Method and system for measuring the profile of the asymmetric distribution of residual stresses in a curved member

Abstract

The present application relates to stress measurement technical field, specifically related to the form inversion measurement method and system of the asymmetric distribution of the residual stress of the bending member, including the following steps: S1, welding the box section bending steel member, obtaining the initial curvature of the bending steel member along the length direction in the unconstrained free state, S2, stress release, cutting operation is carried out at the midpoint of the length direction of the bending steel member or the section where the initial bending rate is maximum, the stress of the bending steel member is released, and two symmetrical steel members are obtained, S3, form inversion measurement after stress release and curvature change calculation, S4, residual stress distribution inversion calculation, the present application can macroscopically and integrally measure the residual stress distribution of such member, and the core method for quantitatively verifying the stress asymmetry of the bending concave side and the convex side. Meanwhile, the macroscopic measurement method and the local point measurement method can be combined, verified and corrected with each other to obtain more accurate and reliable residual stress field data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of structural engineering stress measurement, and particularly relates to a method and system for measuring the morphology of residual stress asymmetric distribution of a curved member. BACKGROUND

[0002] Box section steel members are widely used in large space structures such as buildings and bridges due to their high torsional stiffness and load-bearing capacity. In recent years, with the increasing pursuit of complex spatial curved surface morphology in architectural aesthetics, curved steel members have been widely used. Compared with the traditional method of fitting curves with straight bars, curved members can more directly and accurately achieve the architectural modeling intent, and have become key elements in landmark buildings such as large-span stadiums and transportation hubs. This trend has put higher requirements on the processing precision and stress performance control of curved members.

[0003] The manufacturing process of welded box section curved members is much more complex than that of straight members. Specifically, instead of welding straight boxes and then bending them as a whole, the process involves separate processing and assembly welding. First, the two flange plates that make up the concave and convex sides of the curved member are directly processed into the target curved shape through cold bending. At the same time, the two web plates are cut into corresponding fan-shaped segments. Finally, the four plate pieces with curved shapes are assembled into the final box-shaped curved member through welding.

[0004] Therefore, the residual stress field inside such members is a very complex multiple coupling field: it contains plastic deformation stress introduced by cold bending of flange plates, initial stress in the web plate cutting and forming process, and welding residual stress introduced by uneven thermal cycling during welding of the four main welds. The distribution of these residual stresses of different sources and different natures after superposition in the cross section of the member, especially whether the distribution is symmetric on the concave and convex sides of the curve, becomes extremely complex and difficult to predict by traditional theory.

[0005] The existence of residual stress can significantly affect the stability, fatigue life, and overall mechanical properties of steel members. Accurate measurement of the residual stress distribution inside such welded curved box section members made by complex processes, especially quantitative verification of their asymmetry, is crucial for precise structural safety assessment and has been a long-standing technical difficulty in the field.

[0006] Currently, common residual stress measurement methods (such as the sectioning method, blind hole method, and X-ray diffraction method) have obvious limitations when facing such complex curved members:

[0007]

[0008] ​(1) Point measurement limitation and high cost: These methods (such as blind hole method, X-ray diffraction method) can only obtain the stress values of discrete points on the surface of the component. The stress field of the welded box-shaped bending component has a large gradient in the weld zone, and is unevenly distributed on the bending plate, which requires extremely dense measuring points to approximately describe, which is low in efficiency, high in cost, and cannot macroscopically reflect the overall stress distribution pattern of the section.

[0009] (2) Unable to identify the composite stress source: The traditional point measurement method is difficult to distinguish and quantify the contribution of welding stress and bending forming stress, and cannot effectively evaluate the comprehensive influence of the coupling of the two on the cross-section asymmetry.

[0010] (3) Principle inapplicability of classical destructive method: The basic principle of the strip cutting method, as a classical residual stress measurement method, is to inversely calculate the stress by measuring the distance change (i.e. strain) between the "marking points" before and after cutting. However, this principle is seriously inapplicable in theory for bending components with initial curvature, especially their curved flanges:

[0011] Principle error: When cutting the curved flanges of the bending component, the cutting releases not only the elastic strain caused by the residual stress, but also the rigid body displacement and rotation component of the component from the initial bending shape to the new shape. After cutting, the strip will not only change in length, but also change in spatial arc shape. At this time, the distance change between the two marking points is the result of the combined action of elastic strain, rigid body displacement and arc deformation, which cannot be effectively separated, resulting in a huge error in the inverse calculation based on the one-dimensional linear strain assumption, and the result is seriously inaccurate.

[0012] Loss of macroscopic shape information: The cutting process completely destroys the integrity and initial macroscopic bending shape of the component. The core of the present application is to use the macroscopic shape change to inverse the stress, and the cutting method loses this key information carrier in the first step, so it cannot be used to verify the influence of the residual stress distribution on the overall bending shape of the component, and cannot be used to investigate the macroscopic effect of the asymmetry of the concave side and convex side stress.

[0013] Introducing secondary stress disturbance: The complex cutting process itself will introduce new processing stress and thermal stress, which will interfere with the original residual stress field and affect the accuracy of the final result.

[0014] Therefore, for the welded box-shaped bending component, there is a lack of a reliable measurement method in the prior art that can avoid the limitation of local point measurement, overcome the geometric interference of the curved surface, and effectively evaluate the cross-section residual stress asymmetry from a macroscopic perspective. SUMMARY

[0015] In view of the above-mentioned defects of the prior art, the present application provides a kind of bending member residual stress asymmetric distribution form inversion measurement method and system, the core method of macroscopically, integrally measuring the residual stress distribution of such member, and quantitatively verifying the stress asymmetry of bending concave side and convex side.It can also combine macroscopic measurement method with local point measurement method, verify and correct each other to obtain more accurate and reliable residual stress field data.

