A method for nondestructive testing and structural integrity evaluation of stress concentration points of a key steel structural member

Abstract

This invention discloses a method for non-destructive testing of stress concentration points and structural integrity assessment of key steel structural components. The method first acquires the component's geometry through 3D scanning, reconstructs a 3D model, and performs finite element analysis to theoretically locate high stress concentration areas and fatigue hotspots. Then, targeted non-destructive testing is performed on these areas, integrating multiple testing technologies to identify and quantify actual defects. Finally, the defect information is mapped back to the finite element model to establish a defect-containing structural model. Residual strength and residual fatigue life are assessed based on fracture mechanics, and a remanufacturing decision is output. The residual fatigue life assessment uses the Paris formula for prediction. This invention constructs a closed-loop technical process that deeply integrates digital modeling and simulation analysis, on-site non-destructive testing verification, and data-driven quantitative assessment of structural integrity. This solves the problems of blind testing and inaccurate assessment in existing technologies and is applicable to structural safety determination and quality verification before and after remanufacturing.

Description

Technical Field

[0001] This invention belongs to the field of mechanical structure safety assessment and remanufacturing technology, specifically involving a method for non-destructive testing of stress concentration points and structural integrity assessment of key steel structural components. Background Technology

[0002] Critical steel structural components, such as the frames of large engineering machinery and aggregate bins in mining equipment, are highly susceptible to stress concentration at structural discontinuities (such as welds, holes, and abrupt changes in cross-section) under long-term complex alternating loads, leading to fatigue cracks and structural failure. Remanufacturing these components is a key approach to achieving resource recycling and cost reduction. However, existing technologies for pre-remanufacturing assessment and structural integrity determination of such components suffer from the following main problems: First, it is highly dependent on experience and prone to blind inspection. Existing methods rely heavily on engineers' experience to determine the inspection location, lacking theoretical guidance and easily overlooking potential high-risk points. Traditional non-destructive testing often involves general inspection of the entire structure or major welds, which is inefficient and has a low detection rate for internal defects and micro-cracks, making it difficult to achieve precise location.

[0003] Secondly, the assessment is inaccurate and the remanufacturing process lacks specificity. Existing technologies only determine whether a component is qualified or not based on the detected defect size and general standards, failing to conduct a quantitative safety assessment by considering the actual stress state of the defect at a specific location on a specific component. This makes it impossible to accurately answer the question of how long a crack can operate safely in subsequent use. At the same time, due to the lack of precise understanding of the damage state and stress level, remanufacturing repair schemes (such as welding repair and additive reinforcement) may be poorly designed, leading to new stress concentrations or insufficient repair in the repaired area.

[0004] Therefore, there is an urgent need for an evaluation method that deeply integrates advanced numerical simulation with precise on-site testing to provide a scientific, reliable, and quantitative basis for decision-making regarding the remanufacturing of critical steel structural components. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for non-destructive testing of stress concentration points and structural integrity assessment of key steel structural components.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This application provides a method for non-destructive testing of stress concentration points and structural integrity assessment of critical steel structural components, including the following steps: Step 1: Obtain the geometric shape of the steel structure to be evaluated through 3D scanning, reconstruct it into a 3D geometric model, and perform finite element analysis based on the actual working condition load spectrum to theoretically locate the high stress concentration area and fatigue hot spot area. Step 2: Using the high stress concentration area and fatigue hot spot area as the target area, formulate and implement a targeted non-destructive testing plan to identify and quantify the actual defects and obtain information on the type, size, location and orientation of the defects; Step 3: Map the information of the defect to the three-dimensional geometric model, establish a finite element model of the defective structure, evaluate the remaining strength and remaining fatigue life of the defective structure, and output a remanufacturing decision based on the evaluation results; The remaining fatigue life assessment is based on a fatigue crack propagation model, which uses the Paris formula: In the formula, This represents the amount of crack propagation per load cycle. Crack size The number of load cycles, This represents the stress intensity factor amplitude. and These are material-related fatigue crack propagation parameters.

[0007] Preferably, the three-dimensional scanning in step one is performed using laser scanning or structured light scanning to obtain high-precision three-dimensional point cloud data containing geometric discontinuities. The finite element analysis includes static analysis and fatigue analysis. The static analysis generates a Mises equivalent stress cloud map to identify the stress concentration factor and peak stress region, and the fatigue analysis generates a fatigue life cloud map to identify the fatigue hot spot region with the shortest life. The three-dimensional point cloud data is automatically extracted using a feature recognition algorithm to extract the geometric features of the weld. When reconstructing the three-dimensional geometric model, the weld area is parametrically modeled as an independent sub-model, preserving the actual geometric shape of the weld. Based on the stress gradient change rate in the stress cloud map, the priority classification of the detection area and the scanning path planning are automatically generated. Areas with a stress gradient change rate exceeding a preset threshold are automatically classified as first-level target areas and encrypted scanning paths are generated.

