A method for active combined control of welding distortion and stress

Abstract

The present application relates to the technical field of welding process, and particularly relates to a kind of welding deformation and stress active joint control method.The method comprises the following steps: S1, establishing finite element simulation model obtains predicted deformation δ p , first correction coefficient K1 related to workshop environment temperature and second correction coefficient K2 related to assembly gap are introduced, and inverse deformation δ c Is calculated by combining uncertainty margin coefficient M;S2, actual deformation is obtained by local pre-welding test, deviation rate of actual deformation and predicted deformation is calculated, and if it is over deviation rate threshold, then K1, K2 are closed loop adjusted and inverse deformation is corrected;S3, different hammering parameters are used for root, filling layer and cover layer during welding, and stress is eliminated by welding hammering;S4, surface state after inverse deformation locking is used to initialize piezoelectric compensation zero point, and compensation output is reduced by using attenuation coefficient in welding initial stage.The present application improves inverse deformation precision and stress control level, and improves post-welding precision and quality stability of large structure.

Description

Technical Field

[0001] This invention relates to the field of welding process technology, and in particular to a method for the active joint control of welding deformation and stress. Background Technology

[0002] In shipbuilding, heavy machinery, pressure vessels, and bridge steel structures, large, thick-plate welded structural components are core load-bearing parts. During the welding process, localized high-temperature heating and rapid cooling cause significant compressive plastic strain in the weld and heat-affected zone, leading to macroscopic welding deformation and residual stress. Welding deformation causes the structural components to deviate from design values, requiring larger machining allowances and significantly increasing manufacturing costs and time. Residual stress, on the other hand, reduces the fatigue life, stress corrosion resistance, and structural stability of the structural components.

[0003] To control welding deformation, existing technologies widely employ the reverse deformation method, which involves applying a pre-deformation to the workpiece in the opposite direction to the welding deformation before welding to counteract the deformation generated during welding. The determination of the traditional reverse deformation amount mainly relies on the following two approaches: 1. Experience-based prediction method: The operator manually sets the amount of reverse deformation based on historical welding experience with similar structures. This method is highly subjective, has low accuracy, and is difficult to adapt to fluctuations in material properties and changes in working conditions between different batches, resulting in a low dimensional pass rate after welding. 2. Finite Element Simulation Method: This method uses finite element software to numerically simulate the welding process, predict the welding deformation, and then set the inverse deformation amount. This method is more scientific than empirical prediction, but existing simulation models are usually built under ideal boundary conditions and fail to consider the influence of dynamic factors such as changes in workshop ambient temperature and fluctuations in workpiece assembly gaps on the deformation. There is often a 10% to 20% deviation between the simulated predicted deformation and the actual deformation, leading to unstable inverse deformation effects.

[0004] Furthermore, in controlling welding residual stress, in-process hammering is a commonly used method to eliminate residual stress. Existing hammering processes often employ constant hammering force, frequency, and coverage, neglecting the significant differences in stress state, restraint, and functional requirements between the weld root, filler layer, and capping layer. The root weld has high restraint, and excessive hammering can easily induce microcracks; the filler layer has high residual stress peaks, and insufficient hammering force will result in inadequate residual stress elimination; the capping layer affects surface formation and appearance quality, and hammering parameters must balance residual stress elimination and morphology control. This constant-parameter hammering method is difficult to achieve optimal residual stress control.

[0005] In recent years, piezoelectric active compensation technology has been attempted for real-time cancellation of welding deformation. Piezoelectric stack actuators offer fast response and high precision, enabling them to output reverse displacement during welding to compensate for transient deformation. However, in existing technologies, there is a lack of effective coordination between the pre-weld applied reverse deformation (which remains constant during welding) and the dynamic piezoelectric compensation during welding. If the piezoelectric compensation system uses the original state of the workpiece before reverse deformation is applied as the reference zero point for displacement compensation, the reverse deformation displacement already generated on the workpiece after the application of reverse deformation will be identified by the system as the initial deviation to be compensated. This causes the piezoelectric stack actuator to erroneously output the displacement to cancel the reverse deformation, thus weakening the effect of the reverse deformation. Furthermore, if the compensation gain is improperly set, overshoot or oscillation is prone to occur in the early stages of welding, further affecting the final accuracy.

