Multi-stage adjustment method and system for controlling welding deformation of steel structure

Abstract

The application relates to the technical field of welding deformation control, and discloses a multi-stage adjustment method and system for steel structure welding deformation control. The method comprises the following steps: measuring steel component parameters, establishing a welding deformation prediction model, obtaining pre-deformation data, performing accurate pre-deformation processing according to the data, performing segmented welding to record a temperature field and a deformation amount, performing local heat treatment on a high stress area to obtain stress distribution, correcting a deformation area by using a multi-point force school, performing precision detection and quality evaluation according to correction data, and forming a control file. Through data transmission and feedback adjustment in each stage, high-precision control of complex steel structure welding deformation is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding deformation control, in particular to a multi-stage adjustment method and system for steel structure welding deformation control. BACKGROUND

[0002] In the field of large-scale engineering structures, the welding deformation control of steel structures such as bridges, high-rise buildings, and offshore platforms has always been a key technical problem in engineering construction. Traditional methods for welding deformation control of steel structures mainly include pre-deformation method, rigid restraint method, reasonable welding sequence method, and hot processing technology. The pre-deformation method applies a deformation opposite to the expected welding deformation direction to the structure before welding to offset the welding deformation; the rigid restraint method restricts the free deformation of the structure during welding through external clamps; the reasonable welding sequence method optimizes the welding path and sequence to offset the deformation caused by each weld; and the hot processing technology adjusts the deformation formed by local heating and cooling. These methods have been widely used in practical engineering and have formed relatively mature technical specifications and operation standards.

[0003] However, the existing technology still has obvious deficiencies in dealing with the welding deformation control of large and complex steel structures. First, the effect of a single control method is limited, making it difficult to deal with multiple deformation modes in complex structures; second, traditional methods rely heavily on experience and lack precise quantitative analysis and prediction capabilities, resulting in low control accuracy; third, welding deformation control is often carried out in isolation and does not form an organic link with the entire manufacturing process, lacking data sharing and feedback mechanisms between different stages; fourth, existing methods focus on controlling deformation before or during welding, and lack systematic and effective correction methods for residual deformation after welding; fifth, there is a lack of scientific evaluation system for control effect, making it difficult to provide reliable technical reference for similar structures. These deficiencies result in large steel structures having difficulty in deformation control, low precision, and low efficiency during welding, affecting the assembly precision and service performance of the structure. SUMMARY

[0004] The present application provides a multi-stage adjustment method and system for steel structure welding deformation control, which realizes high-precision control of complex steel structure welding deformation through data transmission and feedback adjustment in each stage.

[0005] In a first aspect, the present application provides a multi-stage adjustment method for controlling welding deformation of a steel structure, comprising: establishing a welding deformation prediction model by measuring geometric parameters of a steel structure and combining welding process parameters to obtain a pre-deformation data set; performing accurate pre-deformation processing on a specified position of the steel structure according to the pre-deformation data set to obtain a pre-processed steel structure; performing segmented welding operation on the pre-processed steel structure, recording temperature field distribution and real-time deformation amount in the welding process to obtain a welding deformation record table; performing local heat treatment on a high stress area of the welded steel structure according to the welding deformation record table to obtain a stress distribution map; applying a correction force to the deformation area by a multi-point force correction device according to the stress distribution map to obtain shape correction data; and performing final accuracy detection and quality evaluation on the steel structure using the shape correction data to form a deformation control file.

[0006] In a second aspect, the present application provides a multi-stage adjustment system for controlling welding deformation of a steel structure, comprising:

[0007] A measurement module is configured to establish a welding deformation prediction model by measuring geometric parameters of a steel structure and combining welding process parameters to obtain a pre-deformation data set.

[0008] A processing module is configured to perform accurate pre-deformation processing on a specified position of the steel structure according to the pre-deformation data set to obtain a pre-processed steel structure.

[0009] A recording module is configured to perform segmented welding operation on the pre-processed steel structure, record temperature field distribution and real-time deformation amount in the welding process to obtain a welding deformation record table.

[0010] An implementation module is configured to perform local heat treatment on a high stress area of the welded steel structure according to the welding deformation record table to obtain a stress distribution map.

[0011] An application module is configured to apply a correction force to the deformation area by a multi-point force correction device according to the stress distribution map to obtain shape correction data.

[0012] An evaluation module is configured to perform final accuracy detection and quality evaluation on the steel structure using the shape correction data to form a deformation control file.

[0013] In a third aspect, the present application provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; and the at least one processor invokes the instructions in the memory to enable the computer device to perform the multi-stage adjustment method for controlling welding deformation of a steel structure.

[0014] The fourth aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores instructions which, when executed on a computer, cause the computer to perform the multi-stage adjustment method for steel structure welding deformation control described above.

[0015] In the technical solution provided in the present application, by measuring the geometric parameters of the steel structure and combining with the welding process parameters, a welding deformation prediction model is established, the traditional empirical formula prediction is changed into data-driven accurate calculation, the accuracy of the pre-deformation calculation is greatly improved, and the over-compensation or under-compensation phenomenon is avoided. Meanwhile, the generation of the pre-deformation data set provides accurate guidance for the subsequent process. According to the pre-deformation data set, accurate pre-deformation processing is performed on the specified position of the steel structure, and the composite technology of laser heating and mechanical preloading is combined to realize accurate control of the deformation amount and the deformation direction. The pre-processed steel member lays a good foundation for the subsequent welding work. The segmented welding operation is performed on the pre-processed steel member, the symmetrical segmented staggered welding method is used to effectively balance the heat input distribution, the overall deformation trend is significantly reduced, and the temperature field distribution and the deformation amount are recorded in real time. The welding deformation record table formed provides accurate data support for the subsequent heat treatment. According to the welding deformation record table, local heat treatment is performed on the high stress area, the heating temperature and the cooling rate are accurately controlled, the residual stress is effectively released, and the stress distribution diagram intuitively reflects the stress release effect. According to the stress distribution diagram, the correction force is applied to the deformation area by the multi-point force correction equipment, the progressive loading strategy and real-time stress monitoring are adopted, and the deformation correction is safely and efficiently realized. The shape correction data formed comprehensively records the correction process and effect. The shape correction data is used for final accuracy detection and quality evaluation of the steel structure, and a complete deformation control file is formed, which verifies the deformation control effect and provides valuable experience reference for subsequent similar structures. Especially, the data mining algorithm applied in the welding deformation prediction and control process can extract the deformation law from historical cases, establish the correlation model between the process parameters and the deformation results, and continuously optimize the prediction accuracy through the deep learning method. The prediction result can adapt to different structure forms and welding conditions, which greatly improves the pertinence and effectiveness of the deformation control. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 FIG. 1 is a schematic diagram of one embodiment of the multi-stage adjustment method for steel structure welding deformation control in the present application;

[0018] Figure 2 Figure 1 is a schematic diagram of an embodiment of a multi-stage adjustment system for steel structure welding deformation control in the present application;

[0019] Figure 3 Figure 2 is a schematic block diagram of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The present application provides a multi-stage adjustment method and system for steel structure welding deformation control. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific flow of the embodiments of the present application is described below. Please refer to Figure 1 An embodiment of a multi-stage adjustment method for steel structure welding deformation control in the present application includes:

[0022] Step S101, a welding deformation prediction model is established by measuring the geometric parameters of the steel structure member and combining the welding process parameters, and a pre-deformation data set is obtained;

[0023] Step S102, according to the pre-deformation data set, a precise pre-deformation process is performed on the specified position of the steel structure member, and a pretreated steel member is obtained;

[0024] Step S103, a segmented welding operation is performed on the pretreated steel member, the temperature field distribution and real-time deformation amount in the welding process are recorded, and a welding deformation record table is obtained;

[0025] Step S104, according to the welding deformation record table, local heat treatment is performed on the high stress area of the welded steel member, and a stress distribution map is obtained;

[0026] Step S105, according to the stress distribution map, a correction force is applied to the deformation area by a multi-point force correction device, and shape correction data is obtained;

[0027] Step S106, using the shape correction data, a final precision detection and quality evaluation are performed on the steel structure, and a deformation control file is formed.

[0028] It can be understood that the execution subject of the present application can be a multi-stage adjustment system for steel structure welding deformation control, and can also be a terminal or a server, and the specific implementation is not limited herein. The embodiments of the present application take the server as an execution subject for example.

[0029] Specifically, a welding deformation prediction model is established by measuring the geometric parameters of the steel structure and combining the welding process parameters. In specific operation, a three-dimensional laser scanner is used to scan the bridge steel structure to obtain key node coordinates and geometric size data. These geometric parameters include the length of the bridge girder, the height of the web plate, the width of the flange, and the length and position of the weld, etc. At the same time, welding process parameters such as welding current, voltage, speed and heat input are collected. These parameters are input into a thermodynamic analysis system to establish a welding deformation prediction model. The model analyzes the heat distribution and transfer law in the welding process through the principle of thermodynamics, and calculates the expected deformation direction and deformation amount. For example, for a bridge steel box girder with a span of 50 meters, after measuring its main geometric parameters, combined with process parameters such as 800A welding current and 30V voltage, the welding heat input is calculated to be 24kJ / cm, and through thermodynamic analysis, it is predicted that the maximum deflection deformation of the middle part of the girder is 18mm, and the flange plate warping deformation is about 5mm. These prediction results together with the deformation direction form a pre-deformation data set.