[0008] As a preferred embodiment, the targeted nondestructive testing scheme in step two includes: Based on the finite element analysis results, the detection area is divided into a primary target area, a secondary target area, and a reference area. For different target areas, multiple techniques such as magnetic particle testing, penetrant testing, ultrasonic testing, or phased array ultrasonic testing are used for cross-validation, wherein at least two non-destructive testing methods are used to test the first-level target area. Data from at least two different types of nondestructive testing methods are fused together, and a unified defect feature vector is established through a data fusion algorithm. When different testing methods produce different quantification results for the same defect, a weighted fusion method is used to determine the final defect size. The weight coefficients are pre-calibrated based on the confidence level of each testing method for that type of defect.

[0009] Preferably, step three, which involves mapping the defect information to the three-dimensional geometric model, specifically includes: Based on the global coordinate system established by 3D scanning, the coordinates of the defect location obtained by non-destructive testing are aligned with the coordinates of the finite element model. At the same time, the stress field distribution and fatigue damage accumulation cloud map calculated in the finite element model are reverse-mapped back to the actual component surface. High-risk areas are then superimposed on the component entity using augmented reality equipment. Based on the type and size of the defects, crack-type defects are simplified in the finite element model as semi-elliptical surface cracks, elliptical buried cracks, or quarter-elliptical hole corner cracks, and volumetric defects are simplified as spherical, cylindrical, or ellipsoidal cavities. The mesh is re-divided in the defect area, and singularity elements are set at the crack tip.

[0010] Preferably, the residual strength assessment in step three includes: applying a load to the finite element model of the defective structure and calculating the stress intensity factor at the crack tip. For a type I crack, the expression is: In the formula, This is a geometric correction factor related to crack shape, structural geometry, and loading mode. For nominal stress, The crack size; The calculated maximum stress intensity factor With the fracture toughness of the material If a comparison is made, If so, the crack is determined to be stable under the current load; By gradually increasing the load until... The critical instability load is determined, and the residual strength ratio is calculated. The residual strength ratio is the ratio of the critical instability load to the rated working load.

[0011] Preferably, the remaining fatigue life assessment in step three specifically includes: Determine the initial size of the crack and critical crack size The critical crack size Through residual strength assessment The crack size was determined at that time; Based on the load spectrum and Paris formula, a numerical integration method is used to... Iterative calculation to The number of cycles remaining in the fatigue life is accumulated. ; The numerical integration method employs a iterative recursive approach, calculating the value based on the current crack size in each increment step. And solve for the corresponding cyclic increment. .

[0012] Preferably, the remanufacturing decision in step three includes: When crack stability, residual strength and residual fatigue life all meet the preset safety threshold, a safe use decision is output and a regular monitoring plan is formulated. When any safety condition is not met but the defect is repairable, a repair decision is output, and different repair schemes are simulated and verified based on the finite element model of the defective structure to optimize the repair scheme. When the defect size exceeds the repair threshold or the safety requirements cannot be met after repair, a scrap decision is output. The repair schemes include grinding to eliminate defects, welding repair, or additive reinforcement. A correlation mapping table between defect feature parameters and remanufacturing process parameters is established. When a repair scheme is determined, process parameters are automatically matched or optimized based on the defect features. The stress improvement effect and heat-affected zone performance changes under these process parameters are verified through finite element simulation.

[0013] Preferably, step three outputs the remaining fatigue life. Following that, it also includes: Based on the ratio of remaining service life to design service life, dynamically plan the next inspection time window: like Detection cycle ; like Detection cycle ; like Included in the short-term monitoring list, detection cycle And not exceeding 6 months; in, The number of cycles corresponding to the design life of the component.

[0014] Preferably, the method further includes step four: The remaining fatigue life, critical crack size, and material property degradation coefficient obtained in step three are fed back into the finite element model in step one to update the digital twin of the component. In the next testing cycle, the updated digital twin model will serve as the starting point for a new round of evaluation, enabling dynamic correction and iterative optimization of model parameters.

[0015] Preferably, the key steel structural component is the frame of engineering machinery or the aggregate bin of mining equipment; The actual working condition load spectrum includes static load, dynamic load, impact load and vibration load, and is quantified according to equipment operation records and working conditions, and converted into concentrated force, pressure, body force or force rectangular application to the finite element model. The method is applied to structural safety assessment before remanufacturing and quality verification after remanufacturing.