[0006] In summary, existing welding deformation and residual stress control technologies have the following shortcomings: (1) The accuracy of the anti-deformation setting is low, and it cannot adapt to the dynamic changes of the workshop environment and assembly status; (2) The welding hammer parameters are singular, and the control of residual stress lacks specificity; (3) There is a lack of effective coordination between pre-welding reverse deformation and in-welding dynamic piezoelectric compensation, making it difficult to give full play to their respective advantages, and they may even interfere with each other.

[0007] Therefore, there is an urgent need to develop a welding shape and property control method that can accurately set the amount of reverse deformation, differentiate the control of welding residual stress, and coordinate the static reverse deformation and dynamic compensation. Summary of the Invention

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a method for active joint control of welding deformation and stress, comprising the following steps: S1. Determination of pre-welding deformation: Establish a finite element simulation model of the workpiece to be welded, and obtain the predicted deformation δ of the workpiece through finite element simulation. p Collect the ambient temperature value T in the workshop, and determine the first correction factor K1 based on T; detect the assembly gap and calculate the average assembly gap G, and determine the second correction factor K2 based on G; based on δ p The inverse deformation δ is calculated using K1, K2, and the preset uncertainty margin coefficient M. c According to δ c Apply reverse deformation to the workpiece and lock it in place; S2. Pre-welding verification: Before formal welding, a local pre-welding test is conducted on a non-critical area of ​​the workpiece to detect the actual deformation amount δ. a Calculate its relationship with δ p The deviation rate η; if the deviation rate η exceeds the preset deviation rate threshold, then adjust K1 and / or K2 and recalculate and adjust the reverse deformation amount δ. cuntil the deviation rate η ≤ the deviation rate threshold; S3, graded hammering: During the welding process, the weld is welded in multiple layers and passes. Depending on the layer of the weld, different preset graded hammering parameters are used to hammer the weld. S4, Piezoelectric Compensation Synergy: Based on the workpiece surface state after applying reverse deformation and locking in S1, initialize the zero point of the piezoelectric active compensation system; during the welding process, apply real-time reverse displacement compensation to the workpiece surface through the piezoelectric stacked actuator in the piezoelectric active compensation system, and reduce the compensation output with an attenuation coefficient at the beginning of the welding stage.

[0009] Preferably, in step S1, the reverse deformation amount δ c The calculation formula is: δ c = δ p × M × K1 × K2, Where, δ p To predict the deformation; M is the uncertainty margin coefficient, whose value is determined according to the weld length L; K1 is the first correction coefficient; K2 is the second correction coefficient.

[0010] Preferably, the rule for determining the value of the first correction coefficient K1 is as follows: When T < 10℃, K1 = 1.05; When 10℃≤T≤20℃, K1=1; When 20℃<T≤30℃, K1=0.95; When T > 30℃, K1 = 0.9.

[0011] Preferably, the rule for determining the value of the second correction coefficient K2 is as follows: When 0 ≤ G ≤ 0.2 mm, K2 = 1.00; When 0.2mm < G ≤ 0.5mm, K2 = 1.03; If G > 0.5mm, the assembly is deemed unqualified and needs to be readjusted.

[0012] Preferably, the rule for determining the value of the uncertainty margin coefficient M is as follows: When L≤500mm, M=1.1; When 500mm < L ≤ 1000mm, M = 1.2; When L > 1000 mm, M = 1.3.

[0013] Preferably, in step S3, the graded hammering parameters are set as follows: Root weld: Hammering force 40-60N, frequency 35-40Hz, coverage width is the root weld width + 15mm on each side; Filler layer weld: hammering force 60-80N, frequency 40-50Hz, coverage width is the width of the filler layer weld plus 10mm on each side; Cover layer weld: hammering force 50-70N, frequency 45-50Hz, coverage width is the width of the cover layer weld plus 8mm on each side; The two sides refer to the two sides along the width of the weld, with the center line of the weld as the reference.

[0014] Preferably, in step S2, the deviation rate threshold ranges from 3% to 8%; when the deviation rate η exceeds the deviation rate threshold, the method for adjusting K1 and / or K2 is as follows: If the actual deformation δ a >Predicted deformation δ p If so, the second correction factor K2 should be increased by 0.02 to 0.05. If the actual deformation δ a <Predicted deformation δ p If so, the second correction factor K2 should be reduced by 0.02 to 0.05. If the deviation rate η is still greater than the deviation rate threshold after adjusting K2, the first correction coefficient K1 will be increased or decreased by 0.01 to 0.03 based on the current value.

