preferred embodiment of the present invention, step 1 above may include:

[0071] Step 1.1: Obtain the original structural data of the engine bolts. The original structural data includes geometric dimensions and material properties. Specifically, this involves: using a 3D scanning method with a scanning accuracy of less than or equal to 0.001 mm, setting scanning sections at fixed intervals of 5 mm along the entire length of the structure, starting from one end of the structure and ending at the other, performing a full-circumference scan section by section, and collecting the surface 3D coordinate data of each section in real time. A continuous 3D coordinate dataset is then formed through data stitching technology. For core dimension parameters, a multi-section repeated measurement method is used, uniformly selecting three independent measurement sections in the critical dimension area, with four symmetrically distributed measurement points set for each section. For each measurement point, measurements were repeated five times. After removing outliers exceeding three times the standard deviation, the arithmetic mean was taken to obtain accurate core dimension data. Material composition detection methods were used, and the elemental composition and mass fraction of each element were determined through spectral analysis. Combined with authoritative material standard manuals, basic attribute information such as density, yield strength, and elastic modulus of the material was supplemented. The three-dimensional coordinate data collected by scanning, the dimensional results of multi-section measurement, and the material attributes obtained by detection were classified and entered into the data management area to establish a complete original structural dataset containing three-dimensional morphological coordinates, key dimensional parameters, and basic material properties.

[0072] Step 1.2: Based on geometric dimensions, the prism volume calculation method is used to equate the complex shape of the bolt to a prism geometry with a regular base surface. Specifically, this includes: based on the dimensional information in the original structural data, dividing the overall structure into irregular contour regions, gradual transition regions, and regular contour regions according to differences in shape characteristics; clarifying the start and end coordinate ranges of each region through boundary threshold analysis of the 3D coordinate dataset; for the irregular contour regions, extracting the maximum outer contour dimension of the region from the 3D coordinate data, calculating its circumcircle diameter, and deriving the side length of the regular polygon base surface based on this diameter. The distance between opposite sides of the regular polygon is equal to the circumcircle diameter. Using the prism volume calculation method, first calculate the base surface area using the regular polygon area formula, which is the product of the square of the side length, the number of sides, and the cotangent value, divided by 2. Then, combine this with the axial length of the region to preliminarily calculate the volume, constructing a regular polygon base surface with a volume error ≤3% compared to the original region. For the gradient transition region, the cross-sectional dimensions at both ends of the region are extracted, the cross-sectional areas at both ends are calculated, and the average value is taken as the equivalent average cross-sectional area. The gradient volume calculation logic is used to multiply the average cross-sectional area by the axial length of the region to obtain the preliminary volume, and a gradient prism structure is constructed. The equivalent volume error is controlled within a preset threshold through multiple iterations. For the regular contour region, the average diameter obtained from multiple cross-section measurements is used as the diameter of the circular base surface. The area of ​​the base surface is calculated using the circle area formula. Combined with the axial length of the region, the preliminary volume is calculated using the prism volume algorithm, which multiplies the base surface area by the axial length to construct a columnar geometry with a circular base surface of equal diameter. During the equivalence process of each region, the axis of each segment of columnar geometry is calibrated based on the coordinates of the central axis of the original structure to ensure that the overlap error of the central axis of all equivalent segments is ≤0.1mm, and finally a complete equivalent combination form is formed.

[0073] Step 1.3: Based on the base projection and axial length parameters of the cylindrical geometry, calculate the spatial volume to obtain volume characteristic parameters. Specifically, this includes: first, setting the projection direction perpendicular to the central axis of each segment of the cylindrical geometry, controlling the projection angle error within ±0.5°, and then performing projection processing on the equivalent cylindrical geometry of each region. Through the two-dimensional projection transformation of the three-dimensional coordinate data, obtain the two-dimensional base contour coordinate data of each segment; for the regular polygon base, extract the coordinates of each vertex from the contour coordinate data, calculate the distance between adjacent vertices to obtain the side length, determine the number of sides of the regular polygon, and accurately calculate the base area using the regular polygon area formula, which is the square of the side length multiplied by the number of sides, multiplied by the cotangent of the ratio of π to the number of sides, and finally divided by 2; for the circular base, use the average diameter obtained from multi-section measurements. Based on this, the base area is calculated using the formula for the area of ​​a circle, which is π multiplied by the square of half the diameter. A multi-point length measurement method is employed, with four measurement points evenly placed at both ends and in the middle along the central axis coordinate direction of each segment. The axial coordinate values ​​of each point are extracted, and the distance between points is calculated using the coordinate differences. Each segment is measured three times, and the average value is taken as the precise axial length of that segment after removing outliers. Based on the core logic of the prism volume algorithm, the base area of ​​each segment is multiplied by its corresponding precise axial length to obtain the volume calculation value of each segment. The volume calculation values ​​of all segments are accumulated sequentially according to the start and end coordinates of each region. Then, the volume deviation of the transition area at the junction of each segment is corrected by verifying the spatial volume corresponding to the original three-dimensional coordinate dataset, ultimately obtaining precise volume characteristic parameters with an error ≤2%.

[0074] In this embodiment of the invention, by first acquiring the original structural data of the engine bolt, including its geometric dimensions and material properties, the integrity and accuracy of the data source for feature calculation are ensured. Then, based on the geometric dimensions, prism volume calculation is used to equate the complex shape of the bolt to a columnar geometry with regular base surfaces, effectively simplifying the volume calculation difficulty of complex structures and avoiding calculation errors caused by directly processing irregular shapes. Finally, based on the base surface projection and axial length parameters of the columnar geometry, the volume feature parameters are accurately calculated, ensuring the reliability and repeatability of the volume features. This provides solid and accurate geometric feature support for the deep integration of volume feature parameters and material mechanical properties, and the construction of key feature parameter sets.

[0075] In a preferred embodiment of the present invention, step 2 above may include:

[0076] Step 2.1: Obtain the mechanical property parameters of the bolt material, including the elastic modulus and Poisson's ratio. Specifically, this involves: selecting a standard specimen of the same material as the target object; using a tensile test to obtain mechanical property data; setting the initial preload force to 50 N; controlling the tensile rate at a strain rate of 0.001 / s; during the test, using a high-precision sensor to collect stress and strain data of the specimen in real time, recording a set of data every 0.01 s until the specimen reaches the elastic limit; identifying linear intervals in the collected stress and strain data; and selecting effective intervals where the stress and strain show a linear relationship. For the data segment, the least squares method was used for linear fitting, and the elastic modulus was calculated from the slope of the fitted line. For the shear test, Poisson's ratio data was obtained. The shear loading rate was set to 0.5 mm / min, and the transverse and longitudinal deformation data during the shear process were recorded in real time. The original Poisson's ratio value was derived from the ratio of the transverse to the longitudinal deformation. Five parallel tests were conducted for each mechanical property parameter. Outliers were identified and removed using the Grubbs criterion, and the arithmetic mean of the remaining valid data was calculated to ensure that the relative error between the elastic modulus and Poisson's ratio was controlled within 0.8%.