[0030] The steel structure is pre-deformed at specified locations according to the pre-deformation dataset. First, the key points that need to be pre-deformed are determined according to the pre-deformation dataset, such as the middle of the main beam and the connection between the support. A high-power fiber laser is used to heat these locations directionally, and the heating temperature is controlled within the range of 600-800°C to form a controllable temperature gradient. At the same time, a mechanical loading system is used to apply a force opposite to the expected deformation direction to generate pre-deformation. For example, for the middle of the main beam that is predicted to deflect 18mm, an upward pre-deformation of about 22mm is applied (considering the elastic rebound factor). The temperature distribution of the heating area is monitored in real time by an infrared thermal imager to ensure that the temperature is controlled within the set range and to avoid overheating that changes the material properties. After pre-deformation is completed, a three-dimensional laser scanner is used to measure the actual pre-deformation, which is compared with the pre-deformation dataset to ensure that the error is controlled within ±1mm, and finally the pre-processed steel structure is obtained. The pre-processed steel structure is subjected to segmented welding operation. First, the welding area is divided into multiple welding segments according to the structural characteristics, and a welding sequence planning table is developed. Pulse welding technology is used to control the pulse frequency between 80-120Hz and the duty cycle between 40%-60% to reduce heat input and thermal stress concentration. According to the symmetry characteristics of the bridge steel structure, a symmetric segmented staggered welding method is used for welding, which is synchronized from the center to both sides, and the length of each welding seam is controlled within 300-500mm. During the welding process, a temperature sensor array is used to monitor the temperature field distribution of the welding area, and when the temperature exceeds 550°C, the welding is paused and cooled. At the same time, a laser interference deformation monitoring system is used to track the structure deformation in real time and record the deformation values. For example, during the welding of the welding seam between the box beam web and the wing plate, the deformation of the wing plate after welding is monitored to be 3.2mm, which is less than the 5mm predicted in the pre-deformation dataset, and accordingly the welding parameters of the subsequent welding seam are adjusted to reduce the welding current to 220A. The temperature field data and deformation values are integrated into time sequence correlation data to form a welding deformation record table.

[0031] According to the welding deformation record table, local heat treatment is performed on the high stress area of the welded steel structure. By analyzing the welding deformation record table, the stress concentration area is determined, such as the connection node of the main beam and the cross beam, the connection area of the support, etc. The residual stress in these areas often exceeds 70% of the material yield strength. These areas are marked by classification, high-frequency induction heating equipment is set up, the heating temperature is controlled between 550-650℃, and the holding time is 15-30 minutes. Gradient heating method is adopted, the center temperature is the highest, and it gradually decreases outward, forming a temperature gradient of 30-50℃ / cm. The acoustic emission detection technology is used to monitor the stress release in the metal. When the acoustic emission signal intensity decreases to less than 30% of the initial value, it indicates that the stress release is sufficient. Then the cooling rate is controlled within 15-25℃ / min to avoid the generation of new stress. Finally, the X-ray diffraction method is used to measure the residual stress distribution in the heat treated area, and the stress distribution map is generated. According to the stress distribution map, the multi-point force correction equipment is used to apply correction force to the deformation area. First, the high-precision three-dimensional laser scanning system is used to scan the structure after heat treatment comprehensively, and the actual deformation point cloud data is obtained, with an accuracy of ±0.1mm. The actual deformation data is compared with the design model to generate a deformation vector diagram, and the key points that need to be corrected are determined. The multi-point synchronous correction system is designed, which is composed of multiple groups of hydraulic servo actuators. Each actuator provides a correction force of 10-50 tons. The correction force adopts a gradual loading mode, with an initial value of 50% of the target force, and then increases by 10% at each step. For example, for the middle part of the main beam with a deformation of 12mm, the initial correction force is 20 tons, which gradually increases to 40 tons until the deformation is restored to the design tolerance range. During the correction process, the strain monitoring system is used to monitor the stress change in real time to ensure that no new excessive stress is introduced. After the correction is completed, the shape correction data containing the deformation comparison data before and after the correction is generated. The shape correction data is used to conduct final precision detection and quality evaluation on the steel structure. The high-precision three-dimensional laser scanning technology is used to measure the corrected structure, and the measured geometric model is generated. The super limit analysis is performed on the original design model to calculate the size deviation value of the key parts, ensuring that the linear deviation of the bridge main beam is controlled within 2mm / m. At the same time, the portable X-ray diffractometer is used to measure the residual stress of the key welds and heat affected zone, ensuring that the residual stress is reduced to less than 30% of the material yield strength. The internal quality of the weld is detected by ultrasonic phased array detection, and the weld quality data is recorded. The vibration modal test is conducted on the bridge steel structure, the first five natural frequencies are obtained, and they are compared with the predicted values of the finite element model to ensure that the deviation is controlled within ±5%. The static load test is conducted on the key connection nodes, and the deformation recovery performance is monitored under the design load of 1.2 times. The technical parameters and measured data in the whole process are integrated to form the deformation control file.

[0032] In the embodiments of the present application, by measuring the geometric parameters of the steel structural member and combining with the welding process parameters, a welding deformation prediction model is established, the traditional empirical prediction is changed into data-driven accurate calculation, the accuracy of the pre-deformation calculation is greatly improved, the over-compensation or under-compensation phenomenon is avoided, and the generation of the pre-deformation data set provides accurate guidance for the subsequent process. According to the pre-deformation data set, the specified position of the steel structural member is accurately pre-deformed, and the composite technology of laser heating and mechanical preloading is combined to realize accurate control of the deformation amount and the deformation direction, and the preprocessed steel member lays a good foundation for the subsequent welding work. The segmented welding operation is performed on the preprocessed steel member, the symmetric segmented staggered welding method is used to effectively balance the heat input distribution, the overall deformation trend is significantly reduced, and the temperature field distribution and the deformation amount are recorded in real time to form a welding deformation record table for providing accurate data support for the subsequent heat treatment. According to the welding deformation record table, local heat treatment is performed on the high-stress area, the heating temperature and the cooling rate are accurately controlled, the residual stress is effectively released, and the stress distribution diagram intuitively reflects the stress release effect. According to the stress distribution diagram, the correction force is applied to the deformation area by the multi-point force correction equipment, the gradual loading strategy and real-time stress monitoring are adopted, the deformation correction is safely and efficiently realized, and the shape correction data formed comprehensively records the correction process and effect. The shape correction data is used for final accuracy detection and quality evaluation of the steel structure, a complete deformation control file is formed, the deformation control effect is verified, and valuable experience reference is provided for the subsequent similar structure. Especially in the welding deformation prediction and control process, the data mining algorithm applied can extract the deformation law from the historical cases, establish the correlation model between the process parameters and the deformation results, continuously optimize the prediction accuracy through the deep learning method, so that the prediction result can adapt to different structure forms and welding conditions, and the pertinence and effectiveness of the deformation control are greatly improved.

[0033] In a specific embodiment, the process of step S101 can specifically include the following steps:

[0034] Three-dimensional scanning is performed on the bridge steel structural member to obtain the geometric parameters and key node coordinates of the structural member;

[0035] The geometric parameters and the welding process parameters are input into a processing system to construct a welding deformation prediction model;

[0036] The welding deformation prediction model is divided into grid elements, the grid encryption processing is performed on the weld area, and a fine grid of the welding area is obtained;

[0037] Based on the fine grid of the welding area, thermal cycle analysis is performed on the steel structure welding process to obtain temperature field distribution data;

[0038] The welding stress and deformation amount are calculated by the welding deformation prediction model according to the temperature field distribution data, and deformation prediction data are generated;

[0039] The deformation prediction data is corrected under bridge load conditions, considering the effects of structural self-weight and service load, to obtain corrected deformation values;

[0040] Based on the corrected deformation values, the pre-deformation compensation amount is determined, and the pre-deformation position parameters of each node are calculated;

[0041] The pre-deformation position parameters are compared and analyzed with the original parameters to generate a pre-deformation data set containing deformation direction, deformation amount, and implementation location.

[0042] Specifically, the bridge steel structure is scanned three-dimensionally to obtain the geometric parameters and key node coordinates of the structure. A high-precision three-dimensional laser scanner is used to comprehensively scan the steel structure, and the spatial coordinates of the laser reflection points are determined by triangulation principle to form point cloud data. The point cloud data density usually reaches 50-100 points per square centimeter, ensuring sufficient measurement accuracy. For the bridge steel structure, the geometric information of key parts such as main girder, cross beam, and connection node is collected, including size, thickness, angle, and other parameters, as well as the position and length of the weld. These geometric parameters constitute the basis of the digital model of the steel structure, providing input data for subsequent analysis. The geometric parameters and welding process parameters are input into the processing system to build a welding deformation prediction model. The welding process parameters include welding current, voltage, speed, heat input, welding sequence, etc. The processing system integrates these data to establish a welding thermodynamic analysis model. This model is based on the theory of thermal elastoplasticity, considering the changes of material physical properties at high temperature, such as thermal expansion coefficient, elastic modulus, yield strength, etc. The establishment process of the welding deformation prediction model integrates experimental data and theoretical calculation, through the analysis of historical cases, the corresponding relationship between key parameters and deformation is established, forming a prediction algorithm.

[0043] The welding deformation prediction model is divided into grid elements, and the grid is encrypted in the weld area to obtain fine grid in the welding area. Adaptive meshing technology is used for grid division, which automatically adjusts the grid density according to the geometric characteristics and stress gradient of the structure. For the weld and its heat-affected zone, the grid size is usually controlled at 1-3mm, while the grid size in the area away from the weld can be relaxed to 10-20mm. This differentiated grid division strategy greatly improves the calculation efficiency while ensuring the calculation accuracy of the weld area. The grid quality is evaluated by indicators such as twist rate and aspect ratio to ensure that the grid quality meets the requirements of numerical calculation.

[0044] The temperature field distribution data is obtained by conducting thermal cycle analysis on the steel structure welding process based on the fine mesh of the welding area. The moving heat source model is adopted for thermal cycle analysis to simulate the movement of the welding arc on the weld. The double-ellipsoid heat source or Gaussian heat source distribution is adopted for the heat source model to consider the heat distribution characteristics in the depth and plane directions. The temperature values of each point of the structure are calculated by solving the heat conduction equation during the welding process. The thermal physical parameters of the material, such as thermal conductivity and specific heat capacity, are considered to change with temperature, and convection and radiation heat dissipation are also considered. The analysis results include temperature time history curves and temperature spatial distribution, which provide input for subsequent stress analysis.