[0015] Preferably, in step S4, the attenuation coefficient ranges from 0.3 to 0.5, and the welding start stage is the first two weld segments.

[0016] Preferably, in step S1, the anti-deformation is applied by an array of jacks arranged below the workpiece, the array of jacks being a hydraulic jack array or a mechanical jack array.

[0017] Preferably, the workpiece is the body of a hydraulic press, and the surface of the workpiece is the upper crossbeam table.

[0018] Compared with existing technologies, the present invention has the following significant advantages: 1. Precise and reliable setting of anti-deformation amount: By introducing first and second correction coefficients related to workshop ambient temperature and assembly gap respectively, and constructing a closed loop through pre-welding verification steps, the determination of anti-deformation amount is upgraded to a precise engineering control method of "model-driven + working condition adaptive + closed-loop calibration". Practical application shows that the flatness of key surfaces after welding can be stably controlled within 0.05mm, improving the product qualification rate; 2. Differentiated and efficient control of residual stress: Graded hammering parameters are developed to address the different residual stress states and process requirements of the root, filler, and cap coat welds. The peak residual stress is reduced by 25%–40% compared to traditional constant parameter hammering. 3. Enhanced Synergy Between Anti-Deformation and Piezoelectric Compensation: By calibrating the workpiece state after anti-deformation and locking as the zero point of the piezoelectric active compensation system, and actively attenuating the compensation gain (attenuation coefficient) in the early stages of welding, conflicts and overshoot between the two shape control mechanisms are effectively avoided. This reduces subsequent machining allowances and significantly lowers manufacturing costs. 4. Strong adaptability and wide range of applications: The method of this invention does not depend on specific welding equipment. The rules for determining the values ​​of the first and second correction coefficients are clear, and the operation process is highly standardized. It is applicable to large structural components such as hydraulic press bodies, as well as various thick plate welding structures such as ship sections and bridge steel box girders. Attached Figure Description

[0019] Figure 1 This is a process flow diagram of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1: This example uses the welding of the frame of an 80mm thick Q355 steel frame hydraulic press as an example. The frame is welded from an upper crossbeam, a lower crossbeam, and two side beams, with a total of 8 welds. Taking the weld connecting the upper crossbeam and the left side beam as an example (weld length 900mm), the specific implementation process of the method of this invention is explained. The active joint control method for welding deformation and stress is shown in the accompanying drawings of the specification. Figure 1 The process includes the following steps: S1, determination of pre-welding deformation amount, S2, pre-welding verification, S3, graded hammering, and S4, piezoelectric compensation coordination.

[0022] S1. Determination of pre-welding deformation amount: A finite element simulation model of the workpiece to be welded is established, and the predicted deformation δ of the workpiece is obtained through finite element simulation. p Collect the ambient temperature value T in the workshop, and determine the first correction factor K1 based on T; detect the assembly gap and calculate the average assembly gap G, and determine the second correction factor K2 based on G; based on δ p The inverse deformation δ is calculated using K1, K2, and the preset uncertainty margin coefficient M. c According to δ cThe workpiece is subjected to reverse deformation and locked. This reverse deformation is applied via a jack array positioned below the workpiece; the jack array can be a hydraulic or mechanical array. The workpiece is the body of a hydraulic press, and its surface is the upper crossbeam platform. The specific operating steps are as follows: 1. Establish a finite element simulation model In this embodiment, the finite element simulation model is an integral welded model of the upper crossbeam, side beam, and lower crossbeam, established using finite element software (such as Sysweld). The welding method is gas metal arc welding (GMAW). A fine mesh (mesh size no larger than 2mm) is used for the weld and heat-affected zone, while a coarser mesh (15-20mm) is used for the far ends. The input material parameters are as follows: Q355 steel: thermal conductivity 18 W / (m·℃), coefficient of linear expansion 11.5 × 10⁻⁶ -6 / ℃, elastic modulus 206GPa, yield strength 355MPa; Welding process parameters: current 320A, voltage 32V, welding speed 5mm / s; Nominal assembly clearance: 0.3mm; Constraint settings: Only the Z-axis translational degree of freedom is constrained at the jack array support points (a total of 6 points, evenly arranged below the upper crossbeam at a spacing of 250mm) in the finite element simulation model, while the X, Y and rotational degrees of freedom are released to accurately simulate the free support state of the workpiece before the application of reverse deformation. The simulation calculation employs a thermo-elastoplastic coupling analysis method: first, the transient temperature field is calculated, and then the calculation result of the transient temperature field is used as a load to map to the stress field for solution. Simulation results show that after welding, an upward arching deformation occurs in the central region of the upper crossbeam platform, with the maximum deformation located at the geometric center of the upper crossbeam platform. The predicted deformation δ... p = 0.040mm.

