A method of welding a superhard material seal body multilayer structure

Abstract

The application relates to the field of intelligent welding processes and control, and discloses a welding method for a multilayer structure of a superhard material sealing body. The method comprises the following steps: acquiring multilayer structure data and generating a three-dimensional digital model; optimizing energy input parameters according to a heat affected zone and a stress concentration area to generate an initial energy tuning parameter set; collecting temperature field data and stress field data in a welding process according to the initial energy tuning parameter set to obtain a dynamic heat affected zone change trend; adjusting energy input and a scanning path according to the dynamic heat affected zone change trend to generate a corrected energy tuning parameter; controlling the depth distribution of energy in the multilayer structure according to the corrected energy tuning parameter to obtain a fusion depth and a heat affected zone distribution; and adjusting energy input and a scanning speed according to the fusion depth and the heat affected zone distribution to generate an energy input parameter set. The application improves the precision of energy control in the welding process.

Description

Technical Field

[0001] This invention relates to the field of intelligent welding processes and control, and in particular to a welding method for a multi-layer structure of a superhard material sealing body. Background Technology

[0002] Multilayer electron beam welding technology faces numerous technical challenges, primarily in precisely controlling energy input to ensure high-quality welding results. Due to the differences in thermophysical properties of the various materials within the multilayer structure, thermal stress concentration and deformation can easily occur during welding, affecting the quality and mechanical properties of the weld joint. The different coefficients of thermal expansion, thermal conductivity, and melting points of the materials result in uneven heat transfer between layers during welding, leading to stress concentration zones and causing weld cracks or deformation. Furthermore, an excessively large heat-affected zone can alter the microstructure of the weld joint, causing grain growth and a decrease in material mechanical properties, thus affecting the strength and toughness of the weld joint.

[0003] Furthermore, the distribution of welding energy in multilayer structures is difficult to control precisely, easily leading to insufficient weld depth or over-melting. Insufficient weld depth may result in incomplete fusion of the joint, affecting its strength; while over-melting causes material loss, compromising the structural stability of the weld zone. These problems are interconnected, forming a complex technical contradiction. To ensure weld quality while minimizing the heat-affected zone and interlayer stress, it is essential to monitor the temperature and stress fields in real time during the welding process and dynamically adjust the energy input and scanning path based on this data. Due to the complexity of multilayer dissimilar material structures, real-time monitoring and analysis of the temperature and stress fields remain urgent problems to be solved. Summary of the Invention

[0004] This application provides a welding method for a multi-layer structure of a superhard material seal, which improves the accuracy of energy control during the welding process.

[0005] This invention provides a welding method for a multi-layer structure of a superhard material sealing body, the method comprising:

[0006] Acquire geometric shape data and material property data of multi-layer structures, and generate three-dimensional digital models of multi-layer structures;

[0007] Based on the three-dimensional digital model, the heat conduction and stress distribution of the multi-layer structure under different energy inputs are simulated to obtain the heat-affected zone and stress concentration area.

[0008] Based on the heat-affected zone and the stress concentration region, optimize the energy input parameters to generate an initial energy tuning parameter set;

[0009] Based on the initial energy tuning parameter set, temperature field data and stress field data during the welding process are collected to obtain the dynamic heat-affected zone variation trend.

[0010] The energy input and scanning path are adjusted according to the changing trend of the dynamic heat-affected zone to generate the corrected energy tuning parameters.

[0011] The energy depth distribution in the multilayer structure is controlled by the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution.

[0012] The energy input and scanning speed are adjusted according to the fusion depth and the heat-affected zone distribution to generate the final energy input parameter set;

[0013] Based on the final energy input parameter set, the energy input and scanning path are adjusted again to obtain the welding results and interlayer stress distribution of the multi-layer structure.

[0014] Furthermore, acquiring geometric shape data and material property data of the multilayer structure to generate a three-dimensional digital model of the multilayer structure includes: acquiring the thickness, melting point, and thermal conductivity data of each layer of material; using a three-dimensional modeling algorithm to generate the three-dimensional digital model based on the thickness, melting point, and thermal conductivity data, obtaining the thickness distribution, melting point distribution, and thermal conductivity distribution of each layer of material; determining the geometric features and material property distribution of each layer of material based on the three-dimensional digital model; and establishing a digital representation of the multilayer structure through the geometric features and material property distribution.