[0045] The welding stress and deformation are calculated based on the temperature field distribution data through the welding deformation prediction model to generate deformation prediction data. This calculation adopts the thermal elastic-plastic finite element method, considers the thermal expansion, phase change, and elastic-plastic deformation of the material based on the temperature field data, and calculates the stress field and final residual stress distribution during the welding process. The welding stress The welding stress can be calculated by the following formula:

[0046] ;

[0047] where, E(t) represents the elastic modulus at the tth time step, α(t) represents the thermal expansion coefficient at the tth time step, T(t) represents the temperature at the tth time step, T0 represents the reference temperature, Δσp represents the stress increment caused by plastic strain, and T is the total time step. The welding deformation is obtained by calculating the node displacement, considering the comprehensive effect of elastic deformation, plastic deformation, and thermal expansion deformation.

[0048] The deformation prediction data is corrected under the load conditions of the bridge to consider the effects of structure self-weight and service load, and the corrected deformation value is obtained. The bridge is subjected to loads such as self-weight, vehicle load, wind load, etc. in service state, which will have additive or offset effect on the welding deformation. The correction process first establishes a structural mechanics model, applies the corresponding load, and calculates the deformation under the load. Then the load deformation and welding deformation are analyzed by superposition to obtain the comprehensive deformation value. The correction method considers the linear characteristics of load deformation and the nonlinear characteristics of welding deformation, and calculates the final deformation state by the superposition principle.

[0049] The pre-deformation compensation amount is determined based on the corrected deformation value, and the pre-deformation position parameters of each node are calculated. The pre-deformation compensation adopts the reverse deformation method, that is, the structure is subjected to a deformation opposite to the expected welding deformation direction before welding. The compensation amount is not a simple equal compensation, but a compensation coefficient K is used to adjust the compensation amount considering the material elastic resilience performance and the structural stiffness distribution. The pre-deformation position parameters include a pre-deformation direction vector and a pre-deformation amount, and the new coordinates of each node after pre-deformation are calculated through coordinate transformation. The pre-deformation position parameters are compared and analyzed with the original parameters to generate a pre-deformation data set containing deformation direction, deformation amount and implementation position. In the comparison and analysis process, the displacement difference and direction difference of each node are calculated to evaluate the effectiveness and accuracy of the pre-deformation. The pre-deformation data set is recorded in the form of a data table and a three-dimensional model, and contains specific operation guidance such as the position of the pre-deformation point, the heating temperature, the mechanical loading force and other parameters.

[0050] Taking a certain steel box girder bridge as an example, the pre-deformation control process of the main girder is described: first, the geometric parameters of the box girder are obtained through three-dimensional scanning, including the top plate width of 2.5 m, the web height of 1.8 m, the steel plate thickness of 25 mm, etc. Combined with the welding process parameters of 800A current and 30V voltage, the welding deformation prediction model is constructed by inputting the processing system. The model is meshed, the mesh size of the weld area is 2mm, and the size gradually increases to 15mm away from the area, and the total number of mesh elements is about 110,000. Through thermal cycle analysis, the highest temperature of the weld area is 1450℃, and the temperature gradient of the heat affected zone is about 300℃ / cm. The expected downward deformation of the main girder in the middle is calculated to be 18mm, and after considering the 7mm upward deformation caused by the self weight of the structure, the corrected deformation value is 11mm downward. Considering the material elastic resilience factor, the pre-deformation compensation amount is determined to be 13mm upward. Finally, the pre-deformation data set is generated, which contains 11 pre-deformation implementation points along the main girder, the pre-deformation direction and value of each point, and the specific heating parameters and mechanical loading force.

[0051] In a specific embodiment, the process of performing step S102 can specifically include the following steps:

[0052] According to the pre-deformation data set, determine the pre-deformation key points of the bridge steel structure, and generate a pre-deformation operation guide;

[0053] Based on the pre-deformation operation guide, configure a high-power fiber laser, and set the laser heating parameters;

[0054] Directionally heat the specified position of the steel structure by laser to form a temperature gradient area;

[0055] According to the pre-deformation amount in the pre-deformation data set, set a mechanical pre-loading system, and determine the loading force value;

[0056] An initial deformation is generated by applying an external force consistent with the pre-deformation direction to the heating area through a mechanical pre-loading system;

[0057] Real-time temperature monitoring of the laser heating area is performed using an infrared thermal imaging device to obtain temperature distribution data;

[0058] Based on the temperature distribution data, the laser power and scanning speed are adjusted to control the pre-deformation progress;

[0059] A three-dimensional laser scanner is used to measure the pre-deformed structure and compare it with the pre-deformation dataset to obtain the pre-processed steel component.

[0060] Specifically, the pre-deformation key points of the bridge steel structure are determined according to the pre-deformation dataset, and the process of generating the pre-deformation operation guide first involves data analysis. The pre-deformation dataset contains deformation direction, deformation amount, and implementation location information. The positions with deformation amount exceeding the critical value are extracted as key points through data filtering algorithms. Gradient analysis is used to select key points, determine the areas with the largest deformation gradient, and set pre-deformation operation points in these areas. For bridge steel structures, the main key points are usually distributed in the main beam mid-span area, the main and secondary beam connection nodes, and the support connection area. The generation of the operation guide uses a parameterization method to integrate the position coordinates, pre-deformation direction, pre-deformation amount, and required operation means of each key point into a structured data table, including heating point position, heating temperature, heating time, and mechanical loading force value and direction parameters.

[0061] Based on the pre-deformation operation guide, a high-power fiber laser is configured. When setting the laser heating parameters, heat calculation is required. First, the heat input required to reach the expected temperature is calculated based on the physical properties of the steel (such as specific heat capacity, thermal conductivity, density, etc.). Laser heating parameters include laser power, spot size, scanning speed, and scanning path. For typical bridge steel such as Q345 steel, the laser power is usually set in the range of 5-10 kilowatts, and the spot diameter is adjusted between 10-30 millimeters according to the heating area size. The scanning speed is set based on heat input calculation, usually in the range of 5-15 millimeters per second, to ensure sufficient heat deposition without overheating. The scanning path planning considers the heat conduction rules in the structure, usually using a spiral or grid path to ensure uniform heat distribution.

[0062] When laser is directed to specified locations on steel structures to form temperature gradient zones, the laser beam is precisely positioned at the pre-deformation key points through an optical path system. During the heating process, the laser energy is converted into heat, forming a high-temperature zone on the surface of the steel. The heat is conducted to the interior and surrounding areas, forming a temperature gradient. The formation of the temperature gradient zone is critical for pre-deformation, as uneven temperature distribution leads to uneven thermal expansion of the material, resulting in internal stress, which is converted into residual stress and plastic deformation after cooling. The heating control uses a closed-loop method, adjusting the laser parameters in real time through temperature monitoring to ensure that the temperature is controlled within the set range, usually below the phase transition temperature of the steel (about 700-800°C), to avoid changes in material organization affecting the structural performance.

[0063] The mechanical pre-loading system is set according to the pre-deformation amount in the pre-deformation data set, and the process of determining the loading force value involves mechanical calculation. According to the elastic-plastic properties of the material and the temperature-dependent yield strength, the size of the external force required to produce the expected deformation at high temperature is calculated. The mechanical pre-loading system consists of a hydraulic actuator and a control unit, and the force value is set considering the reduction of high-temperature strength of the material, usually 50-70% of the force value required at room temperature. The direction of the applied load is opposite to the expected welding deformation direction to achieve compensation. The calculation of the load value also takes into account the stiffness distribution of the structure, and the load is appropriately increased for areas with high stiffness to ensure that the deformation meets the requirements.

[0064] When the initial deformation is generated by applying an external force in the same direction as the pre-deformation direction to the heated area through the mechanical pre-loading system, precise force and heat coordination control is performed. The application of mechanical force is synchronized with the formation of the temperature field, and the maximum mechanical force is applied when the material reaches the optimal plastic state (temperature usually in the range of 500-600°C), achieving plastic deformation. The application of mechanical force uses a gradual loading strategy, starting at 50% of the target force and gradually increasing to the set value as the temperature rises, avoiding local damage caused by sudden force increase. Under the synergistic action of force and heat, plastic deformation occurs at the pre-deformation key points of the steel structure, and this deformation is preserved after cooling, forming the pre-deformation state.

[0065] Real-time temperature monitoring of the laser heating area using infrared thermal imaging equipment is a key step for precise control. The infrared thermal imager captures the infrared radiation emitted by the steel surface and converts it into a temperature image, displaying the temperature field distribution. The data acquisition frequency is usually 10-20 Hz to ensure timely capture of temperature changes. Temperature data is recorded in matrix form, with each pixel corresponding to a temperature value, forming a temperature field distribution map. Data processing includes noise filtering, emissivity correction, and background temperature elimination to improve temperature measurement accuracy. For large bridge structures, multiple thermal imagers often work together to generate a complete temperature field distribution map through data fusion technology. Based on the temperature distribution data, adjust the laser power and scanning speed to control the pre-deformation progress, and use intelligent control algorithms. The control algorithm first analyzes the temperature distribution data in real time, calculates the deviation of the current temperature from the target temperature, and the temperature change rate. According to the deviation size and trend, adjust the laser power and scanning speed to achieve accurate temperature control. The adjustment strategy uses the PID (Proportional-Integral-Derivative) control principle to calculate the control amount based on the temperature deviation, deviation integral, and deviation rate of change, ensuring that the temperature stabilizes within the target range. When the temperature approaches the preset upper limit, automatically reduce the laser power or increase the scanning speed; when the temperature is below the lower limit, increase the power or decrease the speed. Through this closed-loop control, ensure that the temperature field distribution during pre-deformation meets expectations, thereby achieving precise pre-deformation control.