[0023] 2. Determine the correction factor The anti-deformation amount δ c The calculation formula is: δ c = δ p × M × K1 × K2, Where, δ p To predict the deformation; M is the uncertainty margin coefficient, whose value is determined according to the weld length L; K1 is the first correction coefficient; K2 is the second correction coefficient.

[0024] The rule for determining the value of the first correction coefficient K1 is as follows: When T < 10℃, K1 = 1.05; When 10℃≤T≤20℃, K1=1; When 20℃<T≤30℃, K1=0.95; When T > 30℃, K1 = 0.9.

[0025] The rule for determining the value of the second correction coefficient K2 is as follows: When 0 ≤ G ≤ 0.2 mm, K2 = 1.00; When 0.2mm < G ≤ 0.5mm, K2 = 1.03; If G > 0.5mm, the assembly is deemed unqualified and needs to be readjusted.

[0026] The rule for determining the value of the uncertainty margin coefficient M is as follows: When L≤500mm, M=1.1; When 500mm < L ≤ 1000mm, M = 1.2; When L > 1000 mm, M = 1.3.

[0027] In this embodiment, the ambient temperature of the workshop was measured to be T=25℃ using a temperature and humidity recorder. Based on the rules for determining the value of K1, the first correction factor K1=0.95.

[0028] After the workpiece assembly is completed, using a set of feeler gauges with an accuracy of 0.01mm, five evenly spaced inspection points (approximately 150mm apart) are selected on the butt weld between the upper crossbeam and the left side beam. The measured assembly gaps are 0.28mm, 0.32mm, 0.29mm, 0.31mm, and 0.30mm, respectively, and the average assembly gap G is calculated to be 0.30mm. According to the rules for selecting the value of K2, the second correction coefficient K2 is determined to be 1.03.

[0029] In this embodiment, the weld length L = 900 mm, and according to the rules for taking the value of M, the uncertainty margin coefficient M = 1.2 is taken.

[0030] Substitute the inverse deformation amount δ c The calculation formula is obtained as follows: δ c = δ p × M × K1 × K2 = 0.040 × 1.2 × 0.95 × 1.03 ≈ 0.047mm.

[0031] 3. Apply reverse deformation In this embodiment, the jack array consists of six hydraulic jacks arranged below the upper crossbeam platform, spaced 250mm apart, with a lifting accuracy of ±0.005mm. The hydraulic jack control system synchronously lifts the upper crossbeam, causing a 0.047mm upward arching deformation at the center of the platform relative to its edge. A dial indicator with a resolution of 0.001mm is placed at five feature points on the platform (the four corner points and the center point) for displacement monitoring during the pre-welding reverse deformation stage. The vertical Z-axis displacement of each feature point is monitored in real time until the deformation at each feature point stabilizes within the δ... c Within the range of (0.047mm) ±0.002mm, the hydraulic locks of each hydraulic jack are then locked to maintain the anti-deformation state. ±0.002mm refers to the allowable tolerance for the applied anti-deformation precision, i.e., the actual anti-deformation amount and the calculated δ. c The deviation (0.047mm) shall not exceed 0.002mm. This accuracy tolerance is determined comprehensively based on the lifting accuracy of the hydraulic jack and the post-weld flatness of the upper crossbeam platform as required by the design (e.g., 0.05mm), aiming to ensure that the anti-deformation amount is applied accurately and reliably, providing a stable reference for subsequent piezoelectric compensation.