[0015] Furthermore, based on the three-dimensional digital model, the heat conduction and stress distribution of the multilayer structure under different energy inputs are simulated to obtain the heat-affected zone and stress concentration region, including: acquiring the thickness, melting point, and thermal conductivity data of each layer of material; using a three-dimensional modeling algorithm, generating the three-dimensional digital model based on the thickness, melting point, and thermal conductivity data to obtain the thickness distribution, melting point distribution, and thermal conductivity distribution of each layer of material; determining the geometric features and material property distribution of each layer of material based on the three-dimensional digital model; and establishing a digital representation of the multilayer structure through the geometric features and material property distribution.

[0016] Furthermore, the step of optimizing the energy input parameters based on the heat-affected zone and the stress concentration region to generate an initial energy tuning parameter set includes: if the range of the heat-affected zone exceeds a first preset threshold, then using a genetic algorithm to optimize the electron beam energy level parameters; generating a differentiated energy input scheme based on the thickness and melting point of each layer of material; and determining the initial energy level tuning parameter set based on the differentiated energy input scheme.

[0017] Furthermore, the step of collecting temperature field data and stress field data during the welding process based on the initial energy tuning parameter set to obtain the dynamic heat-affected zone variation trend includes: collecting the temperature field data through sensors and calculating the range of the dynamic heat-affected zone based on the temperature field data; collecting the stress field data through sensors and determining the dynamic heat-affected zone variation trend based on the stress field data and the temperature field data.

[0018] Furthermore, the step of adjusting the energy input and scanning path according to the changing trend of the dynamic heat-affected zone to generate corrected energy tuning parameters includes: reducing the electron beam energy level if the temperature of the shallow material exceeds the melting point threshold; adjusting the scanning path according to the changing trend of the dynamic heat-affected zone; generating corrected energy tuning parameters based on the reduced electron beam energy level and the adjusted scanning path; and optimizing the energy input scheme during the welding process using the corrected energy tuning parameters.

[0019] Furthermore, the step of controlling the depth distribution of energy in the multilayer structure according to the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution includes: using a layered scanning algorithm to control the depth distribution of the electron beam in the multilayer structure according to the modified energy tuning parameters; calculating the fusion depth of each layer of material according to the depth distribution; determining the heat-affected zone distribution of each layer of material according to the fusion depth; and verifying the stability of the welding process through the heat-affected zone distribution.

[0020] Furthermore, the step of adjusting the energy input and scanning speed according to the fusion depth and the heat-affected zone distribution to generate a final energy input parameter set includes: if the fusion depth of the deep material is lower than a second preset threshold, then using an adaptive adjustment algorithm to increase the electron beam energy level; optimizing the electron beam scanning speed according to the fusion depth; generating the final energy input parameter set according to the increased electron beam energy level and the optimized scanning speed; and determining welding process parameters through the final energy input parameter set for the complete welding of multi-layer structures.

[0021] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0022] This invention discloses a method for electron beam welding of multilayer structures. By acquiring the geometric shape and material property data of the multilayer structure, a three-dimensional digital model is generated to simulate heat conduction and stress distribution under different energy inputs, thus identifying the heat-affected zone (HAZ) and stress concentration areas. Based on this, energy input parameters are optimized to generate an initial energy tuning parameter set. During the welding process, temperature and stress field data are acquired in real time, and the dynamic HAZ variation trend is analyzed to adjust the energy input and scanning path accordingly. By controlling the depth distribution of energy in the multilayer structure, the fusion depth and HAZ distribution are obtained, further optimizing the energy input and scanning speed, ultimately generating the optimal energy input parameter set. This invention can precisely control the welding process of multilayer dissimilar materials, effectively reduce the HAZ size, lower interlayer stress, and improve welding quality and efficiency. Attached Figure Description

[0023] Figure 1 This is a flowchart of a welding method for a multi-layer structure of a superhard material sealing body according to the present invention;

[0024] Figure 2 This is a schematic diagram of a welding method for a multi-layer structure of a superhard material sealing body according to the present invention;

[0025] Figure 3 The diagram shows the optimization results for two material scenarios of this invention.

[0026] Figure 4 This is a schematic diagram of the adaptive adjustment PID controller structure of the present invention. Detailed Implementation

[0027] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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.

[0028] like Figure 1 and Figure 2 The welding method for a multi-layer structure of a superhard material sealing body in this embodiment may specifically include:

[0029] S1. Obtain the geometric shape data and material property data of the multi-layer structure, and generate a three-dimensional digital model of the multi-layer structure.