[0016] Compared with the prior art, this application has the following beneficial effects: This invention constructs a closed-loop technical process that deeply integrates digital modeling and simulation analysis, on-site non-destructive testing verification, and data-driven quantitative assessment of structural integrity. It deeply integrates the macroscopic stress prediction capability of finite element analysis with the microscopic defect detection capability of non-destructive testing, achieving a scientific closed loop from "theoretical prediction" to "empirical testing" and then to "precise assessment." Specifically, this invention first uses 3D scanning and finite element analysis to accurately locate theoretically high-stress concentration areas, overcoming the blindness of traditional methods that rely on experience to determine the testing location. Then, guided by the finite element analysis results, it implements various on-site non-destructive testing techniques in a targeted manner, transforming "general inspection" into "precision inspection," greatly improving inspection efficiency and defect detection rate. Finally, it feeds back the real defect information found in the on-site inspection and maps it back to the finite element model, performing precise residual strength calculation and fatigue life assessment based on fracture mechanics theory. In particular, it quantitatively predicts the number of cycles required for a defect to expand from its current size to its critical size using the Paris formula, solving the problem that existing technologies cannot accurately answer the question of the remaining life of a component. This invention is particularly applicable to structural safety assessment before remanufacturing and quality verification after remanufacturing. It can effectively avoid over-maintenance or under-maintenance, ensure the long-term service safety of remanufactured steel structural components, and realize the transformation from "post-maintenance" to "predictive maintenance". Attached Figure Description

[0017] Figure 1 : Overall flowchart of the method of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Furthermore, in this invention, an element referred to as fixed to or disposed on another element may be directly disposed on the other element, or there may be an intermediate element. When an element is considered to be connected to another element, it may be directly connected to the other element, or there may be an intermediate element present simultaneously. The terms vertical, horizontal, left, right, and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0020] See Figure 1 This application provides a method for non-destructive testing of stress concentration points and structural integrity assessment of key steel structural components, including the following steps: Step 1: Finite element modeling and theoretical location of stress concentration regions based on 3D scanning Before conducting 3D data acquisition, the surface of the aggregate bin to be measured was thoroughly cleaned. High-pressure air guns were used to blow away dust, metal cleaners were used to wipe away oil stains, and thick rust was removed with a wire brush. For reflective surfaces such as newly welded areas, a white developer was evenly sprayed to form a matte layer to avoid laser overexposure. High-reflectivity positioning targets were attached to stable locations on and around the component surface, with the spacing between adjacent targets controlled at 300 to 500 mm to ensure an overlap of more than 30% between adjacent scanning areas. This provides a common reference system for the accurate stitching of subsequent multi-angle scanning data. When planning the scanning path, special attention was paid to ensuring that all geometrically discontinuous areas such as welds, openings, corners, and stiffener connections were completely covered without any blind spots.

[0021] The scanning was performed using a HandySCAN 3D laser scanner with an accuracy of 0.025 mm / s. The operator held the scanner and moved it around the component at a constant speed. The device projected multiple laser lines onto the object's surface, and three cameras captured the deformation of the laser lines in real time, generating a 3D point cloud at a rate of 30 frames per second. During the scanning process, the system automatically identified and located target points and performed real-time data stitching. The operator monitored the integrity of the scan through a software interface and promptly scanned any missing data areas. After the scan was completed, a complete point cloud dataset was obtained, containing approximately 8 million 3D coordinate points with an overall point cloud accuracy of 0.1 mm, which is sufficient to accurately reflect geometric features such as weld reinforcement, undercut, and minute deformations and wear pits.

[0022] After obtaining the point cloud data, the finite element model reconstruction stage begins. The point cloud data is imported into Geomagic Design X reverse engineering software. First, noise reduction is performed, using an external isolated point filtering function to remove discrete noise points. Then, a unified sampling method is used to compress the number of point clouds to two million. Finally, a global registration function is used to accurately align the multi-view point clouds based on the positioning target points, forming a complete and clean component point cloud model. Based on this, a point cloud feature recognition algorithm automatically extracts the geometric features of the weld. Specifically, a curvature analysis algorithm is used to identify areas with drastic curvature changes in the point cloud as weld candidate areas. Morphological analysis is performed on the candidate areas to extract the weld trajectory lines. Based on local point cloud fitting, the geometric parameters of the weld are calculated, including weld reinforcement height, weld width, undercut depth, and the positions of the arc initiation and termination points. In this embodiment, the reinforcement height of the four main welds is 2.5 to 3.2 mm, the weld width is 12 to 15 mm, and the undercut depth is 0 to 0.8 mm.

