Welding method suitable for tube head and tube plate of heat exchanger

Abstract

The invention discloses a welding method suitable for a tube head and a tube plate of a heat exchanger, which relates to the technical field of welding control, and comprises the following steps of: simulating a welding thermodynamic diagram of a next circle by collecting physical data of the tube head and the tube plate and currently set welding process data as a first welding thermodynamic diagram; determining a high-risk welding area in the welding thermodynamic diagram according to the first welding thermodynamic diagram; performing optimization setting on the welding process data of the high-risk welding area again, and randomly generating a plurality of groups of welding process data according to constraint conditions of the welding process; combining each group of welding process data with physical data simulation of the tube head and the tube plate to obtain a new welding thermodynamic diagram of a next circle, recording the new welding thermodynamic diagram as a second welding thermodynamic diagram, and determining target welding process data according to the first welding thermodynamic diagram and the second welding thermodynamic diagram; the tube head and the tube plate of the next circle are welded according to the target welding process data; after each circle of tube head and the tube plate are welded, optimization is repeated, welding of the remaining circles is completed, the reliability of the heat exchanger is improved, and the service life of the heat exchanger is prolonged.

Description

Technical Field

[0001] The invention relates to the technical field of welding control, and in particular to a welding method suitable for a tube head and a tube plate of a heat exchanger. Background Art

[0002] The welding of heat exchanger tube heads and tube sheets is a critical step in the heat exchanger manufacturing process. A multi-row, porous structure is typically used for this purpose. This structure ensures the efficient arrangement of the heat exchange pipes and achieves good heat exchange. The welding process typically utilizes techniques such as TIG (tungsten inert gas welding) and MIG (metal inert gas welding), which ensure weld strength and sealing. However, the concentrated and uneven distribution of heat input during welding can easily lead to uneven structural stress distribution between the tube heads and tube sheets, which in turn affects weld quality and the service life of the heat exchanger.

[0003] For example, during the welding of multi-row porous structures, uneven heat input can cause deformation of the tube sheet. This is especially true when welding large-size tube sheets, where concentrated heat input becomes more pronounced and the deformation problem becomes more severe. Furthermore, uneven stress distribution in the weld can also lead to fatigue cracking, intergranular corrosion, or stress corrosion. These problems, especially under high temperature and high pressure conditions, can accelerate the generation of cracks, thereby affecting the reliability and service life of the heat exchanger. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide a welding method suitable for the tube head and tube sheet of a heat exchanger.

[0005] The present invention provides a method for welding a tube head and a tube sheet of a heat exchanger, the method comprising:

[0006] S1: collecting physical data of the tube head and tube sheet and currently set welding process data, and obtaining the welding thermodynamic map of the next cycle through a preset heat-stress simulation model as a first welding thermodynamic map, and determining high-risk welding areas in the welding thermodynamic map based on the first welding thermodynamic map;

[0007] S2: re-optimize the welding process data of the high-risk welding zone and randomly generate several sets of welding process data according to the constraints of the welding process;

[0008] S3: Re-simulating each set of welding process data in combination with the physical data of the tube head and the tube sheet using a preset thermal-stress simulation prediction model to obtain a new welding thermogram for the next cycle corresponding to each set of welding process data, recorded as a second welding thermogram, and determining target welding process data based on the first welding thermogram and the second welding thermogram;

[0009] S4: welding the tube head and tube sheet of the next circle according to the target welding process data;

[0010] S5: After each circle of tube heads and tube sheets are welded, steps S1-S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger.

[0011] Optionally, S2: re-optimizing the welding process data of the high-risk welding zone, and randomly generating several groups of welding process data according to the constraints of the welding process is as follows:

[0012] The welding process data includes welding current, voltage, welding speed, shielding gas flow rate, wire feed speed, spot welding duration, cooling interval time and welding path sequence;

[0013] Several groups of welding process data combinations are randomly generated within the range of constraints through Latin hypercube sampling.