[0077] Step 2.2 involves parametrically associating and integrating the volumetric characteristic parameters and mechanical performance parameters to form an integrated parameter set. Specifically, this includes: assigning a unique 16-bit identifier to each target object, containing a feature code corresponding to key information such as material type and specification grade, serving as the unique benchmark for parameter association; normalizing the volumetric characteristic parameters and mechanical performance parameters according to their numerical ranges, mapping all parameter values ​​to the [0,1] interval to eliminate the influence of different dimensions; setting a data validity threshold, with the reasonable range for the elastic modulus determined according to general material standards, and the reasonable range for Poisson's ratio limited to 0.2 to 0.4, eliminating outliers exceeding three standard deviations by comparing parameter values ​​with their reasonable ranges; establishing a one-to-one parameter mapping relationship based on the identifier, precisely matching the volumetric characteristic parameters of each target object with their corresponding elastic modulus and Poisson's ratio, and integrating them into a well-structured parameter set containing complete association relationships according to the structure of identifier, geometric characteristic parameters, and material property parameters.

[0078] Step 2.3: Based on the integrated parameter set, construct a key feature parameter set containing geometric features and material properties. Specifically, this includes: establishing a parameter importance assessment system based on the core requirements of finite element simulation analysis; quantitatively scoring parameters based on their impact weight on simulation results and computational relevance; selecting core parameters with scores higher than a preset threshold; and eliminating redundant information irrelevant to the simulation calculation; sorting the selected parameters according to the finite element simulation calculation process, first arranging geometric feature parameters and then material property parameters, and within the same category, sorting them in the order they participated in the calculation; assigning a unique parameter ID to each parameter; constructing an index directory containing parameter ID, physical meaning, dimension type, accuracy level, and associated parameter IDs to clarify the attributes and interrelationships of each parameter; and using cross-validation to verify the relevance of the parameter set, randomly selecting 30% of parameter combinations for logical consistency checks to ensure that the matching between geometric feature parameters and material property parameters is consistent, while also verifying the rationality and accuracy of parameter values, ultimately forming a key feature parameter set that covers key geometric features and core material properties, with a complete structure and clear correlations.

[0079] In this embodiment of the invention, by acquiring core mechanical performance parameters such as the elastic modulus and Poisson's ratio of the bolt material, reliable material property support is provided for parameter fusion. Then, through a parameterized correlation method, volumetric characteristic parameters and mechanical performance parameters are systematically integrated, breaking the separation between geometric features and material properties. Finally, based on the integrated parameters, a key feature parameter set containing two types of core attributes is constructed, effectively making up for the shortcomings of existing methods that do not deeply integrate geometric and material properties. This provides a more comprehensive and accurate data foundation for finite element simulation analysis, improves the reliability and pertinence of simulation analysis, and provides key data support for the calculation of stress field distribution under bolt tightening conditions.

[0080] In a preferred embodiment of the present invention, step 3 above may include:

[0081] Step 3.1: Based on the key feature parameter set, establish a finite element mesh model of the bolted connection structure. This includes: extracting core parameters such as volumetric features, elastic modulus, and Poisson's ratio from the key feature parameter set; performing coordinate standardization according to the data specifications of the finite element analysis platform; verifying the dimensional consistency and numerical rationality of each parameter; eliminating data exceeding the standard range; setting quadrilateral elements for core structural regions and tetrahedral elements for complex interface regions; and dividing the mesh into three levels according to the structural stress sensitivity. The first-level gradient corresponds to the stress core area with a mesh size of 0.3 mm, the second-level gradient corresponds to the transition area with a mesh size of 1 mm, and the third-level gradient corresponds to the secondary structural area with a mesh size of 2 mm. After division, ensure that the interior angle of each element is between 30° and 150°, the aspect ratio is controlled within 3, and the warpage does not exceed 10°. Perform local densification or topology reconstruction on unqualified elements until all elements meet the quality requirements, and finally generate a finite element mesh model that meets the simulation accuracy.

[0082] The specific construction and training process of the finite element mesh model is as follows:

[0083] First, geometric and material property parameters of the bolted connection structure are extracted from the key feature parameter set. Geometric parameters include precise data such as volumetric features, nominal diameter, and boundary coordinates of key stress areas obtained through multi-step measurements and equivalent calculations. Material property parameters cover the elastic modulus and Poisson's ratio calibrated by tensile and shear tests. All parameters have undergone validity verification to ensure no outliers and dimensional consistency. Based on these parameters, a geometric model is built on the finite element analysis platform. The main structural form is first reconstructed according to the actual assembly relationship of the bolted connection, including the relative positions and contact relationships of the bolts and connectors. Then, the geometric model is preprocessed to remove redundant microstructures and chamfer features, preventing such details from affecting mesh generation quality and computational efficiency.

[0084] Subsequently, the mesh element type selection and partitioning scheme design were carried out. Considering the stress characteristics of bolted connection structures, quadrilateral elements were used for the core stress-bearing areas of the bolt body and connectors. This element type can better transmit stress and strain data and ensure calculation accuracy. Tetrahedral elements were used for complex contact areas and non-critical stress-bearing areas, balancing computational efficiency and structural adaptability. A three-level mesh gradient was set according to the structural stress sensitivity. The first-level gradient corresponds to the critical stress-bearing areas with a high risk of stress concentration, and the mesh size is controlled within 0.3 mm to ensure accurate capture of stress changes in this area. The second-level gradient corresponds to the transition area, and the mesh size is set to 1 mm to achieve a smooth transition between critical and non-critical areas. The third-level gradient corresponds to the secondary structural areas, and the mesh size does not exceed 2 mm to improve computational efficiency without affecting calculation accuracy.

[0085] After mesh generation, a comprehensive verification process begins, focusing on three core indicators: element interior angles, aspect ratio, and warpage. Element interior angles must be between 30° and 150°, aspect ratios must not exceed 3, and warpage must be controlled within 10°. Elements failing verification are then optimized. Defective elements in core load-bearing regions are corrected using localized refinement, while defective elements in transition and secondary regions are adjusted through topology reconstruction until all elements meet the quality requirements. This process ultimately generates a finite element mesh model that accurately reflects the geometric characteristics and material properties of the bolted connection structure.