[0023] To ensure model quality, a manually guided reconstruction method was adopted. Engineers drew key feature lines of the components based on point clouds, extracted the outline boundary lines of the aggregate bins, the center lines of holes, the trajectory lines of stiffeners, and the trajectory lines of welds obtained through feature recognition. Then, they used CAD modeling commands such as extrusion, rotation, and sweep to reconstruct the solid model. In particular, the weld area was parametrically modeled as an independent sub-model. Based on the weld trajectory lines, the weld solid was created using the loft command. Its cross-sectional shape directly adopted the feature recognition results, preserving the actual geometry of the weld rather than simplifying it to an idealized right-angle transition. After completing the surface reconstruction, geometric cleanup was performed to repair holes caused by occlusion and remove small sharp corners, ultimately generating a high-quality STEP format CAD model.

[0024] The STEP model was imported into ANSYS Workbench for mesh generation. Tetrahedral elements with a size of 5 mm were used for non-critical areas, while hexahedral dominant meshes with a size of 1 mm were used for weld areas and areas expected to be under high stress. Anisotropic meshing was also performed in the weld fusion zone, with four layers of elements along the thickness direction. After mesh generation, a quality check was performed to ensure that the warpage was less than 15 degrees, the aspect ratio was less than 5, the Jacobian value was greater than 0.7, and the pass rate reached 98.5%.

[0025] Next, the loads and boundary conditions were determined. By reviewing the equipment design manual, operating records, and interviewing operators, the typical operating conditions of the aggregate silo were identified: the full-load static pressure corresponds to a material pile height of 4 meters and a density of 2.0 tons per cubic meter; the unloading dynamic impact corresponds to a material drop height of 2 meters and an impact coefficient of 2.5; and the vibrating screen excitation force corresponds to a frequency of 16 Hz and an amplitude of 5 mm. The complex actual loads were quantified into a mechanical model suitable for finite element analysis: the material static pressure was calculated using the lateral pressure formula and simplified to a gradient pressure distribution along the silo wall; the unloading impact force applied an equivalent impact load in the discharge port area; the vibration load was converted into an equivalent inertial force applied to the connection point between the aggregate silo and the vibrating screen, with a load spectrum of a sine wave at a frequency of 16 Hz; and standard gravitational acceleration was applied simultaneously. In terms of boundary conditions, fixed constraints were applied to the four support legs to simulate a rigid connection with the foundation, and hinge constraints were applied to the vibrating screen connection point, releasing only the rotational degree of freedom around the connecting axis.

[0026] After completing the above preprocessing, the static analysis was submitted for solution. In terms of result interpretation, the Mises equivalent stress cloud map showed that the root of the four main welds connecting the feed port and the side wall was the area of ​​maximum stress concentration, with a peak stress of 235 MPa and a stress concentration factor of 2.8 in this area. The displacement cloud map showed that the maximum displacement was located at the center of the bottom of the aggregate bin, with a displacement of 1.2 mm, and the stiffness met the design requirements.

[0027] Based on the static analysis results, stress gradient calculations are performed to guide the automatic classification of the detection area. Stress cloud map data is extracted, the rate of change of stress gradient at each node is calculated, and a gradient threshold of 0.5 MPa / mm is set. Areas exceeding this threshold are automatically classified as primary target areas and a denser scanning path is generated, with a scanning interval not exceeding 2 mm. In this embodiment, the roots of the four main welds all meet the threshold condition and are automatically classified as primary target areas. The area 30 mm outward from the weld edge is a secondary target area, and the large area of ​​the base material is a reference area.

[0028] Furthermore, based on the static analysis results and load spectrum, fatigue analysis was performed using the stress-life method. The SN curve of the material was defined in ANSYS nCode DesignLife, the load spectrum was input, and the Miner linear cumulative damage model was selected. The fatigue life cloud map generated by the fatigue analysis showed that the fatigue life at the root of the four main welds was the shortest, at 2.3 x 10⁻⁶. 6 This cycle corresponds to approximately 9.6 years of actual working time, which closely matches the high-stress area in the static analysis. At this point, step one is complete, and a report is output containing a cloud map of the high-stress concentration area, a fatigue hotspot cloud map, a list of stress peak values ​​and stress concentration factors for each area, serving as the action map for step two.

[0029] Step Two: Precise Implementation of On-Site Nondestructive Testing Based on Finite Element Analysis Results Based on the finite element analysis results output from Step 1, a detailed targeted inspection plan is formulated. The inspection area is divided into three levels according to priority: the first-level target area is the stress peak area and the area with the shortest predicted fatigue life, i.e., the root of the four main welds, with the inspection coverage extending 50mm outward from the peak stress point; the second-level target area is the area with high stress level and all geometric discontinuities, including the start and end points of welds, the edges of holes, and the connection of stiffeners; the reference area is the base material area with very low stress level, with a sampling inspection ratio of 5%. Appropriate combinations of non-destructive testing methods are assigned to different levels of target areas. The first-level target area must use both magnetic particle testing and phased array ultrasonic testing for cross-validation, supplemented by ultrasonic thickness measurement to verify the wall thickness; the acceptance standard adopts AWS D1.1 structural welding specifications. No crack-like defects are allowed in the first-level target area. Recording requirements include marking the defect location on the component schematic diagram, saving ultrasonic image screenshots, and archiving phased array scan images to ensure full traceability.