[0014] Optionally, the step of determining target welding process data according to the first welding thermogram and the second welding thermogram is:

[0015] The first welding thermogram and the second welding thermogram are overlapped and compared, and the area corresponding to the original high-risk area in the first welding thermogram is recorded as the target area, and the area corresponding to the target area in the second welding thermogram is marked as the optimized area;

[0016] Calculate the temperature standard deviation and stress standard deviation of the target area and the optimized area respectively, and calculate the reduction ratio TD and SD of the temperature standard deviation T2 and stress standard deviation S2 of the optimized area compared with the temperature standard deviation T1 and stress standard deviation S1 of the target area. Add the ratios as the temperature stress optimization value of the high-risk area. The specific formula is:

[0017] The optimization effect value is calculated according to the temperature stress optimization value of the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

[0018] Optionally, the steps of calculating the optimization effect value according to the temperature stress optimization value of the high-risk area, and taking the welding process data corresponding to the maximum optimization effect value as the target welding process data are:

[0019] Obtain temperature data of each grid area in the first welding thermogram and the second welding thermogram, where the thermogram includes temperature data of each point on the spatial grid;

[0020] Calculate the temperature gradient of each grid area in the first welding thermogram and the second welding thermogram For each grid point, in two dimensions, the temperature gradient is calculated by the difference between the horizontal and vertical directions, and the calculation formula is:

[0021]

[0022] Where, T i，j is the temperature value at position (i, j), Δx and Δy are the horizontal and vertical distances between grid points, respectively;

[0023] Calculate the temperature gradient of each grid point in the first welding thermogram and the second welding thermogram, respectively and Calculate the temperature gradient change: Where, is the temperature gradient change at position (i, j);

[0024] Calculate the temperature gradient change failure coefficient The calculation formula is: Where, and are the sum of the temperature gradient changes of all grid points in the second welding thermodynamic map and the sum of the temperature gradient changes of all grid points in the first welding thermodynamic map respectively;

[0025] The optimization effect value is calculated based on the temperature stress optimization value and the temperature gradient change unqualified coefficient in the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

[0026] Optionally, the step of calculating the optimization effect value based on the temperature stress optimization value and the temperature gradient change unqualified coefficient of the high-risk area is:

[0027] The temperature stress optimization value and the temperature gradient change unqualified coefficient are normalized and mapped to the numerical range of 0-1. The weight values ​​of the normalized temperature stress optimization value and the temperature gradient change unqualified coefficient are both set to 0.5. The optimization effect value of the welding process data is calculated. The calculation formula is: HJ=0.5*yu-0.5*rf, where HJ is the optimization effective coefficient, yu is the temperature stress optimization value after normalization, and rf is the temperature gradient change unqualified coefficient after normalization.

[0028] Optionally, S5: after each circle of tube heads and tube sheets are welded, steps S1 to S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger, further comprising:

[0029] After each round of welding of the tube head and the tube sheet is completed, the actual welding data of the tube head and the tube sheet after welding is obtained, and the actual welding data is compared with the standard welding data range. If the actual welding data is within the standard welding data range, steps S1-S4 are repeated to complete the welding of the remaining rounds of the heat exchanger tube head and the tube sheet;

[0030] If the actual welding data is not within the standard welding data range, an alarm will be issued immediately and the welding of the heat exchanger tube head and the remaining circles of the tube sheet will be stopped.

[0031] Beneficial effects of the present invention:

[0032] 1. The present invention proposes a welding method suitable for the tube head and tube sheet of a heat exchanger. During the welding process of multiple rows of porous structures, the most suitable welding process parameters for the tube head and tube sheet can be selected according to the simulation results of the simulation model, thereby reducing uneven heat input. At the same time, after each round of welding is completed, the simulation process parameters for the next round can be continuously selected, thereby reducing the probability of tube sheet deformation and increasing the reliability and service life of the heat exchanger.

[0033] 2. During the welding process of multi-row and multi-porous structures, it can determine whether to continue welding the subsequent heat exchanger tube heads and tube sheets based on the actual welding conditions, effectively avoiding potential quality problems caused by unqualified welding parameters, ensuring the accuracy and reliability of each welding cycle, and thus ensuring the long-term stable operation of the heat exchanger. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below with reference to the accompanying drawings.