[0066] The process of measuring the pre-deformed structure using a three-dimensional laser scanner and comparing it with the pre-deformation dataset is a key step in quality verification. The three-dimensional laser scanner measures the spatial coordinates of the structure's surface points by emitting laser light and receiving reflected signals, forming point cloud data of the pre-deformed structure. After filtering, registration, and reconstruction, the point cloud data generates a three-dimensional model of the pre-deformed structure. In the comparison and analysis process, the actual pre-deformation model is compared with the target model in the pre-deformation dataset, and the position deviation and direction deviation of each key point are calculated. The deviation analysis results are used to evaluate the pre-deformation quality, and when the deviation exceeds the allowed range, secondary adjustment is needed until the requirements are met. The final confirmed pre-treated steel component becomes the basis for the next stage of welding operations.

[0067] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0068] Segmenting and dividing the welding area of the pre-treated steel component to generate a welding sequence planning table;

[0069] Setting pulse welding current parameters according to the welding sequence planning table, adjusting pulse frequency and duty cycle;

[0070] Using the symmetric segmented staggered welding method to weld the bridge steel structure to form the initial weld;

[0071] Real-time temperature monitoring of the welding area by the temperature sensor array to obtain temperature field dynamic data;

[0072] Judging the heat cycle threshold based on the temperature field dynamic data to control the welding gap cooling time;

[0073] Real-time tracking of structural deformation during welding using a laser interference deformation monitoring system to record deformation values;

[0074] Dynamic adjustment of welding parameters for subsequent welds according to deformation values to perform compensation welding;

[0075] Integrating temperature field dynamic data and deformation values into time sequence correlation data to generate a welding deformation record table.

[0076] Specifically, segmentation is based on structural stress distribution and deformation sensitivity analysis, and the weld is divided into several paragraphs with similar lengths. For bridge steel structures, the segment length is usually controlled within 300-500 millimeters to avoid excessive local deformation caused by excessive heat input in single segment welding. The welding sequence planning adopts the principle of thermal balance, calculates the heat input and cooling time of each segment of the weld, and arranges the welding sequence to balance the overall structural heat distribution. The welding sequence planning table contains the number, position, length, welding parameters and welding sequence of each segment of the weld, forming a detailed guidance document for welding operations. When adjusting the pulse welding current parameters, the parameters are optimized for the characteristics of each segment of the weld. Pulse welding is a welding technology that uses periodically changing current, with pulse frequency defined as the number of pulses per second, measured in hertz (Hz), and duty cycle being the percentage of the pulse duration in the entire cycle. For bridge steel structure welding, the pulse frequency is usually set within 80-120 Hz, and the duty cycle is between 40%-60%. Parameter setting takes into account factors such as weld position, plate thickness, gap size, etc. For key stress parts such as main beam and cross beam joint nodes, lower pulse frequency and duty cycle are used to reduce heat input and residual stress; while for non-critical parts, the frequency and duty cycle can be appropriately increased to improve welding efficiency.

[0077] The initial weld formation process follows the principle of thermal equilibrium. Symmetrical segmented staggered welding is a method where the welds on both sides of the structure's symmetry plane are welded simultaneously at symmetric positions, and adjacent weld segments are staggered. In practice, welding starts from the center of the structure and progresses symmetrically to both sides. After completing a weld segment, the next weld is performed at the corresponding position on the other side of the structure. This method effectively balances the welding heat input and reduces overall distortion. During staggered welding, a certain distance is maintained between adjacent weld segments to avoid heat concentration. The welding direction of each segment also needs to consider stress distribution, typically following the principle of welding from areas with weaker constraints to areas with stronger constraints. Real-time temperature monitoring of the welding area using a temperature sensor array is a key step in controlling welding quality. The temperature sensor array consists of multiple high-precision thermocouples or infrared temperature sensors arranged along the weld and surrounding areas to form a monitoring network. The sensor data acquisition frequency is typically 5-10 Hz to ensure timely capture of temperature changes. The temperature field dynamic data includes time stamp, spatial coordinates, and temperature value, forming a time-space temperature field matrix. Data processing includes noise filtering, outlier removal, and interpolation calculation to generate continuous temperature field distribution maps. By monitoring the temperature changes in the weld area and heat-affected zone in real time, the thermal cycle state can be determined, providing a basis for subsequent welding parameter adjustment.

[0078] The process of determining the thermal cycle threshold based on temperature field dynamic data and controlling the cooling time of the welding gap involves thermodynamic analysis. The thermal cycle threshold refers to the critical point at which the temperature of the welding area drops to a safe value, typically set at 300-350°C. At this temperature, the plastic deformation capacity of steel significantly decreases, and further welding will not cause excessive cumulative deformation. Cooling time control is based on the rate of change of temperature field data. By calculating the temperature drop rate, the time required to reach the threshold temperature can be predicted. When the temperature of the previous weld area is still above the threshold, the next weld operation is delayed, and the temperature is allowed to decrease. Precise control of cooling time avoids excessive accumulation of heat in the structure, reducing thermal stress concentration and plastic deformation. The laser interference deformation monitoring system is used to track the structure's deformation in real time during welding, and recording the deformation values is a core technology for deformation control. The laser interference deformation monitoring system consists of a laser emitter, a mirror, and an interference signal receiver. By measuring the small changes in the length of the laser path, sub-micron level deformation monitoring accuracy is achieved. Monitoring points are set at key locations of the structure, such as the mid-span of the main beam, the cantilever end, and the support area. Deformation data is recorded in time series, including the displacement values of each monitoring point at different times. Data processing includes interference signal demodulation, baseline drift correction, and temperature compensation to obtain the true structure deformation curve. By comparing the actual deformation with the predicted deformation, the effectiveness of deformation control during welding can be evaluated.

[0079] The process of dynamic adjustment of welding parameters for subsequent welds based on deformation values embodies the principle of adaptive control. When a certain area is monitored to have deformation exceeding expectations, the welding parameters for subsequent related welds are adjusted, including reducing welding current, reducing welding speed, or changing welding sequence. The adjustment strategy is based on the correlation model of deformation and welding parameters, and the parameter sensitivity matrix is established by analyzing historical data to quantify the influence of parameter changes on deformation. For example, for an area with a deformation exceeding expectations by 10%, the welding current of the subsequent weld is reduced by 5-8%, or the pulse frequency is increased by 10-15%, reducing the heat input. Compensation welding also includes increasing cooling measures in deformation-sensitive areas, such as local compressed air cooling or CO2 refrigeration, to accelerate heat dissipation and reduce heat accumulation effects.

[0080] The process of integrating temperature field dynamic data and deformation values into time-series correlation data to generate a welding deformation record table is an important part of data fusion and mining. Time-series correlation data associates temperature data and deformation data through timestamps to establish a temperature field-deformation relationship model. Data integration uses multi-source data synchronization technology to ensure the time consistency of data from different sources. Correlation analysis includes time lag effect calculation, which is the delay time of deformation response relative to temperature change. This parameter is crucial for understanding the heat-deformation coupling mechanism. The welding deformation record table is stored in a structured format, containing multi-dimensional information such as weld information, welding parameters, temperature field data, deformation data, and parameter adjustment records. The record table serves as an important basis for subsequent heat treatment and deformation correction, and provides reference data for similar structure welding deformation control.

[0081] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0082] By analyzing the temperature field distribution and deformation data in the welding deformation record table, the residual stress concentration area in the bridge steel member is determined;

[0083] The residual stress concentration area is marked by grade, and a heat treatment area division map is generated;

[0084] According to the heat treatment area division map, configure the high-frequency induction heating equipment, and set the heating power and frequency parameters;

[0085] Based on the heat treatment area division map, implement gradient heating on the marked area to form a temperature gradient field;

[0086] Use an infrared thermal imaging system to monitor the heating process in real time and obtain a heat treatment temperature curve;

[0087] Use an acoustic emission detection device to monitor the stress release process inside the steel member and record the stress release signals;

[0088] Adjusting the cooling system parameters according to the stress release signal to control the cooling rate;

[0089] The residual stress of the steel member after heat treatment is measured by using an X-ray diffraction device to generate a stress distribution map.

[0090] Specifically, by analyzing the temperature field distribution and deformation data in the welding deformation record table, the residual stress concentration area in the bridge steel member is determined as the premise of precise stress release. The welding deformation record table contains temperature field data of each weld section and deformation data of the corresponding position. The temperature-deformation correlation is extracted by data mining technology to identify the residual stress distribution characteristics. Gradient screening method is used for data analysis. First, the deformation gradient, i.e. the change rate of deformation per unit length, is calculated. The area with high deformation gradient usually corresponds to the residual stress concentration area. At the same time, combined with the temperature field data, the highest temperature and cooling rate in the welding process are analyzed, which are closely related to the formation of residual stress. The specific method is to extract the cooling rate in the temperature-time curve and convert it into the estimated value of residual stress by empirical formula. For bridge steel structure, the residual stress concentration area is mainly distributed in the weld intersection, structural stiffness mutation area and edge of heat affected zone. The stress level in these areas is often close to or even exceeds the yield strength of the material. The residual stress concentration area is classified and labeled, and the process of generating the heat treatment area division map involves region clustering and priority sorting. According to the calculated residual stress estimate value, a three-level classification standard is adopted: the first level is the high-risk area with residual stress exceeding 80% of the material yield strength, the second level is the medium-risk area with residual stress between 50%-80% of the yield strength, and the third level is the low-risk area with residual stress between 30%-50% of the yield strength. Region clustering uses spatial correlation analysis to combine points with similar stress values and adjacent positions into a heat treatment area, avoiding excessive fragmentation. At the same time, considering the importance of the structure, the stress area of the key load-bearing parts is upgraded. The heat treatment area division map shows the stress level and boundary of each area in the form of color coding. The first level is marked in red, the second level in yellow, and the third level in green, forming an intuitive guide map for heat treatment operation.