[0030] The on-site inspection team first conducted magnetic particle testing on the primary target areas of the four main welds. Pre-treatment involved using an angle grinder and a grinding wheel to polish the weld surface and 25mm areas on both sides, removing coatings and rust until a metallic luster was exposed, with a surface roughness controlled below Ra 6.3μm. Magnetization was performed using the yoke method, with a yoke spacing of 150mm and a magnetization current of 10A. Magnetization was first performed along the weld length to detect transverse cracks, then perpendicular to the weld direction to detect longitudinal cracks. A gaussmeter was used to measure the magnetic field strength, ensuring it reached above 2000A / m. Simultaneously, a fluorescent magnetic powder suspension with a concentration of 1.2g / L and kerosene as the carrier was sprayed. Under ultraviolet light, the magnetic traces at defects appeared as bright green fluorescent lines. Inspection revealed no cracks on the surface of any of the four welds; only a few dot-like magnetic traces with a diameter less than 1mm were found, which were determined to be irrelevant.

[0031] For the detection of internal defects, the Olympus OmniScan X3 phased array ultrasonic testing equipment was used, along with a 5L64-A32 probe and a miniature wheel scanner. Before testing, the sound velocity and wedge delay were calibrated on the SD-1 standard test block, a TCG curve was established on the SD-2 standard test block to compensate for sound path attenuation, and the sensitivity was calibrated on a φ2mm transverse hole test block with a gain set to 45 dB. A fan-shaped scan was configured with an angle range of 35 to 75 degrees, a step of 1 degree, and a focusing depth of 10 mm. A linear scan was also configured for defect localization. A special ultrasonic coupling agent was used to ensure good contact between the probe and the workpiece surface. The operator moved the probe at a constant speed along the predetermined scanning path, with the speed not exceeding [a certain value]. At a speed of 50 mm / s, the encoder records position information in real time. The instrument acquires and stores the A-scan signal set, generating a B-scan cross-sectional view and a C-scan top view. Offline analysis is performed in OmniPC software. The C-scan image shows a high-echo region at 350 mm from the starting point of weld No. 2. The B-scan cross-section shows that the echo is located at a depth of 12 mm. The weld thickness is 20 mm, and its height is 2 mm. Fan-shaped scan analysis confirms that the echo is a typical crack feature. The final quantitative result is a buried crack with a length of 8 mm, a height of 2 mm, a burial depth of 12 mm, and an orientation perpendicular to the principal stress direction.

[0032] Meanwhile, a CTS-30 digital ultrasonic thickness gauge was used to perform grid-based measurements on the inner wall of the aggregate bin, which is susceptible to material erosion. The grid spacing was 100mm x 100mm. The measurement results showed that the inner wall thickness was between 14 and 16mm, while the original design thickness was 16mm and the maximum thinning amount was 2mm, which is within the allowable range.

[0033] In terms of data fusion, the crack size data obtained by phased array ultrasonic testing is fused with the magnetic particle testing results to establish a unified defect feature vector. Based on the pre-calibrated confidence levels of each testing method on different types of defects, the confidence level of phased array ultrasonic testing for internal buried cracks is 0.95, and the confidence level of magnetic particle testing for surface cracks is 0.98 but it is ineffective for internal defects. Therefore, in the weighted fusion, the phased array ultrasonic data is assigned a weight of 0.95 and the magnetic particle testing data a weight of 0.05. Finally, the defect size is determined to be 8mm in length and 2mm in height. Step two is now complete, and a complete inspection report is output, including a defect list, defect number CR-01, location coordinates associated with the finite element model coordinate system, defect type as buried crack, size 8mm×2mm, and all inspection data, images, and photos are archived.

[0034] Step 3: Integration of test data and finite element model, and structural integrity assessment First, defect information mapping and bidirectional coordinate alignment are performed. Based on the global coordinate system established by the 3D scan in step one, with the origin at the center of the bottom of the aggregate bin, the X-axis along the length direction, the Y-axis along the width direction, and the Z-axis vertically upward, the defect location coordinates in the inspection report are accurately converted into coordinates in the finite element model. Key points are created in ANSYS using the coordinate positioning function to ensure that the position of the defect in the model is completely consistent with the actual component. At the same time, the stress field distribution cloud map calculated in the finite element model is exported and displayed as a superimposed high-risk area on the component entity using augmented reality equipment. Specifically, the AR device scans the positioning target points on the component to identify the component's posture, and then projects the pre-rendered stress cloud map onto the corresponding position on the component in a semi-transparent manner. On-site inspection personnel wearing AR devices can directly see the stress peak area highlighted in red, realizing virtual-real interaction and guiding accurate positioning during re-inspection.