[0016] To achieve the above object, the present application is realized by the following technical solutions:

[0017] The bending member residual stress asymmetric distribution form inversion measurement method comprises the following steps:

[0018] S1, welding the box section bending steel member, obtaining the initial curvature of the bending steel member along its length direction in the unconstrained free state ;

[0019] S2, stress release, cutting operation is carried out at the midpoint of the length direction of the bending steel member or the section with the maximum initial bending rate, the stress of the bending steel member is released, and two symmetrical steel members are obtained

[0020] S3, stress release form inversion measurement and curvature change calculation, obtaining the curvature of the two steel members in the unconstrained free state , the curvature change of the front and rear members after cutting is calculated , that is ;

[0021] S4, residual stress distribution inversion calculation:

[0022] S41, calculate the release bending moment, according to the beam bending theory in material mechanics, curvature change and additional bending moment generated by residual stress release There is a linear relationship:

[0023] ;

[0024] Wherein, E is the elastic modulus of steel, I is the moment of inertia of the box section around the neutral axis;

[0025] S42, establish inversion model, additional bending moment is by section internal self-balanced residual stress Integral, that is, bending moment balance equation:

[0026] ,

[0027] Self-balancing equation is: ,

[0028] Wherein, The cross-sectional area of ​​the box is... The distance from the area of ​​the infinitesimal element on the cross section to the neutral axis;

[0029] S43, Residual stress distribution modeling and inversion solution: Solve the moment equilibrium equations and self-equilibrium equations, and output residual stress distribution contour maps or inverted stress values. A list.

[0030] Optionally, S1 may also include the following steps:

[0031] S11, the bending steel member to be tested is placed on the platform in an unconstrained free state;

[0032] S12, using a 3D laser scanner or photogrammetry system to acquire dense point cloud data of the surface of the curved steel component;

[0033] S13, through point cloud data processing, the spatial curve of the geometric centerline of the curved steel component is extracted and fitted, and its length coordinates are established. Equation of the initial geometric centerline for variables ;

[0034] S14, using the initial geometric equations, calculate the initial curvature of the bent steel member along its length. The calculation formula is: .

[0035] Optionally, straight steel plates are directly processed into the target curved shape through a cold bending process to obtain two flange plates on the concave and convex sides of the curve. At the same time, the two web plates are cut into corresponding fan-shaped segments. Finally, these four plates with curved shapes are assembled into a box-shaped curved steel component by welding.

[0036] Optionally, wire cutting can be used to cut bent steel components.

[0037] Optionally, S3 also includes the following steps:

[0038] S31: Perform the same operation as S12 on the two cut steel components to fit and extract the spatial curves of the geometric center lines of the two steel components, and establish their length coordinates. The equation of the new geometric centerline with variables The curvature corresponding to the new centerline is calculated using the formula in S1. Curvature change It originates directly from the release of residual stress within the bent steel component.

[0039] Optionally, before cutting the curved steel member, a blind hole method is used to measure, according to the conventional blind hole method standard, a strain gauge is pasted on the surface of the middle of the flange plate of the curved steel member, near the weld area, drilling and strain measurement are performed, and the preliminary measurement value of the surface residual stress of the measuring point is obtained .

[0040] Optionally, after the blind hole method measurement and the shape inversion method measurement, the following steps are further included:

[0041] S5, data comparison and correction model establishment, based on the result reflecting the overall stress state of the curved steel member after the residual stress distribution inversion calculation, to identify and correct the system error existing in the blind hole method measurement.

[0042] Optionally, specifically includes:

[0043] S51, data preparation and pairing:

[0044] All measuring points on the curved steel member which are measured by the blind hole method and can be covered by the shape inversion method are selected, to ensure that each measuring point i has two data pairs: the surface residual stress value measured by the blind hole method ; the residual stress value corresponding to the same position and the same direction extracted from the full-field distribution result of the shape inversion method ;

[0045] S52, system error analysis and correction model establishment, all measuring point pairs are plotted on the scatter plot, and is taken as the horizontal coordinate x, is taken as the vertical coordinate y;

[0046] Error pattern identification: observe the distribution trend of the data points;

[0047] A, if the point group is uniformly distributed near the y=x straight line, it indicates that the blind hole method measurement is accurate on the member, and no correction or only simple offset correction is needed;

[0048] B, if the point group shows a significant systematic deviation, in the high compressive stress area is a larger negative value), the absolute value of the blind hole method measurement value is systematically smaller than , which indicates that the blind hole method underestimates the compressive stress due to the plastic effect, and needs to be corrected;

[0049] Establishing a quantitative correction model: based on the identified system error pattern, a numerical method is used to establish a correction function:

[0050] Linear correction model: if the error is linearly related, the correction formula is obtained by linear regression fitting: , wherein is the corrected stress value, the slope and intercept determined by fitting;

[0051] Nonlinear correction model: if the error is nonlinear, use piecewise linear function or quadratic function for fitting, get accurate correction;

[0052] S53, model verification and applicability confirmation, apply the corrected model to all blind hole method measurement points to get the corrected stress value , calculate the residual error between the corrected and If the residual error is significantly reduced and randomly distributed, it indicates that the correction model is effective, successfully separates and eliminates the systematic error;

[0053] S54, generate the fused residual stress field, use the verified correction model to correct the data of all blind hole method measurement points in batches, embed the corrected local stress data points as accurate boundary conditions into the residual stress distribution framework obtained by the shape inversion method, which meets the overall mechanical balance, to generate the final corrected residual stress field.

[0054] The measurement system comprises:

[0055] A three-dimensional morphology acquisition unit is configured to acquire three-dimensional spatial coordinate point cloud data of the surface of the component before and after the component is cut off;

[0056] A data processing and center line extraction unit is configured to receive the point cloud data of the three-dimensional morphology acquisition unit, and reconstruct a geometric model of the component through an embedded algorithm, and extract a geometric center line for calculation;

[0057] A curvature calculation and shape change analysis unit is configured to automatically calculate the curvature and curvature change amount of the geometric center line based on the extracted geometric center line;

[0058] A stress inversion calculation unit is configured to convert the shape change amount into a residual stress distribution result;

[0059] A control and visual human-computer interaction interface unit is configured to integrate the control of the cooperative work of the units and provide intuitive result display.