[0035] Figure 1 The present invention is a flow chart of a welding method suitable for a heat exchanger tube head and a tube sheet. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] The embodiment of the present invention provides a method for welding a tube head and a tube sheet of a heat exchanger. Figure 1 , Figure 1 A flow chart of a method for welding a tube head and a tube sheet of a heat exchanger provided in an embodiment of the present invention. The method comprises the following steps:

[0038] S1: collecting physical data of the tube head and tube sheet and currently set welding process data, and obtaining the welding thermodynamic map of the next cycle through a preset heat-stress simulation model as a first welding thermodynamic map, and determining high-risk welding areas in the welding thermodynamic map based on the first welding thermodynamic map;

[0039] S2: re-optimize the welding process data of the high-risk welding zone and randomly generate several sets of welding process data according to the constraints of the welding process;

[0040] S3: Re-simulating each set of welding process data in combination with the physical data of the tube head and the tube sheet using a preset thermal-stress simulation prediction model to obtain a new welding thermogram for the next cycle corresponding to each set of welding process data, recorded as a second welding thermogram, and determining target welding process data based on the first welding thermogram and the second welding thermogram;

[0041] S4: welding the tube head and tube sheet of the next circle according to the target welding process data;

[0042] S5: After each circle of tube heads and tube sheets are welded, steps S1-S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger.

[0043] Based on an embodiment of the present invention, a welding method suitable for a heat exchanger tube head and a tube sheet is provided. Through the above-mentioned method, during the welding process of a multi-row porous structure, the most suitable welding process parameters for the tube head and the tube sheet can be selected according to the simulation results of the simulation model, thereby reducing uneven heat input. At the same time, after each round of welding is completed, the simulation process parameters for the next round can be continued to be selected, thereby reducing the probability of tube sheet deformation and increasing the reliability and service life of the heat exchanger.

[0044] In one embodiment, S1: physical data of the tube head and the tube sheet and currently set welding process data are collected, and a welding thermogram of the next cycle is obtained using a preset thermal-stress simulation model as a first welding thermogram, and high-risk welding areas in the welding thermogram are determined based on the welding thermogram;

[0045] It should be noted that "a circle" refers to a group of welded units arranged sequentially around the center of the tube holes arranged in a specific pattern (such as concentric circles, polygons, or matrices) along the tube sheet. This refers to a group of tube head and tube sheet welding areas located on the same annular level. Because heat exchangers typically use a multi-arrangement, multi-hole arrangement structure, to avoid local warping or hole position shifts caused by excessive heat input, the welding process often adopts a "circle by circle" strategy, welding one circle at a time from the inside out, in a clockwise or cross-jumping manner. Therefore, "a circle" refers to a relatively independent annular or linear tube hole welding unit to be executed at the current stage of the welding sequence.

[0046] It should be noted that the physical data of the tube head and tube sheet include: tube sheet size: thickness, diameter, material (such as 316L stainless steel), tube head outer diameter, wall thickness, material; tube sheet aperture, arrangement (such as concentric circle, square array), hole spacing; actual deformation of the welded area (such as collected by a laser measurement system); current welding process data include: welding method (such as TIG), current, voltage, welding speed, heat input (kJ / mm) welding sequence (which direction to be welded), cooling strategy (whether there is forced air cooling / water cooling), shielding gas flow rate, type, etc.; these data are usually automatically collected by the system integrated process database or sensor system, such as: using the welding machine controller to upload process parameters; using a laser displacement sensor to measure deformation in real time; using a temperature sensor (thermocouple array) to record temperature field data; extracting structural dimensions and arrangement from the CAD model, etc., the details will not be repeated. The above data is input into a preset thermal-stress coupling simulation model or a trained AI proxy model (such as a neural network proxy model) to perform the following functions: Model Function: Establish a three-dimensional finite element model (including the tube sheet, the welded section, and the tube head to be welded); simulate the movement of the welding heat source along the next path based on the current process parameters; obtain a heat input distribution map (temperature cloud map around the weld); further calculate the temperature field → thermal expansion → residual stress and deformation field; output stress field and deformation field images to form a welding thermodynamic map. The preset thermal-stress coupling simulation model is usually trained or fitted using a large amount of physical simulation data or experimental measurement data from real welding processes. Its essence is to build a computational model based on heat conduction theory, thermal expansion principles, and material constitutive relationships (such as elastic-plastic stress-strain relationships) that can predict the temperature and stress field distribution based on the input welding parameters and structural characteristics. This model can be a physical model based on finite element analysis (such as ANSYS, ABAQUS, etc.) or an AI proxy model built using algorithms such as neural networks and random forests. For example, in order to train an agent model, different combinations of tube sheet thickness, aperture, welding sequence, welding heat input (current, voltage, speed, etc.) can be selected, and finite element simulation can be used to run simulations on each set of data to obtain the corresponding temperature field and residual stress field results. These input and output pairs are then combined to form a training set, which is used to train a U-Net convolutional neural network that can quickly predict the temperature distribution map and stress distribution map under given welding conditions.