[0032] S2, Pre-welding verification: Before formal welding, a local pre-welding test was conducted on a non-critical area of ​​the workpiece to detect the actual deformation δ. a Calculate its relationship with δ p The deviation rate η; if the deviation rate η exceeds the preset deviation rate threshold, then adjust K1 and / or K2 and recalculate and adjust the reverse deformation amount δ. c The process continues until the deviation rate η is less than or equal to the deviation rate threshold; the deviation rate threshold ranges from 3% to 8%; when the deviation rate η exceeds the deviation rate threshold, the method for adjusting K1 and / or K2 is as follows: If the actual deformation δ a >Predicted deformation δ p If so, the second correction factor K2 should be increased by 0.02 to 0.05. If the actual deformation δ a <Predicted deformation δ p If so, the second correction factor K2 should be reduced by 0.02 to 0.05. If the deviation rate η is still greater than the deviation rate threshold after adjusting K2, then the first correction coefficient K1 will be increased or decreased by 0.01 to 0.03 based on its current value. Specifically, if δ a >δ p Decrease K1 by 0.01 to 0.03 from its current value; if δ a <δ p Increase K1 by 0.01 to 0.03 from its current value.

[0033] In this embodiment, the deviation rate threshold is set at 5%. The non-load-bearing area at the end of the weld connecting the upper crossbeam and the left side beam is selected as a non-critical area. A welding length of 100mm is used as a local pre-welding test. The welding parameters are completely consistent with those of the formal welding. After welding, the area is allowed to cool naturally to room temperature. A laser tracker is used to scan the upper crossbeam platform corresponding to the local pre-welding test area to measure the actual deformation δ. a = 0.042mm.

[0034] Calculate the deviation rate The deviation rate η equals the deviation rate threshold of 5%, which meets the requirements. Therefore, there is no need to adjust the correction coefficient, and the actual welding adopts a reverse deformation amount of 0.047mm.

[0035] In another embodiment, δ was measured in a local pre-welding test. a = 0.045mm, then the deviation rate This exceeds the deviation rate threshold of 5%. Due to the actual deformation δ a >Predicted deformation δ p The amount of reverse deformation needs to be increased. Following the above method, K2 should be increased first, from 1.03 to 1.05 (an increase of 0.02). δ should then be recalculated. c : δ c = δ p × M × K1 × K2 = 0.040 × 1.2 × 0.95 × 1.05 ≈ 0.048mm.

[0036] Adjust the lifting capacity of each hydraulic jack to 0.048mm, and then conduct a local pre-welding test. If the second measured δ a = 0.041mm, deviation rate If the deviation rate is less than the 5% threshold, the requirement is met, and the current K2=1.05 and δ are used. c =0.048mm for formal welding.

[0037] If the second δ a = 0.046mm, deviation rate If the deviation rate still exceeds the 5% threshold, K2 has already increased to 1.05 and can be further increased to 1.06–1.08. If the deviation rate is still greater than 5% after adjusting K2 to 1.08, then K1 should be fine-tuned: because δ a >δ p Reduce K1 by 0.01 to 0.03 and recalculate δ. c And verify until the deviation rate is ≤ 5% of the deviation rate threshold.

[0038] S3, graded hammering: During the welding process, the weld seam is welded in multiple layers and passes. Depending on the layer of the weld pass, different preset graded hammering parameters are used for on-the-fly hammering. The graded hammering parameters are set as follows: Root weld: Hammering force 40-60N, frequency 35-40Hz, coverage width is the root weld width + 15mm on each side; Filler layer weld: hammering force 60-80N, frequency 40-50Hz, coverage width is the width of the filler layer weld plus 10mm on each side; Cover layer weld: hammering force 50-70N, frequency 45-50Hz, coverage width is the width of the cover layer weld plus 8mm on each side; The two sides refer to the two sides along the width of the weld, with the center line of the weld as the reference.

[0039] In this embodiment, the weld seam employs a multi-layer, multi-pass welding process. A "layer" refers to a stack of weld layers along the weld thickness direction, and a "pass" refers to a single weld pass arranged along the width direction within the same weld layer. For an 80mm plate thickness and the GMAW process, the weld seam is welded in four layers: the first layer (root) consists of one pass at the root of the bevel; the second layer (filler layer) consists of three passes; the third layer (filler layer) consists of three passes; and the fourth layer (cap coat) consists of two passes, forming the final weld surface. A total of nine passes are involved. This pass arrangement aims to control the heat input and pass width of each pass, providing a clear basis for subsequent graded hammering. The interpass temperature is controlled between 150 and 200°C.