[0086] After the model is built, the first step is to identify the contact area and configure boundary conditions. Based on the node coordinate data of the finite element mesh model, the contact identification algorithm is started with a fitting gap threshold of 0.01 mm. It traverses the surface nodes of the bolt head and the connector, and filters out continuous node areas with a spacing less than the threshold to determine the three-dimensional contour and coordinate range of the effective contact area. The equivalent contact volume is calculated using a sphere volume algorithm. The equivalent radius is determined by extracting the feature dimensions of the contact area. After correction, the accurate equivalent contact volume is obtained by combining it with the sphere volume calculation logic. This data is converted to a platform-compatible format and imported into the model, serving as the initial boundary conditions associated with the corresponding nodes of the contact area to ensure that the boundary conditions are consistent with the actual contact state.

[0087] Next, load conditions were configured and numerical simulations were performed. Based on the standard torque parameters of the bolt tightening process and the actual working conditions required for engine assembly, a weighted average method was used to determine the reference load value. A three-step loading mode was adopted to apply the load: in the preheating stage, 50% of the reference load was applied for 0.1s to allow the model to adapt to the load; in the incremental stage, the reference load was increased from 50% to 100% in a gradient of 0.1 N·m, with each step lasting 0.1s to avoid calculation distortion caused by sudden load changes; in the stabilization stage, the reference load was maintained at 100% for 0.2s to ensure that the stress distribution reached a stable state. An implicit solver was configured, a residual convergence criterion was selected, and a residual threshold of 1e to 4 was set. Numerical simulation calculations were started with a time step of 0.005s, and stress and strain data of each node were collected in real time.

[0088] After the simulation calculation is completed, iterative correction and convergence verification are carried out. The preliminary calculation results are compared with physical test data under the same parameter conditions. No less than 100 uniformly distributed data points are selected, and the deviation between the two is calculated using the mean square error method. If the deviation exceeds 3%, the mesh size of the core region is reduced by 0.1 mm, or the load gradient is adjusted and the calculation is repeated; if the deviation is less than 3%, the convergence verification stage begins. Energy convergence and displacement convergence are used for dual-index verification. First, the total energy change of the system is checked. When the energy change in three consecutive calculation steps is less than 1e to 5, the maximum displacement change of the key parts is checked to ensure it is less than 0.001 mm.

[0089] Repeat the iterative correction and convergence verification process described above until the deviation of the model calculation results is less than 1% and meets the double convergence index requirements. At this point, the model can accurately simulate the stress field distribution under bolt tightening conditions. Finally, solidify the optimized model parameters, boundary condition configurations, load application schemes, and convergence verification results to form a standardized finite element mesh model training process, ensuring that the model has good adaptability and reliability in subsequent analyses of bolted connection structures of different specifications and under different working conditions.

[0090] Step 3.2: In the finite element mesh model, identify the contact area between the bolt head and the connector. Specifically, this includes: based on the three-dimensional node coordinate data of the finite element mesh model, dividing the three-dimensional search mesh into 0.05mm intervals along the X, Y, and Z axes, setting the contact gap threshold to 0.01mm and starting the contact recognition algorithm, traversing layer by layer from the nodes on one side of the structural surface to the nodes on the other side of the structural surface, calculating the spatial straight-line distance between adjacent nodes and retaining four decimal places, filtering out nodes with a distance less than the threshold and marking them as potential contact nodes, statistically analyzing the continuous distribution range of potential contact nodes, and determining the effective contact area when the number of continuous nodes exceeds 50 and forms a closed contour, by extracting the coordinate data of the boundary nodes of this area to generate a three-dimensional contour coordinate matrix, clarifying the spatial range and morphological characteristics of the contact area to complete accurate identification and positioning.

[0091] Step 3.3 employs the principle of spherical volume calculation, determining the equivalent radius based on the characteristic dimensions of the contact area. The equivalent contact volume is then calculated using the product of the cube of the equivalent radius and pi. Specifically, this includes: selecting 36 measurement points at 10° intervals along the three-dimensional contour of the effective contact area, recording the three-dimensional coordinates of each point, calculating the spatial distance between adjacent measurement points, and extracting the maximum value from all distances as a reference for the lateral and longitudinal spans; using the standard deviation method to remove outliers exceeding ±5% of the average, and taking the arithmetic mean of the remaining values ​​as the initial equivalent radius reference value; and comparing the contact area contour with the standard spherical contour... To determine the degree of deviation, a contour correction factor is calculated, with the correction factor ranging from 0.95 to 1.05. The final equivalent radius is obtained by multiplying the initial equivalent radius reference value by the correction factor. Calculations are then performed according to the core logic of the sphere volume algorithm. First, the final equivalent radius is cubed, then the result is multiplied by pi, and finally multiplied by four-thirds to obtain the calculated value of the equivalent contact volume. Simultaneously, the verification volume value is obtained by calculating the product of the two-dimensional projected area of ​​the contact region and the average thickness. The difference between the calculated value and the verification volume value is compared. If the difference exceeds 3% of the verification volume value, the correction factor is readjusted and the calculation is repeated until the calculation error of the equivalent contact volume is controlled within 3%.

[0092] Step 3.4 involves importing the equivalent contact volume as the initial boundary condition into the finite element mesh model. Specifically, this includes: converting the equivalent contact volume data to an ASCII format compatible with the finite element analysis platform; setting the field delimiter of the data file to a tab character; arranging the data in the order of node ID, equivalent contact volume value, and action priority; uploading the data file through the platform's boundary condition import interface; starting the node association verification program; matching all nodes in the contact area one by one according to node ID to ensure that each node is associated with a unique equivalent contact volume parameter, with no omissions or duplicate associations; setting the action priority according to the stress importance of the contact area, dividing it into three levels: the core contact area is set to level one priority, the transition contact area to level two priority, and the edge contact area to level three priority; activating the boundary condition activation mechanism; visually viewing the scope of action and parameter matching of the boundary conditions through the platform's preview function; and completing the import and configuration of the initial boundary conditions after confirming that everything is correct.