[0091] The process of configuring high-frequency induction heating equipment according to the heat treatment region division map needs to consider the material properties and region geometry. High-frequency induction heating is a technology that uses electromagnetic induction principles to generate induced current and eddy current loss in metal through high-frequency alternating current, achieving non-contact heating. For different levels of heat treatment regions, different heating parameters are set: the first level region uses higher power (30-40 kW) and longer holding time (20-30 minutes), the second level region uses medium power (20-30 kW) and medium holding time (15-20 minutes), and the third level region uses lower power (10-20 kW) and shorter holding time (10-15 minutes). The selection of frequency parameters takes into account the skin effect of induced current, with lower frequency (8-12 kHz) for thick plate regions to increase heating depth and higher frequency (20-30 kHz) for thin plate regions to improve heating efficiency. The shape and size of the induction coil are customized according to the geometric characteristics of the heat treatment region to ensure heating uniformity. Gradient heating of the marked region based on the heat treatment region division map is the core step of stress release. Gradient heating refers to the formation of a temperature distribution that decreases from the center to the outside within the heat treatment region, rather than uniform heating. In specific operation, the center of the first level region is heated to the target temperature (550-650℃) first, and then the heating range is expanded outward to form a temperature gradient of 30-50℃ / cm. This gradient heating method can generate internal stress during heating and guide the redistribution and release of residual stress in the softened state of the material. The heating process strictly controls the temperature rise rate, usually 150-200℃ / min, to avoid rapid heating that leads to new thermal stress. For multiple adjacent heat treatment regions, multi-point synchronous heating technology is used to design the optimal heating point position and heating sequence according to the structural symmetry and stiffness distribution, ensuring balanced heating of the overall structure and avoiding new deformation during heating.

[0092] Real-time monitoring of the heating process using an infrared thermal imaging system is a key technology for temperature control. The infrared thermal imaging system captures the infrared radiation emitted by the object's surface and converts it into a temperature distribution image. During monitoring, the thermal imager collects images at a frequency of 5-10 frames per second, covering the entire heat treatment region. Temperature data is stored in matrix form, with each pixel corresponding to a temperature value. Data processing includes radiation rate correction, background temperature compensation, and noise filtering to improve temperature measurement accuracy. The heat treatment temperature curve records the temperature change of each monitoring point over time, including the heating, holding, and cooling stages. Temperature curve analysis focuses on several key parameters: heating rate, maximum temperature, holding time, and cooling rate, which directly affect stress release. When the temperature curve deviates from the preset target, the heating power is dynamically adjusted through a closed-loop control system to ensure temperature control within the set range.

[0093] Recording stress release signals by acoustic emission detection equipment is a direct method to judge the sufficiency of stress release. Acoustic emission detection is a non-destructive testing technology that uses transient elastic waves generated by the release of elastic energy under stress to detect materials. During heat treatment, as the temperature rises and the holding time extends, residual stress gradually releases, producing tiny dislocation slip and micro-crack closure. These micro-changes will produce acoustic emission signals. Acoustic emission sensors are attached around the heat treatment area to collect acoustic signals in real time, with a typical sampling frequency of 1-2 MHz. The acoustic signals are pre-amplified, filtered, and feature-extracted to extract characteristic parameters such as signal amplitude, frequency, energy, and duration. During stress release, acoustic emission signal intensity and frequency decrease as the release degree increases. When the signal intensity drops to 20-30% of the initial value and the signal frequency continues to decrease, it indicates that stress release has been completed, and the cooling stage can begin.

[0094] Adjusting cooling system parameters based on stress release signals and controlling the cooling rate is a key step to prevent new stress from occurring. The cooling control adopts a segmented variable speed strategy, with a slower cooling rate (10-15°C / min) in the high temperature section (650-450°C), a higher cooling rate (15-25°C / min) in the medium temperature section (450-300°C), and natural cooling in the low temperature section (below 300°C). The cooling system includes air cooling and water cooling devices, and the cooling rate is controlled by adjusting the flow and temperature of the cooling medium. Cooling parameter adjustment is based on the trend of acoustic emission signals. When a sudden increase in acoustic emission signal intensity is detected, it indicates that new stress concentration has occurred during cooling, and the cooling rate needs to be reduced immediately. If necessary, reheat to eliminate the newly generated stress. Continuous monitoring of structural deformation during cooling ensures that the deformation caused by cooling is within a controllable range.

[0095] X-ray diffraction device is used to measure the residual stress of steel components after heat treatment, and the stress distribution map is generated to evaluate the stress release effect. X-ray diffraction method uses the principle of crystal diffraction to calculate the stress state inside the material by measuring the change of crystal plane spacing. During measurement, the X-ray beam irradiates the material surface, and according to Bragg's law, a diffraction pattern is generated. By analyzing the change of diffraction angle, the lattice strain is calculated, and then the stress value is converted. The arrangement of residual stress measurement points uses the grid method, with multiple measurement points arranged in the heat treatment area and its surrounding area. Typically, 5-10 measurement points are set in each heat treatment area to form a measurement network. Each measurement point measures the stress component in different directions to obtain the stress tensor. The measurement data is corrected and interpolated to generate a continuous stress distribution map, which intuitively displays the stress size and direction. The stress distribution map is represented in the form of color contour lines, with different colors corresponding to different stress levels, making it easy to visually judge the stress release effect.

[0096] The heat treatment of the main beam and cross beam connecting node of a certain steel box girder bridge is taken as an example to illustrate the whole process: First, analyze the welding deformation record table, find that the temperature peak of the node area reaches 1480℃, the cooling rate is 45℃ / s, and the deformation amount is 5.8mm, which is much higher than the average level of other areas. Through data analysis, the estimated value of residual stress in this area is calculated as 345MPa, which is close to 95% of the yield strength of Q345 steel, and is marked as a first-level stress concentration area. At the same time, the second and third areas are divided around the node to form a heat treatment area division diagram. According to the division diagram, a 35kW high-frequency induction heating equipment is configured, the frequency of the first area is set to 10kHz, the second area is 15kHz, and the third area is 25kHz. When implementing gradient heating, the center of the first area is heated to 600℃ first, and then gradually expanded outward to form a temperature gradient field of 600℃ in the center and 450℃ on the edge. The infrared thermal imaging system monitors the temperature distribution throughout the process and records that the heating rate is 180℃ / min, and the temperature stabilizes at 595-605℃ during the holding stage. At the same time, the acoustic emission sensor monitors that after 8 minutes of heating, the signal strength begins to decrease significantly, and after 15 minutes, it decreases to 25% of the initial value, indicating that the stress release is sufficient. According to the acoustic emission signal change, the cooling rate is controlled at 12℃ / min, and after cooling to 300℃, it is naturally cooled. Finally, through X-ray diffraction method, it is found that the maximum residual stress of the node area after heat treatment is reduced to 125MPa, which is about 36% of the material yield strength, and the average stress of the heat treatment area is reduced by 68%, the generated stress distribution map shows that the stress distribution is more uniform, and the high stress area is significantly reduced, and the heat treatment achieves the expected effect.

[0097] In a specific embodiment, the process of performing step S105 can specifically include the following steps:

[0098] comprehensively scanning the bridge steel structure after heat treatment by a high-precision three-dimensional laser scanning system to obtain actual deformation point cloud data;

[0099] comparing the actual deformation point cloud data with the design model to generate a deformation vector diagram;

[0100] determining correction key points according to the deformation vector diagram and the stress distribution diagram to form a correction point distribution table;

[0101] designing a multi-point synchronous correction scheme based on the correction point distribution table to determine the correction force size and direction;

[0102] configuring parameters of the multi-point force correction equipment to set the loading sequence and loading rate of the hydraulic servo actuators;

[0103] implementing progressive loading on the deformation area by the multi-point force correction equipment to form a preliminary correction state;

[0104] The stress change in the correction process is monitored in real time by using a strain monitoring system, and a stress change curve is recorded.

[0105] The correction process is adjusted based on the stress change curve until the deformation is restored to within the design tolerance range, and shape correction data is generated.

[0106] Specifically, the bridge steel structure after heat treatment is comprehensively scanned by a high-precision three-dimensional laser scanning system to obtain actual deformation point cloud data, which is the basis for accurate correction. The high-precision three-dimensional laser scanning system uses phase distance measurement principle, calculates the three-dimensional coordinates of space points by emitting laser and receiving reflected signals, and calculates the time difference and phase difference of laser propagation. During the scanning process, the laser emitter moves along the preset path to form a scanning grid and collect the space point cloud of the structure surface. The point cloud data density is usually 5000-10000 points per square meter, and the measurement accuracy reaches ±0.1 millimeter. After data collection, preliminary processing is carried out, including noise point filtering, outlier rejection and coordinate system conversion. The statistical outlier analysis method is used for noise point filtering, the average distance of each point and its adjacent points is calculated, and the points deviating from the average distance more than three times the standard deviation are identified as noise points and deleted. Coordinate system conversion converts the local coordinate system data obtained by scanning into the bridge design coordinate system, which is convenient for subsequent comparison and analysis.

[0107] The process of comparing the actual deformation point cloud data with the design model to generate the deformation vector diagram is the key link of deformation quantification. The design model usually exists in the form of three-dimensional CAD model or finite element model, containing the ideal geometric shape and size information of the structure. The comparison and analysis first carries out the registration of point cloud and model, adopts the iterative closest point (ICP) algorithm, minimizes the distance and normal deviation of point cloud data and design model surface, and realizes the accurate alignment of the two. After registration, the distance and direction of each point in the point cloud to the corresponding position of the design model are calculated, and the deformation amount and deformation direction data are formed. The deformation vector diagram intuitively displays the deformation state of each part of the structure in color coding, red represents the serious deformation area exceeding the tolerance, yellow represents the critical deformation area, and green represents the area within the tolerance range. The arrow length of the deformation vector represents the deformation amount, and the arrow direction represents the deformation direction, which intuitively shows the deformation distribution characteristics of the structure.