[0060] Optionally, the three-dimensional morphology acquisition unit is a laser three-dimensional scanner or a photogrammetry system; the data processing and center line extraction unit comprises a point cloud processing software module, the point cloud processing software module adopts a non-uniform rational B-spline based curve fitting algorithm to process the discrete point cloud data and reconstruct a geometric center line equation that is smooth, continuous and consistent with the actual bending shape of the component and ; the curvature calculation and shape change analysis unit comprises a sign operation and numerical differentiation module, which can perform numerical differentiation on the center line equation Perform automatic differentiation and calculate the first derivative. and second derivative Subsequently, the curvature is automatically calculated at each point along the length of the component according to the curvature calculation formula, and the curvature change is finally output. The stress inversion calculation unit has a built-in inversion calculation engine that receives... Input the moment of inertia of the box section about its neutral axis and the elastic modulus of the steel to calculate the released bending moment. The inversion engine solves the bending moment balance and self-balancing equations and outputs the final residual stress distribution cloud map or a list of inverted stress values. It also automatically calculates and identifies the maximum stress amplitude of the concave and convex flanges.

[0061] The technical solution provided by this invention has the following advantages compared with the known prior art:

[0062] 1. The morphological inversion method of this invention breaks through the limitations of traditional point measurement techniques. It does not rely on limited local strain data to infer the overall stress, but uses the overall geometry of the component itself as the measurement signal. It releases stress through a single cutting operation and accurately measures the resulting stress.

[0063] The macroscopic morphological changes (curvature changes) are observed, and the residual stress distribution on the cross-section is directly inverted using a mechanical model. For verifying the macroscopic distribution characteristics of the asymmetry between residual stress on the concave and convex sides of a bend, this invention provides direct and quantitative measurement results, a problem that traditional methods cannot effectively solve due to their localized nature.

[0064] 2. Through high-precision three-dimensional measurement and rigorous mathematical modeling, the residual stress can be quantitatively determined.

[0065] The measurement provides precise input for subsequent structural calculations by providing specific numerical values ​​and distribution cloud maps, rather than qualitative or semi-quantitative judgments. The measurement results reflect the overall, average stress state of the component, effectively avoiding...

[0066] In areas with large stress gradients (such as near welds), even slight deviations in the measurement point location can lead to significant differences in results, making the results more representative and reliable for engineering applications. This method, based on elasticity theory, is unaffected by the absolute value of residual stress. Even when the residual stress approaches or reaches the material's yield strength, it can still be accurately measured, effectively overcoming the measurement errors caused by plastic effects in high-stress areas by traditional methods such as the blind hole method.

[0067] 3、The special measurement system integrates and automates the complex measurement process, greatly reduces the operation difficulty and the technical dependence on the operator, reduces the human error, improves the repeatability and consistency of the measurement results, and the system can generate the component shape change animation, the curvature change curve and the residual stress distribution cloud diagram on the section, so that the abstract stress data becomes intuitive and visible, and the asymmetric characteristics are convenient to understand and analyze. The method provides a clear and repeatable operation process and data processing standard, and lays a foundation for establishing a standard measurement method for residual stress of such complex components in the future.

[0068] 4、The measurement method of the fusion blind hole method plays the respective advantages of the macro-accuracy of the shape inversion method and the local accuracy of the blind hole method, forms a unique complementary enhancement effect, verifies and corrects the local measurement with the overall result, greatly improves the reliability of the final data, and is especially suitable for scientific research and key engineering with extremely high precision requirements. The fusion scheme provides a reliable benchmark method for evaluating and calibrating the applicability of other residual stress measurement methods on complex components, and has important industry guiding value. BRIEF DESCRIPTION OF DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0070] Figure 1 The flowchart of the present application;

[0071] Figure 2 The structural schematic diagram of the curved steel member;

[0072] Figure 3 The structural schematic diagram of the shape change of the curved steel member before and after cutting;

[0073] Figure 4 The center line comparison schematic diagram of the curved steel member before and after cutting;

[0074] Figure 5 The system structure schematic diagram of the present application;

[0075] Figure 6 The residual stress distribution diagram of the curved steel member measured by the blind hole method at the position of the proposed cutting;

[0076] Figure 7 The residual stress distribution schematic diagram of the welded box section in the existing ECCS model;

[0077] Figure 8To improve the residual stress distribution diagram of the welded box section bending steel member in the model;

[0078] Figure 9 To combine the flowchart of the blind hole method measurement;

[0079] Figure 10 To the cross-sectional structure diagram of the bending steel member.

[0080] Reference numerals: 1-bending steel member, 2-flange plate, 3-web. DETAILED DESCRIPTION

[0081] The technical solutions of the present application will be further described below in combination with the drawings and through specific embodiments.

[0082] Wherein, the drawings are only used for exemplary illustration, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation on the present application; in order to better illustrate the embodiments of the present application, some components of the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings can be omitted.

[0083] The present application will be further described below in combination with the embodiments.

[0084] The method for measuring the asymmetric distribution of residual stress of the bending member includes the following steps:

[0085] S1, the bending steel member 1 of the welded box section, the initial curvature of the bending steel member 1 along the length direction thereof in the unconstrained free state is obtained ;

[0086] S2, stress release, cutting operation is performed at the midpoint of the length direction of the bending steel member 1 or the section with the maximum initial bending rate, the stress of the bending steel member 1 is released, and two symmetrical steel members are obtained;

[0087] S3, stress release form inversion measurement and curvature change calculation, the curvature of the two steel members in the unconstrained free state is obtained , the curvature change of the front and rear members after cutting is calculated , that is: ;

[0088] represents the initial curvature of the bending steel member 1 before cutting, represents the curvature of the two members after cutting;

[0089] S4, residual stress distribution inversion calculation:

[0090] S41, calculate the release bending moment, according to the beam bending theory in material mechanics, the curvature change and the additional bending moment generated by the residual stress release There is a linear relationship:

[0091] ;

[0092] Wherein, E is the elastic modulus of steel, I is the moment of inertia of the box section around the neutral axis;

[0093] S42, establish the inversion model, the additional bending moment is generated by the self-balanced residual stress in the section Integrating, that is, the bending moment balance equation:

[0094] ,

[0095] The self-balanced equation is: ,

[0096] Wherein, A is the area of the box section, d is the distance from the micro-area of the section to the neutral axis;

[0097] S43, residual stress distribution modeling and inversion solution, solve the bending moment balance equation and self-balanced equation group, output residual stress distribution cloud or inversion stress value List.