[0040] After each layer of welding is completed, once the weld surface temperature drops to no less than 300℃, welding hammering is immediately performed. A pneumatic hammer gun with a 10mm diameter hammer head made of high-speed steel is used. Using a test plate of the same material (Q355 steel) and thickness (80mm) as the workpiece, single-pass welding tests were conducted on three typical weld passes: root pass, filler pass, and cap pass, using the same parameters as the final welding. After cooling to room temperature, five cross-sections were evenly selected on each weld pass, and measurements were taken using calipers. The arithmetic mean was used as the average width of each single weld pass, which was measured as follows: root pass 8mm, filler pass 10mm, and cap pass 12mm.

[0041] The specific settings for the graded hammering parameters are as follows: Root weld (first layer): Hammering force 50N, frequency 38Hz, coverage width = root weld width + 15mm on each side = 8mm + 30mm = 38mm; Filler layer weld: hammering force 70N, frequency 45Hz, coverage width = filler layer weld width + 10mm on each side = 10mm + 20mm = 30mm; Cover layer weld: hammering force 60N, frequency 48Hz, coverage width = cover layer weld width + 8mm on each side = 12mm + 16mm = 28mm.

[0042] Hammering operation requirements: The pneumatic hammer gun should move at a constant speed along the centerline of the weld, ensuring that the hammer marks evenly cover the set coverage width and overlap with the previous hammered area by 10% to 15%. After each weld pass is hammered, the surface should be visually inspected to show uniform plastic deformation marks, without cracks, peeling, or other defects. Lower hammering force and higher frequency are used for the root pass to effectively suppress the initiation of microcracks; higher hammering force is used for the filler pass to significantly eliminate interlayer residual stress; the cap pass optimizes the frequency and coverage width to balance residual stress elimination and surface quality.

[0043] S4, piezoelectric compensation coordination: Using the workpiece surface state after reverse deformation in S1 as a reference, the zero point of the piezoelectric active compensation system is initialized. During welding, real-time reverse displacement compensation is applied to the workpiece surface through the piezoelectric stack actuator of the piezoelectric active compensation system, and the compensation output is reduced by an attenuation coefficient at the beginning of welding. All components of the piezoelectric active compensation system are conventional equipment, including a piezoelectric stack actuator, a laser displacement sensor, a laser tracker, and a piezoelectric compensation controller. The piezoelectric stack actuator is the actuating element. The attenuation coefficient ranges from 0.3 to 0.5, and the initial welding stage refers to the first two weld segments. Specific operation is as follows: After applying anti-deformation locking to the workpiece, a laser tracker is used to perform a grid scan on the upper crossbeam platform. The measurement point spacing is 50×50mm, and a total of 15×15=225 measurement points are collected. The initial Z-axis coordinate value Z0(x,y) of each measurement point is recorded (where (x,y) is the coordinate of the measurement point in the platform plane), and the Z0(x,y) data of all measurement points is transmitted to the piezoelectric active compensation system as the compensation zero point of the piezoelectric stacked actuator at the corresponding position of each measurement point. The "corresponding position" refers to the fact that each piezoelectric stacked actuator is responsible for compensating the deformation of the measurement points within a radius of 100mm centered on its installation position.

[0044] Piezoelectric stacked actuators (maximum displacement 40μm, output force 5kN, drive voltage 0~150V) are arranged along the length of the weld seam at 200mm intervals. In this embodiment, the weld seam length L=900mm, and a total of 5 piezoelectric stacked actuators are arranged. A spare piezoelectric stacked actuator of the same model is installed in parallel next to each piezoelectric stacked actuator. The top of the piezoelectric stacked actuator contacts the workpiece through a 6mm thick heat insulation pad, and the outer shell of the piezoelectric stacked actuator is equipped with a heat-insulating protective sleeve (temperature resistance ≥300℃).