[0108] The selection of the correction key points involves multi-criteria decision analysis. The selection of the correction key points needs to consider the three factors of deformation, stress level and structural importance. First, the extreme value analysis is performed on the deformation vector diagram to identify the areas with deformation exceeding the design tolerance as candidate areas. Then, in combination with the stress distribution diagram, the areas with low stress level (usually less than 50% of the material yield strength) are screened out, which are not prone to new plastic deformation in the correction process. Finally, considering the structural mechanics characteristics, the positions with moderate structural stiffness and clear stress path are selected as the correction points to avoid applying correction force in the areas with sudden stiffness change. The density distribution of the correction points is related to the deformation gradient, and more correction points need to be set in areas with large deformation gradient. The correction point distribution table records the coordinate position, corresponding deformation, deformation direction and local structural stiffness information of each correction point, providing a data basis for subsequent correction scheme design. Based on the correction point distribution table, a multi-point synchronous correction scheme is designed, and the determination of the correction force size and direction is the core technology of the correction process. The correction scheme design adopts the structural mechanics back calculation method, taking the target deformation as the known condition to inversely calculate the external force required to produce the deformation. The calculation process considers the stiffness matrix and deformation influence matrix of the structure to establish the mapping relationship between the external force and the deformation. The determination of the correction force size considers the elastic-plastic properties of the material, and is usually controlled near the critical value of producing slight plastic deformation, which is about 90% of the yield strength for Q345 steel. The direction of the correction force is opposite to the deformation direction, but needs to consider the structural connection constraints and overall stiffness distribution, and sometimes needs to apply non-directly opposing force to achieve deformation correction. The multi-point synchronous correction scheme also includes the grouping and loading sequence of the correction points, which are usually divided into several groups according to the symmetry and stiffness distribution of the structure, and a synchronous or sequential loading strategy is designed to ensure that the structure is balanced in the correction process and avoid local stress concentration.

[0109] Parameter configuration of the multi-point force correction equipment, setting the loading sequence and loading rate of the hydraulic servo actuators, is the technical guarantee for implementing accurate correction. The multi-point force correction equipment is composed of multiple groups of hydraulic servo actuators, force sensors, displacement sensors and control units, and can provide accurate and controllable correction force. Parameter configuration includes maximum force value setting of the actuator, force control mode selection, loading rate setting and safety limit setting. The force value of the hydraulic servo actuator is usually set to 1.1-1.2 times the theoretical calculation value to reserve a margin for overcoming the deviation between the actual structural stiffness and the calculated stiffness. The loading sequence setting follows the principles of "overall first, local second", "main first, secondary second" and "symmetry first, asymmetry second" to ensure that the structure is balanced in the correction process. Loading rate control is the key to accurate correction. A lower loading rate (usually 2-5% of the maximum force per minute) is used in the initial stage, and the rate is gradually reduced as the correction proceeds, and a very low rate (0.5-1% of the maximum force per minute) is used near the target deformation to achieve accurate control.

[0110] The process of applying progressive loading to the deformation area by multi-point force correction equipment embodies the concept of precise control. Progressive loading refers to gradually increasing the correction force in multiple steps, rather than applying the full force value at once. Typically, 5-8 loading steps are used, with the initial loading being 30-50% of the target force, and each subsequent step increasing by 10-15% until the calculated correction force value is reached. After each loading step, the force value is maintained stable for a period of time (3-5 minutes), and the structural deformation response is observed to verify the consistency of the actual deformation with the expected deformation. Progressive loading avoids structural vibration and local stress concentration caused by sudden force increase, and provides the structure with adaptation time, making the stress distribution more uniform. When implementing multi-point synchronous correction, the loading progress of each actuator needs to be coordinated to ensure that the structure is always in a state of force balance during the correction process, which is usually achieved by a central control unit adjusting the loading rate of each actuator in real time.

[0111] Real-time monitoring of stress changes during the correction process using a strain monitoring system and recording the stress change curve are important means to prevent over-correction. The strain monitoring system is composed of multiple strain gauges, signal conditioning modules and data acquisition units, and can measure the strain changes on the structure surface in real time. Strain gauges are arranged around the key correction points and stress-sensitive areas of the structure, usually in a three-axis rose arrangement to measure strain components in different directions. The strain data acquisition frequency is usually 5-10 Hz to ensure that the subtle process of stress change is captured. After filtering, temperature compensation and zero correction, the original strain data is converted to stress data. The stress change curve records the trend of stress changes at each monitoring point over time and loading force during the correction process, reflecting the response characteristics of the structure to the correction force. Key parameters of stress monitoring include maximum stress value, stress change rate and stress distribution uniformity, which directly reflect the safety status of the structure during the correction process.

[0112] The adjustment of the correction process based on the stress change curve until the deformation is restored to the design tolerance range embodies the closed-loop control thought. The correction adjustment adopts a real-time feedback control strategy, and the size and direction of the correction force are dynamically adjusted by comparing the difference between the actual deformation and the target deformation. When the stress of a certain region is close to the material yield strength, the growth rate of the correction force of the region is reduced or the correction force is stopped increasing to prevent excessive plastic deformation. At the same time, combined with the deformation monitoring data, when the deformation is close to 80-90% of the target value, the fine adjustment stage is entered, and a lower loading rate and a smaller loading increment are adopted to accurately control the final deformation. The termination condition of the correction process is that the structural deformation is restored to the design tolerance range, and the stress level is lower than 70% of the material yield strength. After the correction is completed, the correction force is kept stable for a period of time (usually 2-4 hours) to allow the internal stress of the structure to be fully redistributed, and then the structure is slowly unloaded to avoid rebound deformation. The shape correction data records the deformation comparison before and after the correction, the stress change process and the correction force parameters, which provides a reference for the deformation correction of similar structures in the future.

[0113] The entire process is illustrated by taking the deformation correction of the main girder of a steel box girder bridge as an example. First, a high-precision three-dimensional laser scanning system is used to fully scan the main girder after heat treatment to obtain point cloud data containing about 2 million points. Through data processing, the actual geometric shape of the main girder is obtained, and compared with the design model, it is found that the mid-span sag deformation of the main girder reaches 32 mm, exceeding the design tolerance (10 mm). According to the deformation vector diagram and the stress distribution diagram, 7 correction key points are determined, including 3 points in the mid-span, 2 points in each of the 1 / 4 spans, forming a correction point distribution table. Based on the distribution table, a multi-point synchronous correction scheme is designed, and it is calculated that the upward correction force of 40 tons needs to be applied to the mid-span points, and the force of 25 tons needs to be applied to each of the 1 / 4 span points. According to the correction scheme, hydraulic servo actuators are configured, and the maximum force value is set to 1.15 times the calculated value, and the loading rate is 3% of the maximum force per minute. The correction process adopts a 6-step progressive loading, and the initial loading is 40% of the target force, and each step increases by 12%. The stress change in the correction process is monitored in real time through the strain monitoring system, and it is found that the stress of the main girder in a certain region reaches 320 MPa (close to the yield strength of Q345 steel) at the 4th step of loading, and the correction force growth rate of the region is immediately adjusted to 1 / 3 of the original. After fine adjustment, the mid-span sag deformation of the main girder is restored to 9 mm, which is within the design tolerance range. After maintaining the correction force for 3 hours, the structure is slowly unloaded, and the final shape is measured. The deformation of the main girder is stable at 8 mm.

[0114] In a specific embodiment, the process of performing step S106 can specifically include the following steps:

[0115] The full-size measurement of the corrected bridge steel structure is performed by high-precision three-dimensional laser scanning technology to obtain a measured geometric model;

[0116] The measured geometric model is subjected to pushover analysis with the original design model to calculate the size deviation value of the key parts.

[0117] Based on the size deviation value, a deformation contour map of the bridge steel structure is drawn to determine the deformation distribution state.

[0118] The portable X-ray diffractometer is used to measure the residual stress of the key welds and heat-affected zones, and a stress distribution report is generated.

[0119] The ultrasonic phased array detection equipment is used to perform non-destructive testing on the inside of the welds, and the weld quality data is recorded.

[0120] The vibration modal test is performed on the bridge steel structure to obtain the dynamic characteristic parameters of the structure.

[0121] Based on the dynamic characteristic parameters of the structure, the static load test is performed on the key connection nodes to measure the deformation recovery performance.

[0122] The morphological correction data and the whole-process technical parameters are integrated to establish a digital twin model, forming a deformation control file.

[0123] Specifically, the full-size measurement of the corrected bridge steel structure is performed by high-precision three-dimensional laser scanning technology, and the measured geometric model is the first step of quality acceptance. High-precision three-dimensional laser scanning technology is a non-contact measurement method based on laser ranging principle, which has the characteristics of fast measurement speed, high precision and wide coverage. During the scanning process, the laser measurement equipment moves along the preset trajectory, and the spatial point cloud data of the structure surface is collected. For large bridge steel structures, multiple measurement stations are usually set up for scanning, and the data of each station is spliced into a complete point cloud model through common target points. Point cloud processing includes three steps of denoising, simplification and reconstruction. The statistical outlier analysis method is used in the denoising process to calculate the average distance of each point to the adjacent points and remove the points with large deviation values. The uniform sampling or curvature sampling algorithm is used for point cloud simplification to reduce redundant points while maintaining geometric features. Three-dimensional reconstruction converts discrete point cloud into continuous surface model through point cloud to surface technology, forming the measured geometric model. The measured model is stored in the form of triangular mesh or NURBS surface, accurately describing the actual shape of the structure.