[0098] In this embodiment, as shown in Figure 1 As the modeling basis of the present application, the classical residual stress distribution model (often referred to as the ECCS model or Birkeland model) recognized in the field of welding structure mechanics is used as the reference model, which has been verified by a large number of experiments and is applicable to the welding box section, as shown in Figure 7 (The existing classical welding box section residual stress distribution model) shows,

[0099] Tensile stress zone: distributed in a certain width of belt-shaped area near the four welds, the stress value is the maximum residual tensile stress .

[0100] Compressive stress zone: distributed in the middle of the plate (the area far away from the weld), the stress value is the maximum residual compressive stress .

[0101] Transition zone: linear transition zone between tensile and compressive stress zones.

[0102] Geometric parameters: tensile stress zone width b, transition zone width c, and compressive stress zone width a are constants related to plate thickness, plate width-thickness ratio, and welding process.

[0103] Symmetry: In the reference model, the residual stress distribution pattern is the same for the four plates (two flange plates 2 and two web plates 3) of the box section, and the stress amplitude and is constant, i.e. the model is symmetric.

[0104] Model innovation: The introduction of the curvature effect leads to the amplitude asymmetry between the concave side and the convex side, and the stress amplitude in the classical model is not considered as a constant. For the members subjected to bending, the effect of the bending process will be coupled with the welding effect, resulting in the residual stress amplitude of the bending concave side and the convex side corresponding area not equal. Figure 6 verifies this point by the residual stress distribution diagram of the middle part (the position of the proposed cutting) of the bending steel member measured by the blind hole method, which can be seen that the residual tensile stress amplitude of the bending concave side flange of this position is obviously larger than that of the convex side. At the same time, through a large number of experimental data, the residual stress distribution model of the welded box section bending steel member 1 is proposed, as shown in Figure 8 (improved residual stress distribution pattern of welded box section bending steel member): on the basis of the existing classical distribution pattern (i.e. the area division a, b, c remains unchanged), the stress amplitude of the concave side and the convex side is allowed as an independent parameter to be solved, taking the flange plate 2 as an example:

[0105] The maximum residual tensile stress amplitude of the flange plate 2 of the bending concave side is denoted as P1, and the maximum residual compressive stress amplitude is denoted as P2. The maximum residual tensile stress amplitude of the flange plate 2 of the bending convex side is denoted as P3, and the maximum residual compressive stress amplitude is denoted as P4. The residual stress distribution pattern of the two web plates 3 still adopts the classical symmetric ECCS model.

[0106] Model parameters and solution: the improved model based on the parameter correlation and inversion of physical assumptions introduces four key amplitude parameters (P1, P2, P3, P4), but the independent equations for inversion solution are only two:

[0107] , bending moment balance equation;

[0108] , self-balancing equation;

[0109] In order to make the equation set closed and solvable, reasonable physical assumptions need to be introduced to correlate the parameters and reduce the number of unknowns. Based on the existing technology and research conclusion, for a given welding process (heat input, plate thickness, width-thickness ratio, material, etc.), there is a relatively stable proportional relationship between the tensile stress amplitude generated near a weld and the compressive stress amplitude generated in the middle of the plate. According to this, the following key assumptions are introduced:

[0110] For the same flange plate 2, the ratio of the absolute value of the tensile stress amplitude in the weld zone to the compressive stress amplitude in the middle of the plate is approximately a constant (usually ), which can be expressed as:

[0111] , ;

[0112] where, and are proportional constants, which can be obtained in advance by finite element simulation or test calibration of the test piece. Under this assumption, the four amplitude parameters (P1, P2, P3, P4) are reduced to two independent parameters (e.g. P2 and P4). The tensile stress parameters P1 and P3 can be calculated by P2 and P4 respectively. Thus, two equilibrium equations are sufficient to solve two independent parameters, making the inversion problem mathematically closed and solvable.

[0113] Model solution and asymmetry verification:

[0114] Substitute the above parameterized improved model into the bending moment balance equation and the self-balancing equation, and then perform numerical inversion solution. The solution process is a constrained optimization problem, and the goal is to find a set of parameters (P2, P4) that make the calculated release bending moment closest to the measured bending moment , and strictly satisfy the self-balancing equation. After inversion calculation, the stress amplitudes of the concave side and the convex side corresponding to the characteristic regions solved by the model (such as comparing P2 (concave side compressive stress) and P4 (convex side compressive stress)) can be directly compared to quantitatively verify the asymmetry of residual stress introduced by the bending process. If P2≠P4, it is proved that the asymmetry exists, and the difference can be accurately calculated.

[0115] Further, S1 further includes the following steps:

[0116] S11, placing the bending steel member 1 to be measured in a free state without constraints on the platform;

[0117] S12, using a three-dimensional laser scanner or a photogrammetric system to obtain dense point cloud data of the surface of the bending steel member 1;

[0118] S13, through point cloud data processing, fitting and extracting the spatial curve of the geometric center line of the bending steel member 1, and establishing the initial geometric center line equation with the length coordinate of the bending steel member 1 as the variable;

[0119] S14, combining the initial geometric equation, calculating the initial curvature of the bending steel member 1 along the length direction, and the calculation formula is: .