[0047] In a real-world welding control system, when preparing to execute a specific welding cycle (e.g., the third cycle), the system uses the actual geometric changes in the welded area from the previous two cycles (e.g., warpage and orifice offset) as initial input. This input, combined with the preset welding sequence, current, speed, and other parameters for this cycle, is fed into the trained thermal-stress simulation model. The model then calculates and outputs the predicted temperature and stress fields after the completion of the welding cycle, known as the "next cycle welding thermogram." In this thermogram, areas of concentrated temperature appear as red high-temperature zones, while areas of concentrated stress appear as high-intensity zones. By setting thresholds (e.g., warpage > 0.3mm, orifice offset > 0.05mm, and peak stress > 70% of the material's yield strength), the system automatically identifies "high-risk weld areas." For example, if the model output indicates severe heat accumulation in holes 24-26, and a predicted offset of 0.08mm for hole 25, with a peak stress of 320MPa, these locations are automatically marked as red risk areas by the system for parameter adjustment in the subsequent optimization module. The application of this model significantly improves the controllability, predictive ability and quality stability of the welding process, and avoids the uncertainty caused by relying on manual experience.

[0048] In one embodiment, S2: re-optimizing the welding process data of the high-risk welding zone, and randomly generating several sets of welding process data according to the constraints of the welding process are as follows:

[0049] The welding process data includes welding current, voltage, welding speed, shielding gas flow rate, wire feed speed, spot welding duration, cooling interval time and welding path sequence;

[0050] Several groups of welding process data combinations are randomly generated within the range of constraints through Latin hypercube sampling.

[0051] Specifically, before welding in high-risk areas, the welding process data needs to be optimized. This process first relies on a comprehensive understanding and definition of "welding process data." Typically, these data include welding current, voltage, welding speed, shielding gas flow, wire feed speed (if applicable), spot welding duration, cooling interval time, and welding path sequence, which directly determine the spatial distribution and intensity of welding heat input. Before optimization, clear constraints must be set to prevent the selected parameters from exceeding the safe or feasible range. The sources of constraints mainly include: the rated capacity of the welding equipment (such as maximum allowable current, minimum stable speed, etc.), material process specifications (such as the recommended parameter range for TIG welding for different materials), historical welding data feedback (such as whether the heat input of adjacent welds is too high), and welding path logic restrictions (such as forced cooling zone distribution, skip welding strategy, etc.). For example, the various parameters in the welding process data must meet strict process and equipment constraints to ensure stable and reliable welding quality. Specifically, the welding current should usually be controlled within the allowable range of the equipment, combined with the welding method and material thickness (for example, when TIG welding is used for 1.5mm thick stainless steel pipe heads, the recommended current is between 90A and 130A). If it is too large, it is easy to burn through, and if it is too small, the penetration depth is insufficient; the welding voltage should match the current to maintain a stable arc, generally controlled at 10V to 16V; the welding speed must take into account both the penetration depth and the forming quality, and the range is usually 26mm / s. Too fast speed will result in discontinuous welds, and too slow speed may result in excessive heat input; the shielding gas flow rate (such as argon or argon-helium mixture) needs to be maintained between 1020L / min to prevent oxygen in the weld The arc should be energized and stabilized. If MIG welding is used, the wire feed speed should also be controlled at 1060 mm / s to ensure stable weld fill. In pulse control mode, the spot welding duration is recommended to be set between 0.2 and 2.0 seconds to avoid structural distortion caused by excessive heat input. For multi-hole and multi-row structures, a reasonable cooling interval should be set for continuous high heat input areas, generally between 3 and 15 seconds, to avoid thermal stress accumulation. In addition, the welding path sequence must also be controlled. Continuous welding of multiple high-risk points is not recommended. Path strategies such as "jump spot welding," "symmetrical welding," and "reverse jump welding" should be prioritized to reduce warpage and stress peaks by properly distributing the heat source. All of the above parameters must be sampled within the set physical and process constraints during the optimization generation process to ensure the practical feasibility and welding safety of the generated solution.