[0124] The measured geometric model is compared with the original design model in a pushover analysis to calculate the size deviation values at key locations, which is a quantitative evaluation of the deformation control effectiveness. The pushover analysis first performs accurate registration of the two models using a reference point registration method or an iterative closest point algorithm to ensure that the two models are compared in the same coordinate system. After registration, the differences between the measured and designed values are calculated for the key control points of the bridge (such as the mid-span point of the main girder, the support area, the node connection, etc.) to obtain the size deviation values. The deviation calculation uses the normal distance method, which measures the distance from the measured point to the design surface along the normal direction of the design model surface. This method can more accurately reflect the actual situation of structural deformation. For bridge steel structures, the key control indicators include the main girder linear deviation, transverse inclination, and node position offset. The pushover analysis results are presented in numerical tables and color-coded visual images, which intuitively display the deviation state of each part of the structure.

[0125] Based on the size deviation values, a bridge steel structure deformation contour map is drawn to determine the deformation distribution state, which is a direct expression of the deformation mode. The deformation contour map is a curve graph formed by connecting points with the same deviation value, similar to the contour lines of a topographic map. The drawing process first performs spatial interpolation on the deviation data to generate a continuous deviation field distribution, and then extracts the contour lines according to the preset contour interval. The interpolation algorithm usually uses the Kriging method or the radial basis function method, which can generate smooth transitions between discrete sampling points. The contour interval is set according to the structure size and deformation magnitude, usually 1 / 5-1 / 3 of the design tolerance. The deformation contour map uses color coding to enhance the visual effect, with red representing positive deviation (measured greater than designed) and blue representing negative deviation (measured less than designed), and the color depth representing the deviation magnitude. Through the deformation contour map, the concentrated areas of deformation, the gradient direction of deformation, and the overall deformation mode can be clearly identified, providing important references for the final acceptance and future maintenance of the structure.

[0126] Residual stress measurement of key welds and heat-affected zones using a portable X-ray diffractometer generates a stress distribution report, which is an important means of internal quality assessment. The X-ray diffraction method is based on Bragg's law, which calculates the stress state inside the material by measuring the change in lattice spacing. During measurement, the X-ray beam is incident on the material surface, and the change in diffraction angle reflects the degree of lattice distortion, which is then used to calculate the stress value. The portable X-ray diffractometer is designed in a small size and can perform non-destructive measurement on the structure on site. The measurement point arrangement uses a combination of grid method and key point method, taking points at the center of the weld, the edge of the heat-affected zone, and the matrix area to form a measurement network. Multiple stress components are measured at each location in multiple directions to obtain stress tensor information. The measurement data is statistically processed and spatially interpolated to generate a continuous stress distribution map. The stress distribution report includes numerical data, distribution images, and statistical analysis results, which evaluate the absolute level of residual stress, the uniformity of distribution, and the ratio to the material yield strength, and judge the effect of stress release.

[0127] Ultrasonic phased array testing (EAR) is an advanced non-destructive testing technology that uses controlled ultrasonic waves generated by multiple piezoelectric elements to detect internal defects in materials. Compared to traditional ultrasonic testing, EAR technology offers advantages such as faster scanning speed, more intuitive defect imaging, and more precise defect localization. The testing process begins with equipment calibration, using standard test blocks to determine sound velocity, sensitivity, and resolution. Then, a scan is performed along the weld direction, recording the ultrasonic echo signals. Signal processing includes filtering, gain adjustment, and deconvolution to improve the signal-to-noise ratio and resolution. Data analysis employs defect feature extraction algorithms to identify defect types, such as porosity, inclusions, and lack of fusion, based on the amplitude, duration, and phase characteristics of the echo signals. Weld quality data is presented as B-scan and C-scan images, visually displaying the location, size, and distribution of defects, and is used to assess the weld quality level in accordance with relevant standards.

[0128] Vibration modal testing of bridge steel structures to obtain dynamic characteristic parameters is a scientific method for overall performance evaluation. Vibration modal testing identifies dynamic characteristic parameters such as natural frequencies, mode shapes, and damping ratios by measuring the structure's vibration response under excitation. Testing employs either environmental or artificial excitation methods. Environmental excitation utilizes natural excitation sources such as wind loads and traffic loads, while artificial excitation applies known inputs using impact hammers or vibration exciters. Measurement equipment includes accelerometers, data acquisition systems, and signal analyzers. Sensor placement follows a node densification principle, placing sensors at key structural locations to capture major vibration modes. Data processing uses Fast Fourier Transform to convert time-domain signals to the frequency domain, and extracts natural frequencies and mode shapes through peak identification and modal parameter fitting algorithms. Dynamic characteristic parameters reflect the structure's stiffness distribution, mass distribution, and boundary conditions. By comparing these parameters with finite element model predictions, the overall performance and connection stiffness of the structure are evaluated, and the impact of welding and deformation control on the structural dynamics is verified.

[0129] Static load test on key connection nodes based on structural dynamic characteristics parameters is a direct test of structural reliability. Static load test is a method of applying known loads to a structure and measuring its deformation response to assess its load-carrying capacity and deformation characteristics. The test load is usually set to 1.2 times the design load and is applied by means of weights, water tanks or jacks. Deformation measurement uses displacement sensors or high-precision levels to record the deformation curve during loading and the residual deformation after unloading. Data analysis focuses on three key indicators: maximum deformation, deformation recovery rate and linearity of load-deformation curve. The deformation recovery rate is defined as the ratio of rebound deformation to total deformation after unloading, reflecting the elastic working state of the structure. The linearity of the load-deformation curve is evaluated by the correlation coefficient and the root mean square error of deviation from linearity. By comparing the static load test results before and after welding deformation control, the improvement effect of deformation control on the service performance of the structure is quantitatively evaluated.

[0130] Integrating morphological correction data with whole-process technical parameters to establish a digital twin model and form a deformation control archive is an important part of knowledge accumulation and experience inheritance. The digital twin model is a virtual mapping of the physical object in the digital world, containing geometric model, physical properties and behavior characteristics. Model construction first integrates whole-process data, including pre-deformation design data, welding process records, heat treatment parameters, correction process data and final test results. Data integration uses time series correlation and spatial position mapping to establish the correlation between different stages of data. The modeling process uses a parameterized method to establish a mapping relationship between key process parameters and deformation control effects, forming a parameter sensitivity model. The digital twin model not only records the current state of the structure, but also contains the technical parameters and control strategies of the whole deformation control process, with the ability to analyze and predict. The deformation control archive is stored in the form of database and knowledge base, containing multi-dimensional information such as structural characteristics, process parameters, control methods and quality evaluation, providing scientific basis and experience reference for subsequent deformation control of similar structures.

[0131] The final acceptance of the main girder of a steel box girder bridge is taken as an example to illustrate the whole process. First, a high-precision three-dimensional laser scanner is used to measure the full size of the corrected main girder, and about 3 million point cloud data are collected. After processing, a measured geometric model with an accuracy of ±0.2 mm is generated. The measured model is compared with the design model for out-of-limit analysis, and it is found that the maximum deflection deviation of the main girder in the center is 7 mm, which is within the design tolerance (±10 mm); the maximum torsion angle deviation of the main girder is 0.08 degrees, which is less than the specification limit (0.1 degrees). Based on the deviation data, a deformation contour map is drawn, which shows that the deformation is mainly concentrated on one side of the middle of the main girder, showing an asymmetric distribution characteristic. Then, a portable X-ray diffractometer is used to measure the residual stress of the 8 welds in the connection area between the main girder and the cross beam, and the results show that the average residual stress is 112 MPa, which is about 32% of the yield strength of Q345 steel, which is much lower than the measured value (315 MPa) before deformation control. The ultrasonic phased array detection equipment detects the internal quality of the weld, and no defects above grade II affecting the safety of the structure are found. The first six natural frequencies of the main girder are obtained by vibration modal testing, and the average deviation from the finite element analysis results is 4.2%, indicating that the overall stiffness of the structure meets the design requirements. The static load test of the main cross beam connection joint under 1.2 times the design load shows that the maximum deformation is 92% of the design value, and the residual deformation after unloading is only 3.5% of the total deformation, indicating good deformation recovery performance. Finally, the technical parameters and test data of the whole deformation control process are integrated to establish a digital twin model, which records the complete information of process parameters, control strategies and effect evaluation, and forms a systematic deformation control file, providing valuable technical reference for subsequent welding deformation control of bridge steel structures.

[0132] The above describes the multi-stage adjustment method for welding deformation control of steel structures in the embodiments of the present application. The multi-stage adjustment system for welding deformation control of steel structures in the embodiments of the present application is described below. Please refer to Figure 2 An embodiment of the multi-stage adjustment system for welding deformation control of steel structures in the embodiments of the present application includes:

[0133] A measurement module is configured to establish a welding deformation prediction model by measuring the geometric parameters of the steel structure and combining the welding process parameters, and obtain a pre-deformation data set;

[0134] A processing module is configured to perform accurate pre-deformation processing on the specified position of the steel structure according to the pre-deformation data set, and obtain a pre-processed steel structure;

[0135] A recording module is configured to perform a segmented welding operation on the pre-processed steel structure, record the temperature field distribution and real-time deformation amount in the welding process, and obtain a welding deformation record table;

[0136] An implementation module is configured to perform local heat treatment on the high-stress area of the welded steel structure according to the welding deformation record table, and obtain a stress distribution map.

[0137] an applying module configured to apply a correction force to the deformed area by a multi-point force correction device according to the stress distribution map, to obtain shape correction data;

[0138] an evaluating module configured to perform final precision detection and quality evaluation on the steel structure by using the shape correction data, to form a deformation control file.