[0120] Further, the flat steel plate is directly processed into the target curved shape through a cold bending process to obtain two flange plates 2 with bent concave and convex sides, and two web plates 3 are cut into corresponding fan-shaped segments. Finally, the four plate members with bent shapes are assembled into a bent steel member 1 with a box-shaped cross-section through welding.

[0121] Further, the cutting of the bent steel member 1 is performed by wire cutting, and the cutting accuracy is better than 0.1 mm.

[0122] Specifically, as shown in Figure 2 , Figure 3 , Figure 10 , the cutting operation is performed at a key position (usually selected at the midpoint of the length direction of the member or at the cross-section with the maximum initial bending curvature) of the bent steel member 1. To ensure complete stress release and avoid introducing new processing stress, the member is preferably completely separated by cold processing methods such as wire cutting or water jet cutting.

[0123] Further, S3 further includes the following steps:

[0124] S31, the same operation as S12 is performed on the two cut steel members, as shown in Figure 4 , the spatial curve of the geometric center line of the two steel members is fitted and extracted, and a new geometric center line equation is established with the length coordinate as the variable . The curvature corresponding to the new center line is calculated according to the formula in S1 . The curvature change directly comes from the release of the internal residual stress of the bent steel member 1.

[0125] Further, before cutting the bent steel member 1, a blind hole method is added for measurement. According to the conventional blind hole method standard, a strain rosette is pasted on the surface of the flange plate 2 of the bent steel member 1 near the weld area, and drilling and strain measurement are performed to obtain the preliminary measurement value of the surface residual stress of the measurement point .

[0126] Specifically, as shown in Figure 9 , the blind hole method is widely used as a mature local stress measurement technology, but it has inherent limitations, especially when measuring welded box-shaped cross-section bent members:

[0127] Limited measurement depth: The blind hole method can only measure the residual stress in the shallow layer of the member surface. For thicker steel plates, the stress gradient along the thickness direction in the interior cannot be captured, resulting in measurement results that cannot represent the stress state of the entire plate thickness.

[0128] High stress measurement distortion: when the residual stress level is high (for example, close to or more than a certain proportion of the yield strength of the material), the plastic deformation caused by stress release during drilling will cause significant errors in the measurement results, that is, the so-called "residual stress value / yield strength" cannot be too large" limit.

[0129] Lack of overall representation: the stress value of a single or a few measuring points is difficult to truly reflect the residual stress distribution law of the macro whole of the component, especially cannot truly solve the specific numerical value of the residual stress amplitude asymmetry on the cross-section level.

[0130] The purpose of combining the blind hole method is to give full play to the respective advantages of the blind hole method "local measurement accuracy" and the shape inversion method "overall reflection of reality", form a complementary measurement strategy, and use the overall results of the shape inversion method to verify and correct the local measurement values of the blind hole method, so as to obtain more accurate and reliable residual stress distribution data.

[0131] First, the blind hole method is measured: before the component is cut off, according to the conventional blind hole method standard (such as ASTM E837), strain flowers are pasted on the surface of the key parts (such as the middle of the flange plate 2 and the area close to the weld) of the component, and drilling and strain measurement are performed to obtain the preliminary measurement values of the surface residual stress of these measuring points .

[0132] Then the shape inversion method is measured: as described above, the whole process from the initial shape measurement to the cutting and then to the stress inversion calculation is completed, and the residual stress distribution results of the whole component are obtained, especially the stress distribution mode on the cross-section (such as concave side compressive stress and convex side tensile stress) and the inversion stress value at the key position (corresponding to the blind hole method measuring point) .

[0133] Further, after the blind hole method measurement and the shape inversion method measurement, the following steps are included:

[0134] S5, data comparison and correction model establishment, based on the results reflecting the overall stress state of the curved steel component 1 after the residual stress distribution inversion calculation, to identify and correct the systematic errors existing in the blind hole method measurement.

[0135] This step aims to establish a quantitative relationship to identify and correct the systematic errors that may exist in the blind hole method under specific application scenarios (such as high stress level and thick plate) based on the reliable results obtained by the shape inversion method that can reflect the overall stress state of the component.

[0136] Further, it specifically includes:

[0137] S51, data preparation and pairing:

[0138] Select all key points on the component that are measured by blind-hole method and covered by the shape inversion method. These points should be representative, such as located in the middle of the flange, near the weld area, etc.

[0139] Ensure that each point i has two pairs of data: the surface residual stress value measured by blind-hole method ; the residual stress value corresponding to the same position and direction (such as along the axial direction of the component) extracted from the full-field distribution obtained by shape inversion method ;

[0140] S52, system error analysis and correction model construction, draw all point pairs on the scatter plot , and take as the horizontal coordinate x, as the vertical coordinate y;

[0141] Error pattern recognition: observe the distribution trend of data points;

[0142] A, if the point group is uniformly distributed near the line y=x, it indicates that the blind-hole method measurement is accurate on this component, and no correction or only simple offset correction is needed;

[0143] B, if the point group shows a significant systematic deviation, in the high-pressure stress area is a larger negative value), the absolute value of the blind-hole method measurement value is systematically smaller than , which indicates that the blind-hole method has underestimated the compressive stress due to plastic effect, and needs to be corrected;

[0144] Establish a quantitative correction model: based on the identified system error pattern, use numerical methods to establish a correction function:

[0145] Linear correction model: if the error is linear, the correction formula is obtained by linear regression fitting: , where is the corrected stress value, is the corrected stress value, the slope and the intercept are determined by fitting;

[0146] Nonlinear correction model: if the error is nonlinear, use piecewise linear function or quadratic function for fitting to obtain accurate correction;

[0147] S53, model verification and applicability confirmation, apply the correction model to all blind-hole method measurement points to obtain the corrected stress value , calculate the residual error between the corrected and , if the residual error is significantly reduced and randomly distributed, it indicates that the correction model is effective, successfully separates and eliminates the systematic error;

[0148] S54, generating a fused residual stress field, using the verified correction model to batch correct the data of all blind hole method measuring points, embedding the corrected local stress data points as accurate boundary conditions into the residual stress distribution framework obtained by the shape inversion method, which meets the overall mechanical balance, to generate the final corrected residual stress field.