[0045] During the welding process, a laser displacement sensor (sampling frequency 20Hz, resolution 0.1μm) is used to monitor the Z-axis displacement change ΔZ(t) at each measuring point in real time. The piezoelectric stacked actuator applies real-time reverse displacement compensation to the workpiece surface based on ΔZ(t), and its target output displacement D(t) is determined by the following formula: D(t) = β × ΔZ(t), Where β is the attenuation coefficient. In the initial welding stage (the first two weld segments), β is taken as 0.4; from the third weld segment onwards, β recovers to 1.0, i.e., D(t) = ΔZ(t). This β value was determined through joint calibration of finite element simulation and field tests, and its range is 0.3 to 0.5, effectively avoiding compensation overshoot caused by unstable heat input in the early stages of welding. The piezoelectric active compensation system has a response time ≤5ms, ensuring real-time compensation.

[0046] During welding, the output force fluctuation of the piezoelectric stacked actuator is monitored in real time. When the equivalent displacement change caused by the output force fluctuation exceeds ±0.005mm, the piezoelectric active compensation system automatically switches to the backup piezoelectric stacked actuator and issues an alarm. "±0.005mm equivalent displacement" refers to the equivalent displacement change of the piezoelectric stacked actuator after the output force fluctuation is converted through its stiffness. Because the piezoelectric stacked actuator is a force-displacement coupled device, the output force fluctuation directly manifests as displacement fluctuation; therefore, the equivalent displacement is used as the unified monitoring threshold.

[0047] S4 also includes post-weld treatment and effect verification: After welding all eight seams and allowing them to cool naturally to room temperature, all hydraulic jacks were released. A laser tracker was used to comprehensively inspect the flatness of the upper crossbeam platform, and the maximum flatness error was measured to be 0.038 mm, which is better than the design requirement of 0.05 mm.

[0048] Residual stress in the weld center and heat-affected zone was detected using a portable X-ray stress analyzer. Among 24 measurement points (3 points per weld) across 8 welds, the maximum residual stress was 156 MPa. Subsequent optimization of hammering parameters and interpass temperature control further reduced the peak residual stress to 102 MPa, significantly lower than 50% of the yield strength of Q355 steel (177.5 MPa).

[0049] The blind hole method was used to supplement the detection of deep residual stress in the weld joint area (arc start, arc end, and segmented lap joint). The detection depth was 1.5 mm, the maximum residual stress was 131 MPa, and there were no points exceeding the standard.

[0050] All process parameters, test data, operators, equipment numbers, and other information for this welding process are entered into the individual piece quality file (by scanning the barcode, the workpiece number is bound to the welding parameters, test data, and operators, stored in the database, and the software generates control charts), providing data support for subsequent process iterations.

[0051] Brief description of working principle The working principle of this invention is: (1) The predicted deformation δ is obtained through finite element simulation. p Then, a first correction factor K1 related to the workshop ambient temperature and a second correction factor K2 related to the assembly clearance are introduced to predict the deformation δ. p Dynamic correction is performed to obtain the inverse deformation amount δ. c This eliminates deviations caused by fluctuations in operating conditions; further, closed-loop calibration is formed through pre-welding verification, enabling the setting accuracy of the anti-deformation amount to reach the micrometer level. (2) Based on the residual stress sensitivity and functional requirements of different weld depths, graded hammering is implemented: light hammering at the root to prevent cracking, heavy hammering at the filler layer to relieve stress, and optimized frequency and shape preservation at the cover layer, so as to achieve precise and efficient control of residual stress. (3) The workpiece geometry after applying anti-deformation and locking is used as the zero point of the piezoelectric active compensation system for calibration. In the early stage of welding, the compensation output is reduced by the attenuation coefficient, so that the two different control mechanisms of anti-deformation and piezoelectric compensation can smoothly transition and complement each other, avoiding control conflicts.

[0052] The three elements work together to form a proactive joint control over welding deformation and residual stress throughout the entire process, from pre-welding, during welding to post-welding, significantly improving the dimensional accuracy and internal quality of large welded structural components.