[0052] Within these constraints, the system employs intelligent optimization algorithms, such as the Latin Hypercube Sampling algorithm, to generate multiple welding parameter combinations that satisfy both physical and process boundary conditions. These candidate parameter combinations are then integrated with the simulation model in subsequent steps to calculate their impact on the thermal field and stress distribution. This allows the system to identify the target welding solution with minimal risk and manageable deformation, thereby systematically controlling heat input uniformity and stress concentration risks, improving overall weld quality and structural stability.

[0053] In one implementation, Latin Hypercube Sampling (LHS) is used to randomly generate several sets of welding process data combinations within the constraints of the welding process. This effectively improves sampling uniformity and coverage, avoiding the sample concentration or repetitiveness issues that may occur with traditional random sampling methods. LHS is a space-filling strategy that ensures that each set of process data generated in a high-dimensional parameter space covers the entire parameter space as much as possible, thereby more comprehensively exploring the possibilities of different process parameter combinations. This method is particularly important in welding process optimization, as welding parameters (such as current, voltage, and speed) interact with each other in a complex manner, and welding process performance is affected by multiple factors. The multiple data combinations generated by LHS ensure extensive exploration within the process constraints, thereby increasing the probability of finding the optimal welding process parameter combination and reducing the risk of local optimal solutions. In addition, the LHS method, through uniform sampling, avoids excessive concentration on certain parameter values, ensuring the diversity and reliability of simulation and experimental results. Ultimately, while ensuring welding quality and safety, it optimizes the heat input and stress distribution throughout the welding process, improving the stability and durability of the welded structure.

[0054] In one embodiment, each set of welding process data is combined with the physical data of the tube head and the tube sheet and re-simulated using a preset thermal-stress simulation prediction model to obtain a welding thermogram of the next cycle corresponding to each set of welding process data as a second welding thermogram, and target welding process data is determined based on the first welding thermogram and the second welding thermogram.

[0055] Specifically, the steps of determining target welding process data according to the first welding thermogram and the second welding thermogram are as follows:

[0056] The first welding thermogram and the second welding thermogram are overlapped and compared, and the area corresponding to the original high-risk area in the first welding thermogram is recorded as the target area, and the area corresponding to the target area in the second welding thermogram is marked as the optimized area;

[0057] Calculate the temperature standard deviation and stress standard deviation of the target area and the optimized area respectively, and calculate the reduction ratio TD and SD of the temperature standard deviation T2 and stress standard deviation S2 of the optimized area compared with the temperature standard deviation T1 and stress standard deviation S1 of the target area. Add the ratios as the temperature stress optimization value of the high-risk area. The specific formula is:

[0058] The optimization effect value is calculated according to the temperature stress optimization value of the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

[0059] It should be noted that the data acquisition method involved in the calculation process mainly depends on real-time acquisition and the output of the simulation model. First, the temperature and stress data in the first welding thermogram and the second welding thermogram are generated by a thermal-stress simulation model (such as a physical model based on finite element analysis or an AI agent model). The model performs simulation calculations based on the physical data of the tube head and the tube sheet (such as material, size, aperture, welding process parameters, etc.). Then, the temperature field, stress field and deformation data of the welding process are collected in real time through a sensor system (such as a temperature sensor, stress sensor, laser displacement sensor, etc.) as input to the simulation model. The standard deviation of temperature and stress in the target area can be calculated by combining the temperature and stress data in the statistical thermogram with the numerical value of each grid point to finally obtain an evaluation of the optimization effect. When calculating the optimized value, the change in the standard deviation of temperature and stress reflects the impact of welding process optimization on the heat input distribution, which helps to judge the effectiveness of process improvement.