[0149] Specifically, using the verified correction model to batch correct the data of all blind hole method measuring points (including those detail positions not directly covered by the shape inversion method), embedding these corrected and more reliable local stress data points as accurate boundary conditions into the residual stress distribution framework obtained by the shape inversion method, which meets the overall mechanical balance, to fuse with the overall stress distribution pattern obtained by the shape inversion method, and finally generate a high-confidence residual stress field that is accurate in macroscopic distribution and precise in local details, providing top-quality input data for subsequent fatigue analysis, limit bearing capacity evaluation, etc.

[0150] The blind hole method measurement and the shape inversion measurement are combined with each other, complementary to each other, and the result is more reliable, overcoming the limitations of the blind hole method in measuring high stress and thick plates, and making up for the shortcomings of the shape inversion method in reflecting local stress concentration, realizing the mutual verification and correction of "point" and "surface" measurement results. A new paradigm for measurement under high stress conditions is established, providing an effective solution for obtaining accurate data under measurement scenarios where traditional methods may be distorted (such as high-strength steel, thick plate components). The data usability is improved, and the final fused stress field model contains both macroscopic distribution and local details, providing higher-quality input data for subsequent finite element analysis, fatigue life prediction, and structure safety evaluation.

[0151] A measurement system, comprising:

[0152] A three-dimensional topography acquisition unit for acquiring three-dimensional spatial coordinate point cloud data of the surface of the component before and after the component is cut off;

[0153] A data processing and centerline extraction unit for receiving the point cloud data of the three-dimensional topography acquisition unit and reconstructing the geometric model of the component through the built-in algorithm to extract the geometric centerline for calculation;

[0154] A curvature calculation and shape change analysis unit for automatically calculating the curvature and curvature change of the geometric centerline based on the extracted geometric centerline;

[0155] A stress inversion calculation unit for converting the shape change into residual stress distribution results;

[0156] The control and visual human-computer interaction interface unit is used for integrated control of the units and provides intuitive result display.

[0157] Further, the three-dimensional topography acquisition unit is a laser three-dimensional scanner or a photogrammetry system; the data processing and center line extraction unit comprises a point cloud processing software module, the point cloud processing software module adopts a curve fitting algorithm based on a non-uniform rational B-spline, processes the discrete point cloud data, and reconstructs a smooth, continuous and actual bending shape of the component geometric center line equation and ; the curvature calculation and shape change analysis unit comprises a sign operation and numerical differentiation module, can automatically derive the center line equation , calculate the first derivative and the second derivative , then automatically complete the curvature calculation of each point along the length direction of the component according to the curvature calculation formula, and finally output the curvature change amount ; the stress inversion calculation unit built-in inversion calculation engine receives , inputs the moment of inertia of the box section around the neutral axis and the elastic modulus of the steel material to calculate the released bending moment, the inversion engine solves the bending moment balance and self-balancing equation set, outputs the final residual stress distribution cloud diagram or the list of inversion stress values, and automatically calculates and identifies the maximum stress amplitude of the concave side and convex side flange plate 2.

[0158] In the embodiment, as shown in Figure 5 , the three-dimensional topography acquisition unit is responsible for high-precision and rapid acquisition of the three-dimensional space coordinate point cloud data of the surface of the component before and after the component is cut off. The unit includes a high-precision optical scanning device, such as a laser three-dimensional scanner or an industrial photogrammetry system. In order to keep the measurement reference uniform, the system can integrate an automatic positioning platform for fixing and accurately moving the scanning device, to ensure that the coordinate systems of the two scans are strictly consistent. Its core performance index is to be able to capture the small shape change of the component, so its absolute accuracy should be better than 0.1 millimeter.

[0159] The data processing and center line extraction unit receives the point cloud data of the three-dimensional topography acquisition unit, and reconstructs the geometric model of the component through the built-in algorithm, and accurately extracts the geometric center line for calculation. The unit is one of the "brains" of the system, and comprises a dedicated point cloud processing software module. The module adopts a curve fitting algorithm based on a non-uniform rational B-spline (NURBS), processes the discrete point cloud data, and can accurately reconstruct a smooth, continuous and actual bending shape of the component geometric center line mathematical model and . The core of the unit is the robustness of the algorithm, which can effectively filter measurement noise and accurately process the edge characteristics of the box section, to ensure the accuracy of the center line positioning.

[0160] The curvature calculation and morphological change analysis unit automatically calculates curvature and its changes based on the extracted centerline equation. This unit includes a symbolic computation and numerical differentiation module, which can perform symbolic computation and numerical differentiation on the centerline equation. Perform automatic differentiation and calculate the first derivative. and second derivative Subsequently, the curvature is automatically calculated at each point along the length of the component based on the precise formula for curvature, and the change in curvature is finally output. .

[0161] The stress inversion calculation unit is the core calculation unit of the system, responsible for converting morphological changes into residual stress distribution results. This unit has a built-in dedicated inversion calculation engine. It receives data from upstream... The user-inputted material parameters (elastic modulus) and cross-sectional geometric parameters (moment of inertia) are first processed according to the formula... Calculate the release moment.