[0035] Based on the defect type and size, the crack is modeled equivalently in the finite element model. The buried crack is simplified to an elliptical buried crack, with the crack size set as follows: 4mm for the major semi-axis corresponding to the half-length, and 2mm for the minor semi-axis corresponding to the half-height. The crack plane is perpendicular to the principal stress direction. For volumetric defects, the modeling can be simplified to a spherical, cylindrical, or ellipsoidal cavity, depending on the actual situation. After introducing the defect geometry, the area around the defect is re-meshed with extremely fine meshing. ANSYS's local mesh refinement function is used, with a 5mm radius dense sphere centered on the crack front, and the element size refined to 0.1mm. In particular, singular elements, i.e., 1 / 4 node elements, are set at the crack tip, and three layers of ring meshes with 8 elements each are arranged around the crack front to accurately capture the extremely high stress gradient at the crack tip. After the mesh quality check is passed, the total number of model elements is increased from the initial 500,000 to 850,000.

[0036] Next, a reanalysis and residual strength assessment of the defective structure are performed. Loading conditions, including full-load static pressure and unloading impact, are reapplied to the finite element model of the defective structure for static analysis. In ANSYS post-processing, the stress intensity factor at the crack tip is extracted using the fracture mechanics module, and ten integration points are uniformly selected along the crack front for calculation. The value shows the deepest point of the crack. Maximum, is .

[0037] The calculated maximum stress intensity factor equals 15.2 MPa Fracture toughness of material 16Mn Equals 45 Comparison, because Less than Furthermore, the ratio is only 0.34, far less than 1, indicating that the crack is stable under the current working load and will not undergo rapid unstable propagation; by gradually increasing the load in the finite element model, calculations were performed under different load levels. The load factor starts at 1.0 and increases in increments of 0.1 until... equal The calculation results show that when the load factor reaches 2.3, 45.2 MPa ,near At this point, the critical buckling load is 2.3 times the rated working load; the remaining strength ratio is equal to the critical buckling load divided by the rated working load, which is 2.3, greater than the safety factor of 1.8 required by the specification, indicating that the remaining strength of the structure is sufficient to safely withstand the working load.

[0038] During the remaining fatigue life assessment phase, the initial crack size is determined. The critical crack size is 2 mm, derived from phased array ultrasonic testing. The residual strength assessment determined that, under the critical instability load, the back-calculation... equal The crack depth at that time was calculated. The thickness is 8.5 mm. Consulting material handbooks and literature, the Paris formula parameters for 16Mn steel were obtained, with C equal to 1.2 × 10⁻⁶. -12 m equals 3.0, stress intensity factor amplitude The calculation considers the minimum and maximum values ​​of the load spectrum. In this embodiment... equal reduce ,in The corresponding no-load condition is approximately 0.2. .

[0039] The remaining fatigue life is calculated using numerical integration with a cyclic recursive method; the crack is then introduced from... equal to 2mm to The 8.5mm increment is divided into one hundred steps, each step It is 0.065mm; for the first Step, based on the current crack size calculate The crack propagation rate in this step is calculated using the Paris formula. Equal to C× Calculate the number of iterations required for this step, then raise the value to the power of m. equal Divide by Finally, the total number of iterations is calculated. The calculation process was completed using AFGROW fracture mechanics analysis software, with the load spectrum and material parameters input. and The initial crack size is 2mm, the critical crack size is 8.5mm, and the software automatically calculates the remaining fatigue life iteratively. equals 1.2 × 10 6 This cycle; based on 8 hours of operation per day, 300 days per year, and a frequency of 16Hz, the actual working time is approximately 8 years.

[0040] Remanufacturing decision analysis based on assessment results; crack stability aspects. Less than The remaining strength is 2.3 greater than 1.8, which meets the requirements; however, the remaining fatigue life is 8 years, which is less than the design life of 10 years, and does not meet the requirements. Therefore, it is determined that the product needs to be repaired before use.

[0041] A mapping table was established between defect characteristic parameters and remanufacturing process parameters. Based on the defect characteristics of this embodiment—buried cracks and small size—a carbon fiber composite bonding reinforcement scheme was recommended. Simulations were performed in ANSYS to verify the two candidate repair schemes. Scheme A was welding repair, simulating the performance degradation of the heat-affected zone during the welding process, reducing the material strength of the welded area by 10%, and applying a post-weld residual stress of 120 MPa. Scheme B was carbon fiber bonding reinforcement, establishing a 3 mm thick composite material layer with an elastic modulus of 235 GPa on the surface of the defect area, simulating perfect bonding with the matrix using shared topology functions. Static analysis results showed that the peak stress in the original defect area was 235 MPa. After repair with Scheme A, the peak stress decreased to 198 MPa, a decrease of 15.7%, but a new stress concentration of 210 MPa appeared in the heat-affected zone. After reinforcement with Scheme B, the peak stress decreased to 142 MPa, a decrease of 39.6%, and the stress distribution was more uniform. Scheme B was selected as the optimal repair scheme after comparison.