[0093] Step 3.5: After importing the finite element mesh model, based on the initial boundary conditions, configure the load conditions corresponding to the tightening condition for numerical simulation calculation to obtain the numerical simulation results. Specifically, this includes: referring to the standard load range of the corresponding specification structure and combining the assembly process requirements, using the weighted average method to determine the reference load value for the tightening condition, and taking 1.05 times the midpoint of the range as the load reference; adopting a three-step loading mode, the first step is the preheating stage, applying 50% of the reference load for 0.1s; the second step is the incremental stage, applying... The load gradient is increased from 50% to 100% with a gradient of 0.1 N·m, each step lasting 0.1 s. The third step is the stabilization phase, maintaining the 100% reference load for 0.2 s. The load application area is determined using a coordinate positioning method, which is the central region of the stress-bearing end face of the structure. The central region is defined as a circular area with a radius of 5 mm centered at the center point of the end face. The load application direction is set to be completely coincident with the central axis of the structure, with a deviation not exceeding 0.5°. An implicit solver is configured, a residual convergence criterion is selected, and the residual threshold is set to 1e to 4. The calculation time step is set to 0.005 s according to the time step of load application. The numerical simulation program is started, and the stress and strain data of each node at each time step are collected in real time and stored as a structured result file.

[0094] Step 3.6 involves iteratively correcting and verifying the convergence of the numerical simulation results to obtain the stress field distribution of the bolted connection structure. Specifically, this includes: obtaining physical test stress distribution data under the same parameter conditions as a reference benchmark; extracting stress data from key areas and comparing it with the reference benchmark from the preliminary numerical simulation results; calculating the deviation value using the mean square error method; selecting no fewer than 100 uniformly distributed data points for calculation; if the deviation value exceeds 3%, reducing the core area mesh size by 0.1 mm or adjusting the load gradient and restarting the calculation; repeating the above correction process until the deviation value is less than 1%. The convergence verification program is then started. First, the total energy change of the calculation system is verified according to the energy convergence criterion. When the energy change in three consecutive calculation steps is less than 1e to 5, the maximum displacement change of key parts is verified according to the displacement convergence criterion. When the maximum displacement change is less than 0.001 mm, the convergence requirement is met, the iterative calculation is stopped, and a stress field distribution state file containing the coordinates of each node, the corresponding stress value, and the stress distribution gradient is output.

[0095] In this embodiment of the invention, a basic finite element analysis model is established based on a set of key feature parameters including geometric features and material properties, ensuring accurate mapping of the model to the actual structure. Then, by accurately identifying the contact area and calculating the equivalent contact volume using the principle of spherical volume calculation, the initial boundary conditions are set to better fit the actual contact state. Subsequently, the boundary conditions are imported and the load conditions corresponding to the tightening condition are configured to carry out numerical simulation calculations. The calculation results are optimized by combining iterative correction and convergence verification methods. This not only overcomes the simulation error problem caused by the rough setting of boundary conditions and incomplete consideration of parameters in existing methods, but also improves the reliability and accuracy of numerical simulation. Finally, the stress field distribution state of the bolted connection structure under the tightening condition can be accurately obtained.

[0096] In a preferred embodiment of the present invention, step 4 above may include:

[0097] Step 4.1: Based on the stress field distribution, identify and extract stress concentration regions. This includes: traversing all node stress data in the stress field distribution by dividing it into three-dimensional coordinate grids along the X, Y, and Z axes. The grid size for each partition is set to 0.5mm × 0.5mm × 0.5mm to ensure no omissions. Sort all node stress values ​​in descending order, calculate the stress difference between adjacent sorted nodes, and then divide by the spatial distance between nodes to obtain the stress gradient change rate. Set the gradient change rate threshold to 50MPa / mm, and select nodes with a change rate exceeding this threshold as candidate stress concentration nodes. Perform region clustering analysis on the candidate nodes. When the number of consecutive adjacent candidate nodes exceeds 80 and forms a closed contour, it is defined as a potential stress concentration region. Calculate the average stress value of all nodes in this region and compare it with the average stress value of the overall stress field. When the average stress value of the potential region reaches twice or more of the overall average stress value, it is officially determined as a stress concentration region. Extract all node coordinates, corresponding stress values, and boundary contour features of this region to form a complete stress concentration region dataset.

[0098] Step 4.2: Based on the preset stress threshold, stress concentration areas are screened to locate key stress areas. Specifically, this includes: retrieving the yield strength data of the bolt material from the acquired mechanical performance parameters, and taking 70% of this yield strength as the initial preset stress threshold; combining the actual working conditions of engine assembly, setting correction coefficients according to the stress level of different assembly parts, with a correction coefficient of 1.05 for core stress parts and 0.95 for secondary stress parts, and obtaining the final preset stress threshold by multiplying the initial threshold by the correction coefficient; comparing the maximum nodal stress value of each stress concentration area with the final preset stress threshold one by one, and screening out areas where the maximum stress value exceeds the threshold; calculating the area of ​​the screened area, which is obtained by summing the areas of the grid cells corresponding to all nodes in the statistical area, and then calculating the ratio with the total stress area of ​​the bolt (the two-dimensional projected area of ​​the stress end face of the bolt), retaining only areas with an area ratio of 5% or more, and completing the accurate location of key stress areas by extracting the boundary coordinates and stress distribution characteristics of these areas.

[0099] Step 4.3: After determining the key stress area, based on the nominal diameter in the bolt specification parameters, determine the reference calculation radius for calculating the circular area. This includes: extracting the nominal diameter data of the corresponding structure from the key feature parameter set. This data is the arithmetic mean of outliers after multi-section measurement in Step 1. First, compare the nominal diameter data with the original measured diameter data of each section to ensure that the deviation does not exceed 0.001mm, thus completing the accuracy verification. According to the basic logic of the circular area algorithm, divide the verified nominal diameter value by 2 to obtain the initial calculation radius. Analyze the offset between the center coordinates of the key stress area and the central axis of the structure. If the offset is within 0.1mm, keep the initial calculation radius unchanged. If the offset exceeds 0.1mm, fine-tune the initial calculation radius by 10% of the offset. The adjusted radius should not exceed 50% ± 1% of the nominal diameter. Finally, determine the reference calculation radius for area calculation.

[0100] Step 4.4: Based on the baseline calculation radius, a circular area algorithm is used to obtain the initial bearing area by multiplying the square of the radius by pi. Specifically, this includes: using the determined baseline calculation radius as the core input parameter, retaining six significant digits of the radius value to ensure calculation accuracy; performing the calculation according to the specific principles of the circular area algorithm: first, squaring the baseline calculation radius, retaining eight decimal places to avoid precision loss; second, selecting a commonly used industry standard value for pi to ensure consistency in the calculation basis; third, multiplying the radius squared result by the pi value, performing the operations from left to right to obtain the initial bearing area calculation result; retrieving the standard bearing area range for structures of the same specifications, which is obtained by consulting industry technical manuals and combining extensive experimental data statistics; comparing the initial bearing area with this range; if the calculation result is within the range, it passes verification; if it exceeds the range, returning to step 4.3 to readjust the baseline calculation radius until the result meets the requirements.