[0060] It should be noted that the temperature stress optimization value of the high-risk area is used to measure the optimization effect of the welding process in controlling the thermal stress and temperature distribution in the welding area. Specifically, this value reflects whether the temperature and stress distribution in the high-risk area has been effectively improved under the optimized welding process, especially whether the temperature gradient and stress concentration have been reduced. A larger temperature stress optimization value means that the heat input and stress distribution have been more effectively adjusted during the welding process, successfully reducing the accumulation of thermal stress and reducing the high-temperature area that may cause welding deformation, cracks and other problems. Therefore, when the temperature stress optimization value of the high-risk area is larger, it means that the effect of the welding process in improving welding quality, controlling deformation and improving structural reliability is more obvious, indicating that the welding process data is more in line with the optimization requirements and has a higher probability of being selected as the target welding process data to ensure the stability of the welding process and the quality of the final weld joint.

[0061] In one implementation, calculating optimized temperature and stress values ​​for high-risk areas offers the benefit of quantifying the actual effects of welding process optimization and providing actionable data support for the welding process. By optimizing the temperature and stress distribution in high-risk areas, problems such as deformation and cracking caused by uneven heat input and stress concentration can be effectively reduced during welding. A higher optimized value indicates more precise control of temperature and stress in the welding process, thereby improving the stability and consistency of weld quality.

[0062] In one embodiment, the steps of calculating the optimization effect value according to the temperature stress optimization value of the high-risk area and using the welding process data corresponding to the maximum optimization effect value as the target welding process data are as follows:

[0063] Obtain temperature data of each grid area in the first welding thermogram and the second welding thermogram. Typically, a thermogram includes temperature data of each point on the spatial grid.

[0064] Calculate the temperature gradient of each grid area in the first welding thermogram and the second welding thermogram For each grid point, in two dimensions, the temperature gradient is calculated by the difference between the horizontal and vertical directions, and the calculation formula is:

[0065]

[0066] Where, T i，j is the temperature value at position (i, j), Δx and Δy are the horizontal and vertical distances between grid points, respectively;

[0067] Calculate the temperature gradient of each grid point in the first welding thermogram and the second welding thermogram, respectively and Calculate the temperature gradient change: Where, is the temperature gradient change at position (i, j);

[0068] Calculate the temperature gradient change failure coefficient The calculation formula is: Where, and are the sum of the temperature gradient changes of all grid points in the second welding thermodynamic map and the sum of the temperature gradient changes of all grid points in the first welding thermodynamic map respectively;

[0069] The optimization effect value is calculated based on the temperature stress optimization value and the temperature gradient change unqualified coefficient in the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

[0070] It should be noted that the temperature gradient variation failure coefficient measures the extent of temperature gradient variation during the welding process, particularly in high-risk areas. Specifically, it reflects whether the optimized welding process effectively reduces abrupt temperature gradient changes and avoids unbalanced or drastic temperature fluctuations. Excessive temperature gradient variations can lead to localized thermal stress accumulation, increasing risks to the weld joint, such as cracks, deformation, and fatigue failure. Therefore, a smaller temperature gradient variation failure coefficient indicates that the optimized welding process is more effective in balancing the temperature field and reducing temperature gradient variations in high-risk areas. A smaller temperature gradient variation failure coefficient in high-risk areas indicates enhanced temperature field stability during the welding process, controlled thermal stress accumulation, and reduced the risk of potential weld defects. Consequently, the corresponding welding process data becomes more stable and reliable, with higher quality assurance, and is more likely to be used as target welding process data. This demonstrates that the process effectively optimizes the welding process, reduces potential thermal stress and deformation issues, and improves the quality and stability of the weld joint.

[0071] In one implementation, the benefit of calculating the temperature gradient variation failure coefficient is that it can quantify the temperature gradient fluctuations during the welding process, especially the degree of change in high-risk areas. Through this coefficient, the optimization effect of the welding process on the temperature distribution can be objectively evaluated, and then it can be determined whether there are excessive temperature changes or gradient mutations during the welding process. These factors are usually the cause of thermal stress accumulation and welding defects such as cracks and deformation. By reducing the temperature gradient variation failure coefficient, these problems can be effectively avoided, thereby improving the quality and structural stability of the welded joint. In addition, this coefficient also provides a clear direction for further optimizing the welding process, ensuring that better temperature control effects can be obtained under the combination of multiple welding process parameters, thereby reducing defective products in production and improving production efficiency and product consistency.