[0139] Through the cooperation of the above components, by measuring the geometric parameters of the steel structure and combining the welding process parameters, a welding deformation prediction model is established, the traditional empirical prediction is changed into data-driven accurate calculation, the accuracy of the pre-deformation calculation is greatly improved, and the over-compensation or under-compensation phenomenon is avoided. Meanwhile, the generation of the pre-deformation data set provides accurate guidance for the subsequent process. According to the pre-deformation data set, the specified position of the steel structure is accurately pre-deformed, and the composite technology of laser heating and mechanical preloading is combined to realize the accurate control of the deformation amount and the deformation direction. The pre-processed steel member lays a good foundation for the subsequent welding work. The segmented welding operation is performed on the pre-processed steel member, the symmetric segmented staggered welding method is used to effectively balance the heat input distribution, the overall deformation trend is significantly reduced, and the temperature field distribution and the deformation amount are recorded in real time. The welding deformation record table formed provides accurate data support for the subsequent heat treatment. According to the welding deformation record table, local heat treatment is performed on the high stress area, the heating temperature and the cooling rate are accurately controlled, the residual stress is effectively released, and the stress distribution map intuitively reflects the stress release effect. According to the stress distribution map, a correction force is applied to the deformed area by a multi-point force correction device, an incremental loading strategy and real-time stress monitoring are adopted, and the deformation correction is safely and efficiently realized. The shape correction data formed records the correction process and effect. The shape correction data is used to perform final precision detection and quality evaluation on the steel structure, and a complete deformation control file is formed. The deformation control effect is verified, and valuable experience reference is provided for the subsequent similar structure. Especially in the welding deformation prediction and control process, the data mining algorithm applied can extract the deformation law from historical cases, establish the correlation model between the process parameters and the deformation results, and continuously optimize the prediction accuracy by using the deep learning method. The prediction result can adapt to different structure forms and welding conditions, which greatly improves the pertinence and effectiveness of the deformation control.

[0140] Reference Figure 3 In the embodiment of the present application, a computer device, which can be a server, is also provided. The internal structure of the computer device can be as shown in Figure 3The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store corresponding data in the embodiment. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement the above method.

[0141] Those skilled in the art can understand that, Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied.

[0142] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program. The computer program is executed by the processor to implement the above method. It can be understood that the computer readable storage medium in the embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0143] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to the memory, storage, database or other medium provided by the present application and used in the embodiments can include non-volatile and / or volatile memory. The non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. The volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM, etc.

[0144] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0145] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0146] The above-described and above-mentioned embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-stage adjustment method for controlling welding distortion of a steel structure, characterized by, The multi-stage adjustment method for controlling welding deformation of the steel structure comprises: A welding deformation prediction model is established by measuring the geometric parameters of the steel structure and combining the welding process parameters to obtain a pre-deformation data set, including: performing three-dimensional scanning on the bridge steel structure to obtain the geometric parameters and key node coordinates of the structure; inputting the geometric parameters and welding process parameters into a processing system to construct a welding deformation prediction model; dividing the welding deformation prediction model into grid elements, and performing grid encryption processing on the weld area to obtain fine grid of the welding area; based on the fine grid of the welding area, performing thermal cycle analysis on the steel structure welding process to obtain temperature field distribution data; calculating the welding stress and deformation amount by the welding deformation prediction model according to the temperature field distribution data to generate deformation prediction data; correcting the deformation prediction data under the bridge load condition to consider the influence of the structure weight and service load to obtain the corrected deformation value; determining the pre-deformation compensation amount based on the corrected deformation value to calculate the pre-deformation position parameters of each node; comparing the pre-deformation position parameters with the original parameters to generate a pre-deformation data set containing deformation direction, deformation amount and implementation position; According to the pre-deformation data set, precise pre-deformation processing is performed on the specified position of the steel structure to obtain a pre-processed steel member, including: determining the pre-deformation key points of the bridge steel structure according to the pre-deformation data set to generate a pre-deformation operation guide; configuring a high-power fiber laser based on the pre-deformation operation guide and setting laser heating parameters; performing directional laser heating on the specified position of the steel structure to form a temperature gradient area; setting a mechanical preloading system according to the pre-deformation amount in the pre-deformation data set to determine the loading force value; applying an external force consistent with the pre-deformation direction to the heating area by the mechanical preloading system to generate initial deformation; using an infrared thermal imaging device to monitor the temperature of the laser heating area in real time to obtain temperature distribution data; adjusting the laser power and scanning speed based on the temperature distribution data to control the pre-deformation progress; using a three-dimensional laser scanner to measure the pre-deformed structure and comparing it with the pre-deformation data set to obtain a pre-processed steel member; Performing segmented welding operation on the pre-processed steel member, recording the temperature field distribution and real-time deformation amount during the welding process to obtain a welding deformation record table; According to the welding deformation record table, local heat treatment is performed on the high stress area of the welded steel member to obtain a stress distribution map; According to the stress distribution map, a multi-point force correction device is used to apply a correction force to the deformation area to obtain shape correction data; Using the shape correction data, the steel structure is subjected to final precision detection and quality evaluation to form a deformation control file.

2. The multi-stage adjustment method of steel structure welding deformation control according to claim 1, characterized in that, The segmented welding operation on the pre-processed steel member, recording the temperature field distribution and real-time deformation amount during the welding process to obtain a welding deformation record table, comprises: Segmented division is performed on the welding area of the pre-processed steel member to generate a welding sequence planning table; According to the welding sequence planning table, pulse welding current parameters are set to adjust the pulse frequency and duty cycle; The bridge steel structure is welded by using a symmetric segmented staggered welding method to form an initial weld; Real-time temperature monitoring of the welding area is performed by a temperature sensor array to obtain dynamic temperature field data; A heat cycle threshold is determined based on the dynamic temperature field data to control the welding gap cooling time; A laser interferometric deformation monitoring system is used to track the structural deformation in real time during welding to record the deformation values; The welding parameters of the subsequent welds are dynamically adjusted based on the deformation values to perform compensation welding; The dynamic temperature field data and deformation values are integrated into time sequence correlation data to generate a welding deformation record table.

3. The multi-stage adjustment method of steel structure welding distortion control according to claim 1, wherein, According to the welding deformation record table, local heat treatment is performed on the high stress area of the welded steel member to obtain a stress distribution map, including: By analyzing the temperature field distribution and deformation data in the welding deformation record table, the residual stress concentration area in the bridge steel member is determined; The residual stress concentration area is marked in stages to generate a heat treatment area division map; According to the heat treatment area division map, a high-frequency induction heating device is configured, and the heating power and frequency parameters are set; Based on the heat treatment area division map, gradient heating is performed on the marked area to form a temperature gradient field; An infrared thermal imaging system is used to monitor the heating process in real time to obtain a heat treatment temperature curve; An acoustic emission detection device is used to monitor the stress release process inside the steel member to record the stress release signals; According to the stress release signals, the cooling system parameters are adjusted to control the cooling rate; An X-ray diffraction device is used to measure the residual stress of the steel member after heat treatment to generate a stress distribution map.

4. The multi-stage adjustment method of steel structure welding distortion control according to claim 1, characterized in that, According to the stress distribution map, a multi-point force correction device is used to apply a correction force to the deformation area to obtain shape correction data, including: A high-precision three-dimensional laser scanning system is used to fully scan the bridge steel structure after heat treatment to obtain actual deformation point cloud data; The actual deformation point cloud data is compared with the design model to generate a deformation vector map; According to the deformation vector map and the stress distribution map, key correction points are determined to form a correction point distribution table; Based on the correction point distribution table, a multi-point synchronous correction scheme is designed to determine the correction force size and direction; The multi-point force correction device is parameter configured to set the loading sequence and loading rate of the hydraulic servo actuators; The multi-point force correction device is used to implement gradual loading on the deformation area to form a preliminary correction state; A strain monitoring system is used to monitor the stress changes during the correction process in real time to record the stress change curve; Based on the stress change curve, the correction process is adjusted until the deformation is restored to within the design tolerance range to generate shape correction data.

5. The multi-stage adjustment method of steel structure welding distortion control according to claim 1, wherein, The shape correction data is used to perform final precision detection and quality evaluation on the steel structure to form a deformation control file, including: A high-precision three-dimensional laser scanning technology is used to measure the full size of the corrected bridge steel structure to obtain a measured geometric model; The measured geometric model is compared with the original design model for overrun analysis to calculate the size deviation value of the key parts; Based on the size deviation value, a bridge steel structure deformation contour map is drawn to determine the deformation distribution state; The residual stress of the key weld and heat-affected zone is measured by a portable X-ray diffractometer, and a stress distribution report is generated; The internal weld is non-destructively tested by an ultrasonic phased array detection device, and weld quality data is recorded; The bridge steel structure is tested for vibration modal, and the dynamic characteristic parameters of the structure are obtained; Based on the dynamic characteristic parameters of the structure, a static load test is performed on the key connection node to measure the deformation recovery performance; The morphological correction data and the whole-process technical parameters are integrated to establish a digital twin model, and a deformation control file is formed.

6. A multi-stage adjustment system for steel structure welding distortion control for implementing the multi-stage adjustment method for steel structure welding distortion control according to any one of claims 1 to 5, characterized in that, The multi-stage adjustment system for steel structure welding deformation control comprises: A measurement module is configured to establish a welding deformation prediction model by measuring the geometric parameters of the steel structure and combining the welding process parameters, and obtain a pre-deformation data set; A processing module is configured to perform precise pre-deformation processing on the specified position of the steel structure according to the pre-deformation data set, and obtain a pre-processed steel member; A recording module is configured to perform segmented welding operation on the pre-processed steel member, record the temperature field distribution and real-time deformation amount during the welding process, and obtain a welding deformation record table; An implementation module is configured to perform local heat treatment on the high stress area of the welded steel member according to the welding deformation record table, and obtain a stress distribution map; An application module is configured to apply a correction force to the deformation area by a multi-point force calibration device according to the stress distribution map, and obtain morphological correction data; An evaluation module is configured to use the morphological correction data to perform final precision detection and quality evaluation on the steel structure, and form a deformation control file.

7. A computer device, characterized by A memory and a processor are included, and the memory stores a computer program that can run on the processor, characterized in that the processor executes the computer program to realize the multi-stage adjustment method for steel structure welding deformation control according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and the computer program makes the processor execute the multi-stage adjustment method for steel structure welding deformation control according to any one of claims 1 to 5 when the processor runs.

Images