[0101] Step 4.5 involves geometrically correcting the initial bearing area based on the actual contour of the key stress region using a polygon approximation calculation method to obtain an accurate equivalent bearing area. Specifically, this includes: extracting the boundary node coordinate data of the key stress region; prioritizing nodes with a curvature change rate exceeding 10° / mm on the boundary as key sampling points; and then supplementing sampling points on the remaining boundary at uniform intervals of 0.1mm to ensure that the sampling points fully cover the region contour without redundancy; using a polygon approximation calculation method, connecting all sampling points sequentially in a clockwise direction, while determining the angle between three adjacent points; if the angle is between 175° and 185°, discarding intermediate nodes to simplify the process. Polygonal structure; the simplified closed polygon is divided into several non-overlapping triangles according to the vertex order. The side length is calculated using the vertex coordinates of each triangle, and then the area of ​​each triangle is calculated. The sum of the areas of all triangles is used to obtain the actual area of ​​the polygon. The difference between the actual area of ​​the polygon and the obtained initial bearing area is calculated to obtain the area deviation value. The correction coefficient is obtained by dividing the actual area of ​​the polygon by the initial bearing area. If the deviation value accounts for more than 2% of the actual area of ​​the polygon, the sampling interval is adjusted to 0.05mm and resampling is performed. If the deviation ratio is within 2%, the initial bearing area is multiplied by the correction coefficient to obtain the accurate equivalent bearing area.

[0102] Step 4.6: Associate and mark the equivalent bearing area with the corresponding key stress-bearing areas to generate bearing area data with area identifiers. Specifically, this includes: assigning position numbers in a clockwise direction from the center to the edge according to the distribution of key stress-bearing areas on the structure, with the numbers increasing sequentially from 1; classifying stress levels based on the maximum stress value of the key stress-bearing areas: Level 1 for maximum stress values ​​between 70% and 80% of the yield strength, Level 2 for between 80% and 90%, and Level 3 for above 90%, corresponding to stress level codes A, B, and C; and generating dimensional specification codes based on the structure's specifications, linking the position number, stress level code, and dimensional specification code. The inch specification codes are combined to form a unique regional identifier code; a hash table structure is established to map the regional identifier code as the key and core information such as equivalent bearing area, regional boundary coordinates, maximum stress value, and average stress value as the values ​​for associated storage; the data is organized in CSV format, with fields in the following order: regional identifier code, equivalent bearing area, boundary coordinate set, maximum stress value, and average stress value, and the data format of each field is uniform and standardized; 10% of the regional identifier codes are randomly selected for retrieval and verification to ensure that all corresponding associated data can be quickly retrieved, and finally, bearing area data with regional identifiers and accurate association relationships are generated.

[0103] In this embodiment of the invention, by accurately identifying and extracting stress concentration areas based on the stress field distribution, and combining this with a preset stress threshold to screen out key stress areas, the problem of existing methods being unable to accurately locate core stress-bearing parts is solved. After the key stress areas are determined, the baseline calculation radius of the circular area algorithm is determined based on the nominal diameter, and the initial bearing area is obtained by multiplying the radius squared by pi. At the same time, a polygon approximation algorithm is combined to perform geometric correction based on the actual contour, which effectively improves the accuracy of the equivalent bearing area calculation and overcomes the calculation deviation caused by the single algorithm ignoring contour differences. Finally, the equivalent bearing area and the key stress areas are associated and marked to generate bearing area data with region identification, providing a precise region and area correspondence for multi-dimensional data fusion and group analysis. This allows subsequent analysis steps to more specifically reflect the bolt bearing characteristics under different scenarios, providing reliable core data support for bolt selection optimization and assembly process adjustment, and further reducing the risk of stress concentration.

[0104] In a preferred embodiment of the present invention, step 5 above may include:

[0105] Step 5.1 involves fusing the bearing area data with regional identifiers with the stress field distribution status to establish a mapping relationship between stress values ​​and bearing area, generating a fused dataset. Specifically, this includes: using the regional identifier code as a unified association benchmark, which includes location sequence number, stress level code, and size specification code to ensure the uniqueness of data association; extracting the equivalent bearing area, regional boundary coordinate set, and regional identifier code from the bearing area data; simultaneously extracting the stress values, 3D coordinates of each node, and corresponding regional identifiers from the stress field distribution status; matching the two sets of data one by one according to the regional identifier code; calibrating the node coordinates of the matched data using the least squares method; calculating the coordinate deviation between the bearing area data and stress data under the same regional identifier; and removing discrete data with deviations exceeding 0.1 mm; repeatedly matching and verifying the calibrated valid data to ensure that the equivalent bearing area corresponding to each regional identifier corresponds one-to-one with the stress values ​​of all nodes within that region; integrating the data according to a fixed structure of regional identifier code, equivalent bearing area, 3D coordinates of nodes, corresponding stress values, and stress distribution gradient to establish a precise mapping relationship between stress values ​​and bearing area, ultimately generating a fused dataset without data conflicts and with clear association logic.

[0106] Step 5.2: Based on the fused dataset, the data is divided according to three dimensions: engine type, functional component classification, and bolt specification grade, to form several data groups. Specifically, this includes: First, establishing a three-level classification standard system. The first level of classification defines engine type based on engine power rating and cylinder count. Power ratings are divided into 1.0L and below, 1.0 to 2.0L, and above 2.0L. Cylinder counts are divided into 3-cylinder, 4-cylinder, and 6-cylinder and above. The combination of these two categories forms the basis for engine type classification. The second level of classification determines functional component classification based on the functional attributes of the bolt assembly parts, specifically into crankshaft connection parts, cylinder head sealing parts, transmission connection parts, auxiliary support parts, oil circuit sealing parts, etc. The components correspond to clearly defined assembly functional requirements; the three-level classification defines the specification level based on the nominal diameter and strength grade of the bolts. The nominal diameter is divided into three categories: below 10mm, 10 to 16mm, and above 16mm, and the strength grade is divided into three categories: 8.8, 10.9, and 12.9. Based on this standard system, the fused dataset is hierarchically divided. First, datasets with the same combination are selected according to the model category to form a first-level group. Then, within the first-level group, it is divided into second-level subgroups according to functional parts. Finally, within the second-level subgroups, it is further subdivided into third-level groups according to specification level. Each third-level group is named using a combination of model code, functional part code, and specification code. Cross-checking ensures that all data is assigned to the corresponding group without omissions or duplications.