[0042] Based on remaining fatigue life equals 1.2 × 10 6 Next cycle, design life equals 1.5 × 10 6 In the next cycle, the ratio of remaining life to design life is calculated to be 0.8, which is greater than 0.5; according to the dynamic testing cycle planning rule, when Greater than 0.5 At that time, the detection cycle Equal to 0.3 times Calculations yielded 3.6×10 5 Each cycle, converted to actual time, is approximately 2.5 years. It is recommended to conduct a re-inspection after 2.5 years.

[0043] The remaining fatigue life, critical crack size, and material property degradation coefficient obtained from this assessment are fed back into the finite element model of Step 1 to update the digital twin of the aggregate bin. Specifically, a material property database is established in ANSYS, the fatigue strength is adjusted from the original 345MPa to 328MPa, the initial crack size is updated to the actual size of 2mm detected in this test as the initial condition for the next assessment, the load spectrum is updated to the actual operation record data, and the average daily unloading frequency is corrected from 200 times to 220 times. The updated digital twin model is used as the starting point for the assessment of the next inspection cycle to achieve dynamic correction and iterative optimization of model parameters. Step 3 is thus completed, and a structural integrity assessment and remanufacturing decision report is output, including a defect mapping diagram, mechanical analysis results of the defective structure, quantitative values ​​of remaining strength and remaining fatigue life, a remanufacturing decision of repair required, the optimal repair scheme and its simulation verification results, and a dynamic inspection cycle plan of re-inspection in 2.5 years.

[0044] After completing the carbon fiber composite bonding reinforcement, the same method was used again to verify the quality of the repaired area. First, a three-dimensional scan of the repaired area was performed to obtain the actual geometry after reinforcement. Then, phased array ultrasonic testing was used to detect the original crack area to confirm that there was no crack propagation. Infrared thermal imaging was used to detect the bonding quality of the composite material to confirm that there was no debonding. The repaired geometric model was imported into ANSYS, and static analysis was performed with applied loads. It was confirmed that the peak stress after repair was 142 MPa, which was 39.6% lower than before repair, and the stress concentration factor was reduced to 1.7. Finally, the repaired model and evaluation results were updated to the digital twin as the initial state for the next life cycle.

[0045] As can be seen from the above specific implementation methods, the present invention constructs a complete technical chain of digital twin, empirical verification, quantitative evaluation, and closed-loop iteration, realizing the full-process digitalization and intelligentization from theoretical prediction to precise detection and then to scientific evaluation, so that remanufacturing is no longer a craft based on experience, but a precise science based on data and models.

[0046] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0047] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for non-destructive testing of stress concentration points and structural integrity assessment of key steel structural components, characterized in that, Includes the following steps: Step 1: Obtain the geometric shape of the steel structure to be evaluated through 3D scanning, reconstruct it into a 3D geometric model, and perform finite element analysis based on the actual working condition load spectrum to theoretically locate the high stress concentration area and fatigue hot spot area. Step 2: Using the high stress concentration area and fatigue hot spot area as the target area, formulate and implement a targeted non-destructive testing plan to identify and quantify the actual defects and obtain information on the type, size, location and orientation of the defects; Step 3: Map the information of the defect to the three-dimensional geometric model, establish a finite element model of the defective structure, evaluate the remaining strength and remaining fatigue life of the defective structure, and output a remanufacturing decision based on the evaluation results; The remaining fatigue life assessment is based on a fatigue crack propagation model, which uses the Paris formula: In the formula, This represents the amount of crack propagation per load cycle. Crack size The number of load cycles, This represents the stress intensity factor amplitude. and These are material-related fatigue crack propagation parameters.

2. The method according to claim 1, characterized in that, The three-dimensional scanning described in step one uses laser scanning or structured light scanning to obtain high-precision three-dimensional point cloud data containing geometric discontinuities. The finite element analysis includes static analysis and fatigue analysis. The static analysis generates a Mises equivalent stress cloud map to identify the stress concentration factor and peak stress region, and the fatigue analysis generates a fatigue life cloud map to identify the fatigue hot spot region with the shortest life. The three-dimensional point cloud data is automatically extracted using a feature recognition algorithm to extract the geometric features of the weld. When reconstructing the three-dimensional geometric model, the weld area is parametrically modeled as an independent sub-model, preserving the actual geometric shape of the weld. Based on the stress gradient change rate in the stress cloud map, the priority classification of the detection area and the scanning path planning are automatically generated. Areas with a stress gradient change rate exceeding a preset threshold are automatically classified as first-level target areas and encrypted scanning paths are generated.