[0072] In one embodiment, the steps of calculating the optimization effect value based on the temperature stress optimization value and the temperature gradient change unqualified coefficient of the high-risk area, and using the welding process data corresponding to the maximum optimization effect value as the target welding process data are as follows:

[0073] The temperature stress optimization value and the temperature gradient change unqualified coefficient are normalized and mapped to the numerical range of 0-1. The weight values ​​of the normalized temperature stress optimization value and the temperature gradient change unqualified coefficient are both set to 0.5. The optimization effect value of the welding process data is calculated. The calculation formula is: HJ=0.5*yu-0.5*rf, where HJ is the optimization effective coefficient, yu is the temperature stress optimization value after normalization, and rf is the temperature gradient change unqualified coefficient after normalization.

[0074] It should be noted that the larger the optimization effect value, the more effective the corresponding welding process data is in reducing thermal stress and optimizing temperature gradients, thereby improving the quality and stability of the weld joint. Ultimately, the welding process data with the largest optimization effect value is selected as the target process data. This process ensures the scientific and accurate nature of the welding process and provides a quantitative basis for further improvement of welding quality.

[0075] In one embodiment, S4: welding the next round of tube heads and tube sheets according to the target welding process data; welding the next round of tube heads and tube sheets according to the target welding process data means precisely adjusting various parameters in the welding process, such as welding current, voltage, welding speed, etc., based on the optimal process parameters previously optimized and analyzed, to ensure that the welding process can accurately control heat input, temperature gradient, and stress distribution. This process not only relies on the welding thermodynamic map obtained in the early stage through the thermal-stress simulation prediction model, but also requires real-time adjustment based on the actual thermodynamic map after each welding to avoid the emergence of high-risk areas. Through continuous iterative optimization, the quality of the welded joint is gradually improved, the welding stress concentration and crack risk are reduced, and the stability and durability of the welded joint in high temperature and high pressure environments are guaranteed. This method not only improves the controllability of welding quality, but also optimizes the efficiency of the entire production process, so that each round of welding process can develop towards the optimal state.

[0076] In one embodiment, S5: after each circle of tube heads and tube sheets are welded, steps S1-S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger, further comprising:

[0077] After each round of welding of the tube head and the tube sheet is completed, the actual welding data of the tube head and the tube sheet after welding is obtained, and the actual welding data is compared with the standard welding data range. If the actual welding data is within the standard welding data range, steps S1-S4 are repeated to complete the welding of the remaining rounds of the heat exchanger tube head and the tube sheet;

[0078] If the actual welding data is not within the standard welding data range, an alarm will be issued immediately and the welding of the heat exchanger tube head and the remaining circles of the tube sheet will be stopped.

[0079] It should be noted that after each weld of the tube head and tube sheet is completed, actual welding data, such as temperature and stress, is first collected. This actual data is rigorously compared with a preset standard welding data range. Standard welding data ranges are typically set based on experience, simulation results, and industry standards, encompassing the reasonable fluctuation range of parameters such as temperature, pressure, and speed during the welding process. If the actual welding data falls within the standard welding data range, it indicates that the welding process meets the expected requirements and the weld quality is guaranteed. The system will automatically execute steps S1-S4 and continue welding the remaining welds. Conversely, if the actual welding data deviates from the standard range, the system will immediately issue an alarm and stop the welding operation. This mechanism effectively avoids potential quality issues caused by unqualified welding parameters, ensuring the accuracy and reliability of each weld, thereby ensuring the long-term stable operation of the heat exchanger. After stopping welding, the operator can check the welding equipment parameters, adjust the process, or recalibrate the equipment to ensure compliance with subsequent welds.

[0080] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A welding method for a heat exchanger tube head and a tube sheet, characterized in that: The following steps are involved: S1: collecting physical data of the tube head and tube sheet and currently set welding process data, and obtaining the welding thermodynamic map of the next cycle through a preset heat-stress simulation model as a first welding thermodynamic map, and determining high-risk welding areas in the welding thermodynamic map based on the first welding thermodynamic map; S2: re-optimize the welding process data of the high-risk welding zone and randomly generate several sets of welding process data according to the constraints of the welding process; S3: Re-simulating each set of welding process data in combination with the physical data of the tube head and the tube sheet using a preset thermal-stress simulation prediction model to obtain a new welding thermogram for the next cycle corresponding to each set of welding process data, recorded as a second welding thermogram, and determining target welding process data based on the first welding thermogram and the second welding thermogram; S4: welding the tube head and tube sheet of the next circle according to the target welding process data; S5: After each circle of tube heads and tube sheets are welded, steps S1-S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger.