[0107] Step 5.3 calculates the geometric distribution boundary of each data group in the parameter space, specifically including: for each third-level data group, extracting the equivalent bearing area and the maximum stress value of the corresponding region, using both as the horizontal and vertical coordinates of the two-dimensional parameter space to construct the coordinate point set of that group; preprocessing the coordinate point set by first calculating the mean and standard deviation of all coordinate points, using the Grubbs criterion to calculate the outlier threshold, identifying coordinate points that deviate from the mean by more than 3 times the standard deviation as outliers and removing them, retaining valid coordinate points that conform to the data distribution pattern; performing boundary calculations by first selecting the point corresponding to the minimum x-axis (equivalent bearing area) from the valid coordinate points as the initial pole, and if there are multiple minimum x-axis points, selecting the minimum y-axis (maximum stress value) point. Point; taking the initial pole as the origin, calculate the vectors of all other points to the initial pole, and solve the angle between each vector and the positive x-axis using the arctangent function. Sort the points counterclockwise in ascending order of the angles. Connect the sorted points in sequence, and calculate the cross product of the vectors formed by three adjacent points. If the cross product is positive, it means that the next point is outside the current convex hull boundary, and the point is retained. If the cross product is negative, it means that the point is inside the convex hull, and it is discarded. If the cross product is zero, it means that the three points are collinear, and the point farthest from the initial pole is retained. After traversing all sorted points, connect the last point to the initial pole to form a closed convex polygon. The outline of this convex polygon is the geometric distribution boundary of the data group in the two-dimensional parameter space, completely enclosing the entire distribution range of the data group.

[0108] Step 5.4: Based on the geometric distribution boundaries, perform cross-dimensional integration of the divided grouped data to construct a hierarchical multidimensional grouped data set. Specifically, this includes: first, defining the parameter units: the equivalent bearing area unit is mm², and the maximum stress value unit is MPa; setting 0.5 parameter units means either an equivalent bearing area of ​​0.5 mm² or a maximum stress value of 0.5 MPa; calculating the overlap and shortest distance between the geometric distribution boundaries of different dimensional groups; the overlap is obtained by the ratio of the intersection area to the union area of ​​two boundary convex polygons; the shortest distance is obtained by calculating the minimum straight-line distance between all pairs of points on the two boundaries; when the overlap exceeds 30% or the shortest distance is less than 0.5 parameter units, the group is considered to have strong correlation; and constructing a three-level hierarchical framework, with the engine model category as the core of the first level, and each model category corresponding to a unique first-level index. The system employs a multi-level index system. The second-level index is categorized by functional components under the first-level index, with each functional component corresponding to a unique second-level index code, associated with the first-level index code. The third-level index is categorized by bolt specification grade under the second-level index, with each specification grade corresponding to a unique third-level index code, also associated with the second-level index code. Data from different strongly correlated levels are linked and mounted. Under each level index, the corresponding geometric distribution boundary information, core parameter range, and data volume are labeled. A hierarchical index directory is established, clearly specifying the index codes for each level, the corresponding group name, the set of boundary coordinates, the parameter value range, and the associated lower-level index codes. Data traceability verification is performed by randomly selecting 20% ​​of the level nodes to ensure that each node from the first to the third level can quickly retrieve the corresponding complete data. Ultimately, this constructs a multi-dimensional grouped data set that is hierarchically distinct, tightly correlated, and highly traceable.

[0109] In this embodiment of the invention, by deeply fusing the load-bearing area data with regional identifiers and the stress field distribution state, a precise mapping relationship between stress values ​​and load-bearing area is established, solving the problems of scattered and loosely correlated data in the original method. The generated fused dataset achieves the organic unity of key mechanical parameters and geometric load-bearing parameters. Furthermore, the data is divided based on three core dimensions: engine type, functional part classification, and bolt specification grade, breaking the limitations of single-dimensional analysis in existing methods and making the data grouping more in line with actual application scenarios. By accurately calculating the geometric distribution boundary of each data group in the parameter space, the value range and distribution characteristics of each group of data are clarified, providing a clear boundary basis for cross-dimensional integration. Finally, based on the geometric distribution boundary, the grouped data is integrated across dimensions to construct a multi-dimensional grouped data set with a hierarchical structure. This ensures the clarity of data classification and realizes the correlation and interoperability of data from different dimensions, providing structured and systematic data support for conducting bolt tightening performance difference analysis, precise selection optimization, and assembly process adjustment in multiple scenarios.

[0110] In a preferred embodiment of the present invention, step 6 above may include:

[0111] Step 6.1: Extract stress distribution parameters and equivalent bearing area data for each group from the multidimensional grouped data set. Specifically, this includes: traversing the three-level hierarchical structure of the multidimensional grouped data set sequentially, first entering the first-level engine model category level, then progressively entering the second-level functional part classification level and the third-level bolt specification level, ensuring coverage of all independent data groups; for each third-level data group, extract stress distribution parameters, including the maximum stress value, average stress value, stress gradient change rate, and distribution range of stress concentration areas for all nodes within the group; simultaneously extract equivalent bearing area data, including the equivalent bearing area value, area calculation error, correction coefficient, and corresponding area boundary coordinates; perform integrity verification on the extracted data, checking for missing parameters or abnormal values. If missing data is found, return to the previous step to re-extract; if abnormal values ​​are found, remove them according to the Grubbs criterion, ultimately forming a subset of stress distribution parameters and a subset of equivalent bearing area parameters for each group.

[0112] Step 6.2: Based on the stress distribution parameters and equivalent bearing area data, calculate the statistical characteristics of each group, including the stress mean, stress standard deviation, and area distribution characteristic value. Specifically, this includes: for each subset of stress distribution parameters in each group, collecting stress value data from all nodes, first removing abnormal stress values ​​exceeding three times the standard deviation, then calculating the arithmetic mean of the remaining effective stress values ​​to obtain the stress mean of that group; based on the stress mean, calculating the difference between each effective stress value and the mean, squaring all differences and taking the arithmetic mean, then taking the square root of the result to obtain the stress standard deviation, which reflects the dispersion of the stress distribution; for the equivalent bearing area data, calculating the arithmetic mean of all equivalent bearing area values ​​within the group as the area mean, then calculating the area standard deviation, dividing the area standard deviation by the area mean to obtain the area coefficient of variation, and combining the area mean and the area coefficient of variation as the area distribution characteristic value to comprehensively characterize the central tendency and dispersion characteristics of the equivalent bearing area; all statistical characteristic calculations are performed with four decimal places to ensure that the calculation accuracy meets the analysis requirements.