3. The method according to claim 1, characterized in that, The targeted nondestructive testing scheme described in step two includes: Based on the finite element analysis results, the detection area is divided into a primary target area, a secondary target area, and a reference area. For different target areas, multiple techniques such as magnetic particle testing, penetrant testing, ultrasonic testing, or phased array ultrasonic testing are used for cross-validation, wherein at least two non-destructive testing methods are used to test the first-level target area. Data from at least two different types of nondestructive testing methods are fused together, and a unified defect feature vector is established through a data fusion algorithm. When different testing methods produce different quantification results for the same defect, a weighted fusion method is used to determine the final defect size. The weight coefficients are pre-calibrated based on the confidence level of each testing method for that type of defect.

4. The method according to claim 1, characterized in that, Step three, which involves mapping the defect information to the three-dimensional geometric model, specifically includes: Based on the global coordinate system established by 3D scanning, the coordinates of the defect location obtained by non-destructive testing are aligned with the coordinates of the finite element model. At the same time, the stress field distribution and fatigue damage accumulation cloud map calculated in the finite element model are reverse-mapped back to the actual component surface. High-risk areas are then superimposed on the component entity using augmented reality equipment. Based on the type and size of the defects, crack-type defects are simplified in the finite element model as semi-elliptical surface cracks, elliptical buried cracks, or quarter-elliptical hole corner cracks, and volumetric defects are simplified as spherical, cylindrical, or ellipsoidal cavities. The mesh is re-divided in the defect area, and singularity elements are set at the crack tip.

5. The method according to claim 1, characterized in that, The residual strength assessment described in step three includes: applying loads to the finite element model of the defective structure and calculating the stress intensity factor at the crack tip. For a type I crack, the expression is: In the formula, This is a geometric correction factor related to crack shape, structural geometry, and loading mode. For nominal stress, The crack size; The calculated maximum stress intensity factor With the fracture toughness of the material If a comparison is made, If so, the crack is determined to be stable under the current load; By gradually increasing the load until... The critical instability load is determined, and the residual strength ratio is calculated. The residual strength ratio is the ratio of the critical instability load to the rated working load.

6. The method according to claim 1, characterized in that, The remaining fatigue life assessment described in step three specifically includes: Determine the initial size of the crack and critical crack size The critical crack size Through residual strength assessment The crack size was determined at that time; Based on the load spectrum and Paris formula, a numerical integration method is used to... Iterative calculation to The number of cycles remaining in the fatigue life is accumulated. ; The numerical integration method employs a iterative recursive approach, calculating the value based on the current crack size in each increment step. And solve for the corresponding cyclic increment. .

7. The method according to claim 1, characterized in that, The remanufacturing decision mentioned in step three includes: When crack stability, residual strength and residual fatigue life all meet the preset safety threshold, a safe use decision is output and a regular monitoring plan is formulated. When any safety condition is not met but the defect is repairable, a repair decision is output, and different repair schemes are simulated and verified based on the finite element model of the defective structure to optimize the repair scheme. When the defect size exceeds the repair threshold or the safety requirements cannot be met after repair, a scrap decision is output. The repair schemes include grinding to eliminate defects, welding repair, or additive reinforcement. A correlation mapping table between defect feature parameters and remanufacturing process parameters is established. When a repair scheme is determined, process parameters are automatically matched or optimized based on the defect features. The stress improvement effect and heat-affected zone performance changes under these process parameters are verified through finite element simulation.

8. The method according to claim 1, characterized in that, Step 3 outputs the remaining fatigue life. Following that, it also includes: Based on the ratio of remaining service life to design service life, dynamically plan the next inspection time window: like Detection cycle ; like Detection cycle ; like Included in the short-term monitoring list, detection cycle And not exceeding 6 months; in, The number of cycles corresponding to the design life of the component.

9. The method according to claim 1, characterized in that, The method also includes step four: The remaining fatigue life, critical crack size, and material property degradation coefficient obtained in step three are fed back into the finite element model in step one to update the digital twin of the component. In the next testing cycle, the updated digital twin model will serve as the starting point for a new round of evaluation, enabling dynamic correction and iterative optimization of model parameters.

10. The method according to any one of claims 1-9, characterized in that, The key steel structural components are the frames of engineering machinery or the aggregate bins of mining equipment; The actual working condition load spectrum includes static load, dynamic load, impact load and vibration load, and is quantified according to equipment operation records and working conditions, and converted into concentrated force, pressure, body force or force rectangular application to the finite element model. The method is applied to structural safety assessment before remanufacturing and quality verification after remanufacturing.

Images