[0162] The key is that it includes a pre-defined residual stress distribution model for welded box-section bending members, introducing the asymmetry in stress amplitude between the concave and convex sides caused by the bending effect, and not treating the stress amplitude in the classical model as constant. Instead, it recognizes that for bending members, the effects of the bending process are coupled with the welding effect, resulting in unequal residual stress amplitudes in corresponding regions on the concave and convex sides of the bend. Figure 6 This is a residual stress distribution map of the middle part (the proposed cut-off position) of a bent steel component, measured by the blind hole method in the laboratory. It can be seen that the residual tensile stress amplitude of the flange 2 on the concave side of the bend is greater than that on the convex side, while its residual compressive stress amplitude is smaller. The distribution pattern of residual stress in each flange 2 and web 3 conforms to the aforementioned model benchmark. Therefore, the final distribution model used for inversion is based on the classical distribution model, allowing the stress amplitudes on the concave and convex sides to be treated as independent parameters. For example, the compressive stress in the middle of the flange has different values ​​on the concave and convex sides; similarly, the tensile stress amplitude near the weld is also different on the concave and convex sides. The inversion engine solves the moment balance and self-balancing equations, outputting the final residual stress distribution cloud map or numerical list, and automatically calculates and identifies the maximum stress amplitudes on the concave and convex sides, thus directly providing a quantitative conclusion on the asymmetry.

[0163] Finally, the control and visualization human-machine interface integrates the control of the collaborative work of each unit and provides an intuitive display of results. The system provides a graphical user interface for setting measurement parameters (such as component size and material) and starting the measurement process. The interface can visualize the comparison of the three-dimensional model before and after cutting, the curvature change curve, and the distribution cloud map of the final residual stress on the cross section, making the asymmetry results clear at a glance.

[0164] Compared with the prior art, the trinity technical solution provided by the present application brings significant and multi-level positive effects, which are embodied in the following aspects:

[0165] (1) The paradigm shift from "local inference" to "overall inversion" is realized, which is a fundamental breakthrough. The "shape inversion method" invented by the present application breaks through the thinking limitations of traditional point measurement technology. It does not rely on limited local strain data to "speculate" the overall stress, but uses the overall geometric shape of the component itself as a measurement signal, releases the stress through one cutting operation, accurately measures the macroscopic shape change (curvature change) caused thereby, and directly inverts the residual stress distribution on the cross section by using a mechanical model. For the problem of "verifying the asymmetry of residual stresses on the concave side and the convex side of the bending", the present application can provide a technical means for directly and quantitatively measuring the results, and the traditional method cannot effectively solve this problem due to its locality.

[0166] (2) Significant improvement of measurement performance: precision, macroscopic, and high stress measurement

[0167] Through high-precision three-dimensional measurement and strict mathematical model, the specific numerical value and distribution cloud diagram of residual stress can be quantitatively given, rather than qualitative or semi-quantitative judgment, which provides accurate input for subsequent structure calculation. The measurement result reflects the overall and average stress state of the component, effectively avoiding the huge difference in results caused by slight deviation of the measurement point position in the area with large stress gradient (such as near the weld), and the result is more representative and reliable in engineering. The present method is based on the theory of elasticity, and its measurement accuracy is not affected by the absolute value of the residual stress. Even if the residual stress is close to or reaches the yield strength of the material, it can still be accurately measured, effectively overcoming the measurement error of traditional methods such as blind hole method in high stress area due to plastic effect.

[0168] (3) Outstanding advantages of technology integration: automation, visualization, standardization

[0169] The special measurement system integrates and automates the complex measurement process, greatly reduces the operation difficulty and the technical dependence on the operator, reduces the human error, and improves the repeatability and consistency of the measurement results. The system can generate component shape change animation, curvature change curve and residual stress distribution cloud diagram on the cross section, making the abstract stress data intuitive and visible, and facilitating the understanding and analysis of characteristics such as asymmetry. The method provides a clear and repeatable operation process and data processing standard, laying a foundation for establishing a standard measurement method for residual stress of such complex components in the future.

[0170] (4) Significant expansion of application value: from detection to guiding design and optimization

[0171] The measurement result can be directly used for feedback and evaluation of rationality of manufacturing processes such as bending forming and welding. By comparing the residual stress level and asymmetry of the component under different process parameters, scientific decision basis can be provided for optimizing the manufacturing process and reducing harmful residual stress from the source. The accurate residual stress data obtained can be used as the initial state, input into the finite element analysis software, for predicting stress redistribution, creep deformation and fatigue performance of the component during long-term use, and providing key data support for safety and durability evaluation of the structure. The present application realizes the "measurement-evaluation-control" closed loop, and makes the residual stress change from a difficult-to-quantify "implicit" parameter to an accurately measurable, evaluatable and controllable "explicit" index, thereby providing a core technology for realizing the quality control closed loop of high-end steel structure manufacturing.

[0172] (5) Multiplication effect of comprehensive solution: fusion complementation, value multiplication

[0173] The fusion measurement method plays the respective advantages of the "shape inversion method" macro-accuracy and the "blind hole method" local precision, forms a unique complementary enhancement effect, verifies and corrects the local measurement with the overall result, greatly improves the reliability of the final data, and is especially suitable for scientific research and key engineering with extremely high precision requirements.

[0174] The above examples are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that the technical solutions recorded in the foregoing examples can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of measuring the profile of the asymmetric distribution of residual stresses in a curved member, characterized by, Comprising the following steps: S1, welding a bent steel member (1) of a box-shaped cross section, obtaining an initial curvature of the bent steel member (1) in a length direction thereof in a free state without constraint ; S2, stress release, cutting operation at the midpoint of the length direction of the curved steel member (1) or the section with the maximum initial bending rate, releasing the stress of the curved steel member (1), and obtaining two symmetrical steel members; S3, morphology inversion measurement after stress release and curvature change calculation, obtain the curvature of the two steel members in the free state without constraint , calculate the curvature change of the front and rear members after cutting , that is: ; S4, residual stress distribution inversion calculation: S41, calculating the releasing bending moment, according to the beam bending theory in material mechanics, the curvature change amount and the additional bending moment generated by the residual stress releasing There is a linear relationship: ； wherein E is the modulus of elasticity of the steel material, I is the moment of inertia of the box section about its neutral axis; S42, establish the inversion model, the additional bending moment is balanced by the residual stress in the section Integrating, we get the bending moment equilibrium equation: ， The self-balancing equation is: , wherein, A is the cross-sectional area of the box, is the distance from the cross-sectional element area to the neutral axis; S43, residual stress distribution modeling and inversion solving, solving bending moment balance equation and self-balanced equation set, output residual stress distribution nephogram or inversion stress value of the list.