[0113] Step 6.3: Based on statistical characteristics, conduct a multi-dimensional comparative analysis, including stress distribution comparison between engine models, load-bearing characteristics comparison between parts, and performance impact comparison of specification levels, to obtain multi-dimensional comparative analysis results. Specifically, this includes: comparing stress distribution between engine models by selecting different engine models with the same functional parts and bolt specification levels, comparing the mean stress and standard deviation of each group, calculating the difference and ratio of the same parameter between different engine models, analyzing the consistency and degree of difference in stress distribution, and summarizing the influence of engine operating conditions on bolt stress distribution by combining information such as engine power rating and cylinder count; comparing load-bearing characteristics between parts by selecting the same engine model and... The bolt specifications and grades are grouped into different functional parts. The area distribution characteristics and equivalent bearing area values ​​of each group are compared to analyze the differences in bearing capacity of different functional areas such as core transmission parts and auxiliary support parts, the assembly stress requirements of related parts, and the correspondence between functional attributes and bearing characteristics. The performance impact of specifications and grades is compared. Different specifications and grades of the same engine model and functional parts are grouped together. The stress distribution parameters and area distribution characteristics of each group are compared to analyze the impact of changes in nominal diameter and strength grade on bolt stress state and bearing capacity. The correlation between specification parameters and performance indicators is clarified through trend fitting. Finally, the three types of comparison results are integrated to form a multi-dimensional comparative analysis dataset.

[0114] Step 6.4: Based on the results of the multi-dimensional comparative analysis, generate a comprehensive analysis report including model comparison analysis, part difference analysis, and specification impact analysis. Specifically, this includes: organizing the multi-dimensional comparative analysis dataset according to a pre-defined report framework; the main body of the report is divided into three core modules: model comparison analysis, part difference analysis, and specification impact analysis; the model comparison analysis module includes a stress mean comparison table for different models, a stress standard deviation trend chart, and an analysis of the reasons for differences, clarifying the optimal range and suitable scenarios for bolt stress distribution under different model operating conditions; the part difference analysis module presents a comparison of the characteristic values ​​of the bearing area of ​​each functional part, a ranking of bearing capacity, and related parameters. The report prioritizes key stress-bearing components and proposes targeted load-bearing optimization directions based on assembly functional requirements. The specification impact analysis module provides bolt specification selection recommendations for different scenarios through correlation curves between specification levels and performance indicators, and parameter sensitivity analysis results. Data traceability identifiers are added to the report, with each analysis conclusion corresponding to specific data groups and original parameters. Terminology explanations and analysis method notes are also included to ensure readability and traceability. Ultimately, a comprehensive analysis report with a complete structure, detailed data, and clear conclusions is formed, providing direct reference for engine bolt selection optimization, assembly process adjustment, and reliability assessment.

[0115] In this embodiment of the invention, by accurately extracting the stress distribution parameters and equivalent bearing area core data of each group from a multi-dimensional grouped dataset, targeted and highly correlated basic data support is provided for subsequent analysis. By calculating statistical features such as the mean stress, standard deviation of stress, and characteristic values ​​of area distribution, the scattered raw data is transformed into key indicators with quantitative representation significance, improving the analyzability and representativeness of the data. Based on the statistical features, a multi-dimensional comparative analysis of stress distribution between engine models, bearing characteristics between parts, and the performance impact of specification levels is carried out, breaking the limitations of single-dimensional analysis and comprehensively revealing the differences and patterns of the mechanical properties of bolts under different scenarios. Finally, a comprehensive analysis report covering engine model comparison, part differences, and specification impact is generated, transforming complex analysis results into clear and practical references. This provides comprehensive and accurate decision support for the selection and optimization of engine bolts, adjustment of assembly processes, and suitability assessment under different working conditions, effectively reducing the risk of stress concentration during bolt use and improving the overall reliability and stability of engine assembly.

[0116] A multi-dimensional analysis system for engine assembly bolt tightening includes:

[0117] The receiving module is used to receive key feature parameters after data preprocessing and feature engineering. In the process of constructing key feature parameters, based on the geometric structural parameters of the bolt, the prism volume calculation method is used to convert the three-dimensional shape of the bolt into a regular prism for spatial volume calculation, thus forming volume feature parameters.

[0118] The module is used to fuse volumetric characteristic parameters with the material mechanical properties of bolts to construct a set of key characteristic parameters for finite element simulation analysis.

[0119] The analysis module is used to perform finite element analysis based on a set of key feature parameters, and to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation.

[0120] The calculation module is used to identify stress concentration areas based on the stress field distribution to locate key stress areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius based on the nominal diameter; and the equivalent bearing area of ​​the key stress area is calculated using the reference calculation radius.

[0121] The integration module is used to fuse the equivalent bearing area and stress field distribution to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set.

[0122] The execution module is used to perform statistical analysis on multidimensional grouped datasets to calculate the statistical characteristics of each group and generate multidimensional analysis results including model comparison analysis, part difference analysis, and specification impact analysis.

[0123] The multidimensional analysis system according to embodiments of the present invention can correspond to the execution of the methods described in the embodiments of the present invention, and the above and other operations and / or functions of each module of the multidimensional analysis system are respectively for implementing Figure 1 The corresponding process of the method in the illustrated embodiment will not be described in detail here for the sake of brevity.

[0124] This application also provides a computing device. This computing device can utilize a server.

[0125] like Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0126] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0127] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0128] Communication interface 703 is used for external communication. Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD). Memory 704 stores executable code, which processor 702 executes to perform the aforementioned multi-dimensional analysis method for engine assembly bolt tightening.

[0129] Specifically, in implementing the embodiment of the engine assembly bolt tightening multi-dimensional analysis system described above, and where each module or unit of the engine assembly bolt tightening multi-dimensional analysis system described above is implemented by software, the software or program code required to execute the functions of each module / unit in the engine assembly bolt tightening multi-dimensional analysis system described above can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704 to execute the aforementioned engine assembly bolt tightening multi-dimensional analysis method.

[0130] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned multi-dimensional analysis method for tightening engine assembly bolts.

[0131] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0132] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0133] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods of the multi-dimensional analysis method for engine assembly bolt tightening. The computer program product can be a software installation package; when any of the aforementioned methods of the multi-dimensional analysis method for engine assembly bolt tightening needs to be used, the computer program product can be downloaded and executed on the computer.