2. A welding method for a heat exchanger tube head and a tube sheet according to claim 1, characterized in that: S2: Re-optimize the welding process data of the high-risk welding zone and randomly generate several sets of welding process data according to the constraints of the welding process: The welding process data includes welding current, voltage, welding speed, shielding gas flow rate, wire feed speed, spot welding duration, cooling interval time and welding path sequence; Several groups of welding process data combinations are randomly generated within the range of constraints through Latin hypercube sampling.

3. A welding method for a heat exchanger tube head and a tube sheet according to claim 1, characterized in that: The steps of determining target welding process data according to the first welding thermogram and the second welding thermogram are as follows: The first welding thermogram and the second welding thermogram are overlapped and compared, and the area corresponding to the original high-risk area in the first welding thermogram is recorded as the target area, and the area corresponding to the target area in the second welding thermogram is marked as the optimized area; Calculate the temperature standard deviation and stress standard deviation of the target area and the optimized area respectively, and calculate the reduction ratio TD and SD of the temperature standard deviation T2 and stress standard deviation S2 of the optimized area compared with the temperature standard deviation T1 and stress standard deviation S1 of the target area. Add the ratios as the temperature stress optimization value of the high-risk area. The specific formula is: The optimization effect value is calculated according to the temperature stress optimization value of the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

4. A welding method for a heat exchanger tube head and a tube sheet according to claim 3, characterized in that: The steps of calculating the optimization effect value based on the temperature stress optimization value of the high-risk area and taking the welding process data corresponding to the maximum optimization effect value as the target welding process data are as follows: Obtain temperature data of each grid area in the first welding thermogram and the second welding thermogram, where the thermogram includes temperature data of each point on the spatial grid; Calculate the temperature gradient of each grid area in the first welding thermogram and the second welding thermogram For each grid point, in two dimensions, the temperature gradient is calculated by the difference between the horizontal and vertical directions, and the calculation formula is: Where, T i，j is the temperature value at position (i, j), Δx and Δy are the horizontal and vertical distances between grid points, respectively; Calculate the temperature gradient of each grid point in the first welding thermogram and the second welding thermogram, respectively and Calculate the temperature gradient change: Where, is the temperature gradient change at position (i, j); Calculate the temperature gradient change failure coefficient The calculation formula is: Where, and are the sum of the temperature gradient changes of all grid points in the second welding thermodynamic map and the sum of the temperature gradient changes of all grid points in the first welding thermodynamic map respectively; The optimization effect value is calculated based on the temperature stress optimization value and the temperature gradient change unqualified coefficient in the high-risk area, and the welding process data corresponding to the maximum optimization effect value is used as the target welding process data.

5. The welding method for heat exchanger tube head and tube sheet according to claim 1, characterized in that: The steps for calculating the optimization effect value based on the temperature stress optimization value and the temperature gradient change unqualified coefficient in the high-risk area are as follows: The temperature stress optimization value and the temperature gradient change unqualified coefficient are normalized and mapped to the numerical range of 0-1. The weight values ​​of the normalized temperature stress optimization value and the temperature gradient change unqualified coefficient are both set to 0.

5. The optimization effect value of the welding process data is calculated. The calculation formula is: HJ=0.5*yu-0.5*rf, where HJ is the optimization effective coefficient, yu is the temperature stress optimization value after normalization, and rf is the temperature gradient change unqualified coefficient after normalization.

6. A welding method for a heat exchanger tube head and a tube sheet according to claim 1, characterized in that: S5: After each circle of tube heads and tube sheets are welded, steps S1 to S4 are repeated to complete the welding of the remaining circles of tube heads and tube sheets of the heat exchanger, which also includes: After each round of welding of the tube head and the tube sheet is completed, the actual welding data of the tube head and the tube sheet after welding is obtained, and the actual welding data is compared with the standard welding data range. If the actual welding data is within the standard welding data range, steps S1-S4 are repeated to complete the welding of the remaining rounds of the heat exchanger tube head and the tube sheet; If the actual welding data is not within the standard welding data range, an alarm will be issued immediately and the welding of the heat exchanger tube head and the remaining circles of the tube sheet will be stopped.

Images