2. The method according to claim 1, wherein S1 further comprises the following steps: S11, placing the curved steel member (1) to be measured in a free state without constraints on the platform; S12, using a three-dimensional laser scanner or a photogrammetry system to obtain dense point cloud data of the surface of the curved steel member (1); S13, through point cloud data processing, the spatial curve of the geometric center line of the curved steel member (1) is fitted and extracted, and a length coordinate of the initial geometric center line equation is established as a variable S14, in combination with the initial geometric equation, calculates the initial curvature of the curved steel member (1) along the length direction , the calculation formula is: .

3. The method according to claim 2, wherein Two flange plates (2) with curved concave and convex sides are obtained by directly processing flat steel plates through cold bending process to obtain the target curved shape, at the same time, two web plates (3) are cut into corresponding fan-shaped sections, and finally, the four plate pieces with curved shapes are assembled into a box-shaped section curved steel member (1) through welding.

4. The method according to claim 3, wherein The cutting of the curved steel member (1) adopts a wire cutting processing method.

5. The method according to claim 4, wherein S3 further comprises the following steps: S31, the same operation as S12 is performed on the two cut steel members, the spatial curve of the geometric center line of the two steel members is fitted and extracted, and the new geometric center line equation with its length coordinate as a variable is established The curvature corresponding to the new center line is calculated according to the formula in S1 The curvature change is directly derived from the release of the residual stress inside the curved steel member (1).

6. The method according to claim 5, wherein Before the bending steel member (1) is cut off, the blind hole method is added for measurement. According to the conventional blind hole method standard, a strain roset is pasted on the surface of the middle part of the flange plate (2) of the bending steel member (1) near the weld area, and drilling and strain measurement are performed to obtain the preliminary measurement value of the surface residual stress of the measurement point .

7. The method according to claim 6, wherein After the blind hole method measurement and the shape inversion method measurement, the following steps are further included: S5, data comparison and correction model establishment, based on the results reflecting the overall stress state of the curved steel member (1) after the residual stress distribution inversion calculation, to identify and correct the system error existing in the blind hole method measurement.

8. The method according to claim 7, wherein Specifically comprising: S51, data preparation and pairing: Select all the measuring points on the curved steel member (1) which are both measured by the blind hole method and covered by the shape inversion method, to ensure that each measuring point i has two data pairs: the surface residual stress value measured by the blind hole method ; From the full-field distribution results obtained by the morphing method, the residual stress values corresponding to the same position and the same direction are extracted ; S52, system error analysis and correction model construction, draw all the measuring points on the scatter diagram , and take as the horizontal coordinate x, as the vertical coordinate y; Error mode identification: observe the distribution trend of the data points; A, if the point group is uniformly distributed near the y=x straight line, it indicates that the blind hole method measurement is accurate on this member, and no correction or only simple offset correction is needed; B, if the point cloud exhibits deviation in the high-pressure stress area is negative, the measurement value by the blind hole method is less than , correction is required; Establishing a quantitative correction model: based on the identified system error mode, a numerical method is used to establish a correction function: Linear correction model: if the error is linear, the correction formula is obtained by linear regression fitting: wherein, is the corrected stress value, the slope and the intercept are determined by fitting; Nonlinear correction model: if the error is nonlinear, a piecewise linear function or a quadratic function is used for fitting to realize correction; S53, model verification and applicability confirmation, apply the modified model to all blind hole method measurement points to obtain the modified stress value , calculate the residual error between the modified and , if the residual error is reduced and randomly distributed, it indicates that the modified model is effective, successfully separates and eliminates the system error; S54, generate fused residual stress field, use the verified correction model to correct the data of all blind hole method measurement points in batches, embed the corrected local stress data points into the residual stress distribution framework obtained by the shape inversion method, which meets the overall mechanical balance, to generate the final corrected residual stress field.

9. A measuring system for realizing the form inversion measurement method of the asymmetric distribution of the residual stress of the curved member according to any one of claims 1 to 8, characterized by, Comprising: A three-dimensional topography acquisition unit is used to acquire three-dimensional spatial coordinate point cloud data of the surface of the member before and after cutting; A data processing and center line extraction unit is used to receive the point cloud data of the three-dimensional topography acquisition unit, and reconstruct the geometric model of the member through the built-in algorithm to extract the geometric center line for calculation; A curvature calculation and shape change analysis unit automatically calculates the curvature and curvature change of the geometric center line based on the extracted geometric center line; A stress inversion calculation unit is used to convert the shape change into residual stress distribution results; A control and visual human-computer interaction interface unit is used to integrate the control of the cooperative work of each unit and provide intuitive result display.

10. The measurement system of claim 9, wherein, The three-dimensional topography acquisition unit is a laser three-dimensional scanner or a photogrammetry system; the data processing and center line extraction unit comprises a point cloud processing software module, the point cloud processing software module adopts a curve fitting algorithm based on non-uniform rational B-spline, processes the discrete point cloud data, and reconstructs a smooth, continuous and actual bending shape conforming geometric center line equation of the component and ; the curvature calculation and shape change analysis unit comprises a sign operation and numerical differentiation module, can automatically derive the center line equation , calculates the first derivative and the second derivative , then automatically completes the curvature calculation of each point along the length direction of the component according to the curvature calculation formula, and finally outputs the curvature change amount ; The inversion calculation engine built in the stress inversion calculation unit receives , the input box section around the neutral axis of the moment of inertia and the elastic modulus of steel material to calculate the release bending moment, the inversion engine solves the bending moment balance and self-balancing equation set, outputs the final residual stress distribution cloud picture or the list of inversion stress values, and automatically calculates and identifies the maximum stress amplitude of the flange plate (2) of the concave side and the convex side.

Images