[0134] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multi-dimensional analysis method for engine assembly bolt tightening, characterized in that, The method includes: Receive key feature parameters after data preprocessing and feature engineering; in the process of constructing key feature parameters, based on the geometric structural parameters of the bolt, the spatial volume is calculated by equating the three-dimensional shape of the bolt to a regular prism, thus forming volume feature parameters; The volumetric characteristic parameters and the material mechanical properties of the bolts are fused together to construct a set of key characteristic parameters for finite element simulation analysis. Based on the key feature parameter set, a finite element analysis process is carried out to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation. Stress concentration areas are identified based on the stress field distribution to locate critical stress areas; the reference calculation radius is determined based on the nominal diameter dimension in the bolt specification parameters; and the equivalent bearing area of ​​the critical stress area is calculated using the reference calculation radius. The equivalent bearing area and stress field distribution are fused to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set. The statistical analysis process is applied to multidimensional grouped datasets to calculate the statistical characteristics of each group and generate multidimensional analysis results that include model comparison analysis, part difference analysis, and specification impact analysis.

2. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 1, characterized in that, The system receives key feature parameters after data preprocessing and feature engineering. During the construction of these key feature parameters, based on the bolt's geometric parameters, a prism volume calculation method is used to equate the bolt's three-dimensional shape to a regular prism for spatial volume calculation, forming volume feature parameters, including: Obtain the original structural data of the engine bolts, including their geometric dimensions and material properties; Based on geometric dimensions, the complex shape of the bolt is equivalent to a columnar geometry with regular base surfaces; Based on the base projection and axial length parameters of the columnar geometry, the spatial volume is calculated to obtain the volume characteristic parameters.

3. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 2, characterized in that, Volumetric characteristic parameters and bolt material mechanical properties are fused to construct a set of key characteristic parameters for finite element simulation analysis, including: Obtain the mechanical property parameters of the bolt material, including the elastic modulus and Poisson's ratio; The volumetric characteristic parameters and mechanical performance parameters are parametrically correlated and integrated to form an integrated parameter set; Based on the integrated parameter set, a key feature parameter set containing geometric features and material properties is constructed.

4. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 3, characterized in that, Based on a set of key feature parameters, a finite element analysis process is used to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculations, including: A finite element mesh model of the bolt connection structure is established based on the key feature parameter set. In the finite element mesh model, identify the contact area between the bolt head and the connector; Using the principle of sphere volume calculation, the equivalent radius is determined based on the characteristic dimensions of the contact area, and the equivalent contact volume is calculated by multiplying the cube of the equivalent radius by pi. The equivalent contact volume is used as the initial boundary condition and imported into the finite element mesh model; After importing the finite element mesh model, based on the initial boundary conditions, the load conditions corresponding to the tightening condition are configured for numerical simulation calculation to obtain the numerical simulation results. The numerical simulation results are iteratively corrected and convergence verified to obtain the stress field distribution of the bolted connection structure.

5. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 4, characterized in that, Stress concentration areas are identified based on the stress field distribution to locate key stress areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius based on the nominal diameter; Calculate the equivalent bearing area of ​​key stress-bearing regions using the baseline calculation radius, including: Based on the stress field distribution, identify and extract stress concentration regions; Based on the preset stress threshold, stress concentration areas are screened to locate key stress areas; After the key stress area is determined, the reference calculation radius for calculating the circular area is determined based on the nominal diameter in the bolt specification parameters. Based on the baseline calculation radius, the circular area calculation method is adopted, and the initial bearing area is obtained by multiplying the square of the radius by pi. By combining polygon approximation calculation method, the initial bearing area is geometrically corrected according to the actual contour of the key stress area to obtain the accurate equivalent bearing area; The equivalent bearing area is associated with the corresponding key stress area to generate bearing area data with area identification.

6. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 5, characterized in that, The equivalent bearing area and stress field distribution are fused to construct a fused dataset containing the relationship between stress and area. Based on the fused dataset, the data was grouped and integrated according to multiple dimensions such as engine type, functional part classification, and bolt specification grade, forming a multi-dimensional grouped dataset, including: Data on the bearing area with regional identifiers is fused with the stress field distribution to establish a mapping relationship between stress values ​​and bearing area, generating a fused dataset. Based on the fused dataset, the data is divided according to three dimensions: engine type, functional part classification, and bolt specification grade, to form several data groups; Calculate the geometric distribution boundary of each data group in the parameter space; Based on the geometric distribution boundary, the grouped data after division is integrated across dimensions to construct a multidimensional grouped data set with a hierarchical structure.

7. The multi-dimensional analysis method for engine assembly bolt tightening according to claim 6, characterized in that, A statistical analysis workflow is performed on multidimensional grouped datasets to calculate the statistical characteristics of each group, generating multidimensional analysis results that include model comparison analysis, part difference analysis, and specification impact analysis, including: From the multidimensional grouped dataset, extract the stress distribution parameters and equivalent bearing area data for each group; Based on stress distribution parameters and equivalent bearing area data, the statistical characteristics of each group are calculated, including the mean stress, standard deviation of stress, and area distribution characteristic value. Based on statistical characteristics, a multi-dimensional comparative analysis is conducted, including comparison of stress distribution between machine models, comparison of load-bearing characteristics between parts, and comparison of the performance impact of specification levels, in order to obtain the results of the multi-dimensional comparative analysis. Based on the results of multi-dimensional comparative analysis, a comprehensive analysis report is generated, which includes model comparison analysis, part difference analysis, and specification impact analysis.

8. A multi-dimensional analysis system for engine assembly bolt tightening, characterized in that, The system performs the method as described in any one of claims 1 to 7, comprising: The receiving module is used to receive key feature parameters after data preprocessing and feature engineering. In the process of constructing key feature parameters, based on the geometric structural parameters of the bolt, the prism volume calculation method is used to convert the three-dimensional shape of the bolt into a regular prism for spatial volume calculation, thus forming volume feature parameters. The module is used to fuse volumetric characteristic parameters with the material mechanical properties of bolts to construct a set of key characteristic parameters for finite element simulation analysis. The analysis module is used to perform finite element analysis based on a set of key feature parameters, and to obtain the stress field distribution of the bolted connection structure under tightening conditions through numerical simulation calculation. The calculation module is used to identify stress concentration areas based on the stress field distribution to locate key stress areas; based on the nominal diameter dimension in the bolt specification parameters, a circular area calculation method is used to determine the reference calculation radius based on the nominal diameter; and the equivalent bearing area of ​​the key stress area is calculated using the reference calculation radius. The integration module is used to fuse the equivalent bearing area and stress field distribution to construct a fused dataset containing the correspondence between stress and area. Based on the fused dataset, the data is grouped and integrated according to multiple dimensions such as engine type, functional part classification and bolt specification level to form a multi-dimensional grouped data set. The execution module is used to perform statistical analysis on multidimensional grouped datasets to calculate the statistical characteristics of each group and generate multidimensional analysis results including model comparison analysis, part difference analysis, and specification impact analysis.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 7.

Images