[0053] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for active joint control of welding deformation and stress, characterized in that, Includes the following steps: S1. Determination of pre-welding deformation: Establish a finite element simulation model of the workpiece to be welded, and obtain the predicted deformation δ of the workpiece through finite element simulation. p Collect the ambient temperature value T in the workshop, and determine the first correction factor K1 based on T; Detect assembly clearance and calculate average assembly clearance G, then determine the second correction factor K2 based on G; based on δ p The inverse deformation δ is calculated using K1, K2, and the preset uncertainty margin coefficient M. c According to δ c Apply reverse deformation to the workpiece and lock it in place; S2. Pre-welding verification: Before formal welding, a local pre-welding test is conducted on a non-critical area of ​​the workpiece to detect the actual deformation amount δ. a Calculate its relationship with δ p The deviation rate η; if the deviation rate η exceeds the preset deviation rate threshold, then adjust K1 and / or K2 and recalculate and adjust the reverse deformation amount δ. c until the deviation rate η ≤ the deviation rate threshold; S3, graded hammering: During the welding process, the weld is welded in multiple layers and passes. According to the layer of the weld, different preset graded hammering parameters are used to hammer the weld. S4, Piezoelectric Compensation Synergy: Using the workpiece surface state after applying reverse deformation and locking in S1 as a reference, initialize the zero point of the piezoelectric active compensation system. During the welding process, the piezoelectric stacked actuator in the piezoelectric active compensation system applies real-time reverse displacement compensation to the workpiece surface, and the compensation output is reduced by attenuation coefficient at the beginning of the welding stage.

2. The active joint control method for welding deformation and stress according to claim 1, characterized in that, In step S1, the reverse deformation amount δ c The calculation formula is: d c = d p × M × K1 × K2, Where, δ p M is the uncertainty margin coefficient, whose value is determined based on the weld length L; K1 is the first correction coefficient; K2 is the second correction coefficient.

3. The active joint control method for welding deformation and stress according to claim 2, characterized in that, The rule for determining the value of the first correction coefficient K1 is as follows: When T < 10℃, K1 = 1.05; When 10℃≤T≤20℃, K1=1; When 20℃<T≤30℃, K1=0.95; When T > 30℃, K1 = 0.

9.

4. The active joint control method for welding deformation and stress according to claim 2, characterized in that, The rule for determining the value of the second correction coefficient K2 is as follows: When 0 ≤ G ≤ 0.2 mm, K2 = 1.00; When 0.2mm < G ≤ 0.5mm, K2 = 1.03; If G > 0.5mm, the assembly is deemed unqualified and needs to be readjusted.

5. The active joint control method for welding deformation and stress according to claim 2, characterized in that, The rule for determining the value of the uncertainty margin coefficient M is as follows: When L≤500mm, M=1.1; When 500mm < L ≤ 1000mm, M = 1.2; When L > 1000 mm, M = 1.

3.

6. The active joint control method for welding deformation and stress according to claim 1, characterized in that, In step S3, the parameters for the graded hammering are set as follows: Root weld: Hammering force 40-60N, frequency 35-40Hz, coverage width is the root weld width + 15mm on each side; Filler layer weld: hammering force 60-80N, frequency 40-50Hz, coverage width is the width of the filler layer weld plus 10mm on each side; Cover layer weld: hammering force 50-70N, frequency 45-50Hz, coverage width is the width of the cover layer weld plus 8mm on each side; The two sides refer to the two sides along the width of the weld, with the center line of the weld as the reference.

7. The active joint control method for welding deformation and stress according to claim 1, characterized in that, In step S2, the deviation rate threshold ranges from 3% to 8%; when the deviation rate η exceeds the deviation rate threshold, the method for adjusting K1 and / or K2 is as follows: If the actual deformation δ a >Predicted deformation δ p If so, the second correction factor K2 should be increased by 0.02 to 0.

05. If the actual deformation δ a <Predicted deformation δ p If so, the second correction factor K2 should be reduced by 0.02 to 0.

05. If the deviation rate η is still greater than the deviation rate threshold after adjusting K2, the first correction coefficient K1 will be increased or decreased by 0.01 to 0.03 based on the current value.

8. The active joint control method for welding deformation and stress according to claim 1, characterized in that, In step S4, the attenuation coefficient ranges from 0.3 to 0.5, and the welding start stage refers to the first two weld segments.

9. The active joint control method for welding deformation and stress according to claim 1, characterized in that, In step S1, the anti-deformation is applied by an array of jacks arranged below the workpiece, the array of jacks being either a hydraulic jack array or a mechanical jack array.

10. The active joint control method for welding deformation and stress according to claim 1, characterized in that, The workpiece is the body of a hydraulic press, and the surface of the workpiece is the upper crossbeam table.

Images