[0030] In one specific embodiment, the step of acquiring the geometric shape data and material property data of the multi-layer structure and generating a three-dimensional digital model of the multi-layer structure includes:

[0031] (1) Obtain the thickness, melting point and thermal conductivity data of each layer of material;

[0032] (2) Using a three-dimensional modeling algorithm, the three-dimensional digital model is generated based on the thickness, melting point and thermal conductivity data to obtain the thickness distribution, melting point distribution and thermal conductivity distribution of each layer of material;

[0033] (3) Based on the three-dimensional digital model, determine the geometric characteristics and material property distribution of each layer of material;

[0034] (4) A digital representation of the multilayer structure is established by means of the geometric features and the distribution of the material properties.

[0035] Specifically, the thickness data of each layer of the multi-layer structure is collected using measuring equipment. For example, an ultrasonic thickness gauge is used to scan the interlayer thickness of the superhard material seal, obtaining values ​​such as a shallow layer thickness of 0.5 mm and a deep layer thickness of 1.2 mm. Melting point and thermal conductivity data are obtained by querying a material database. For example, the melting point of the shallow material is 1500 degrees Celsius and the thermal conductivity is 200 W / m·K, while the melting point of the deep material is 1800 degrees Celsius and the thermal conductivity is 150 W / m·K.

[0036] Geometric features are converted into vector data, such as converting surface geometry into NURBS curve representations. Property layers are created by combining attribute distributions, such as adding melting point and thermal conductivity values ​​to each geometric unit, forming a complete digital representation. This representation is then imported into simulation software, such as ANSYS, for heat conduction simulation. After inputting heat source parameters, the temperature field is calculated. For example, in a sealed body welding scenario, using defined geometric features such as the interlayer interface curvature of 0.1 mm⁻¹ and attribute distributions such as thermal conductivity gradients, a digital representation can be established to simulate the heat-affected zone when an electron beam input of 10 kW is applied, showing that the shallow layer temperature rises to 1600 degrees Celsius while the deeper layer remains below 1200 degrees Celsius. This representation is beneficial for predicting stress concentration and avoiding welding defects.

[0037] For thick structures, the digital representation can be enhanced by adding boundary conditions. For example, environmental temperature constraints can be incorporated into the geometric features, and density data can be added to the attribute distribution as an auxiliary measure even if it is not obtained. This ensures more accurate heat conduction during simulation and is beneficial for controlling the depth of multilayer fusion.

[0038] S2. Based on the three-dimensional digital model, simulate the heat conduction and stress distribution of the multi-layer structure under different energy inputs to obtain the heat-affected zone and stress concentration area.

[0039] In one specific embodiment, based on the three-dimensional digital model, the heat conduction and stress distribution of the multilayer structure under different energy inputs are simulated to obtain the heat-affected zone and stress concentration region, including:

[0040] (1) The three-dimensional digital model is meshed using the finite element analysis method;

[0041] (2) Based on the grid division, the heat conduction process under different electron beam energy levels is simulated to obtain the temperature field distribution of each layer of material;

[0042] (3) Calculate the range of the heat-affected zone of each layer of material based on the temperature field distribution, and determine the heat-affected zone;

[0043] (4) Based on the temperature field distribution and material properties, simulate the interlayer stress distribution and determine the stress concentration region.

[0044] Specifically, using the finite element method (FEM), the multi-layered structure is divided into a three-dimensional mesh. For example, the overall structure is divided into 1000 mesh elements, each assigned a corresponding thickness value; for instance, the thickness of the shallow mesh gradually changes from 0.4 mm to 0.6 mm. Based on the mesh elements, interpolation algorithms are applied to calculate the melting point distribution. For example, the Kriging interpolation method is used to extrapolate the melting point field of the entire layer based on known melting point data points, ensuring continuous distribution; for example, the melting point distribution of the shallow layer is between 1450 degrees Celsius and 1550 degrees Celsius. Similarly, the thermal conductivity distribution is calculated; for example, the radial basis function interpolation algorithm is used to generate a thermal conductivity distribution map from 180 W / m·K to 220 W / m·K. This distribution data is directly used for model visualization, improving the accuracy of subsequent simulations.

[0045] For example, a multilayer structure includes three layers: an outer tungsten alloy layer, a middle ceramic layer, and an inner titanium alloy layer. Data were obtained for the outer layer with a thickness of 0.8 mm, a melting point of 3422°C, and a thermal conductivity of 174 W / m·K; the middle layer with a thickness of 1.0 mm, a melting point of 2050°C, and a thermal conductivity of 20 W / m·K; and the inner layer with a thickness of 0.6 mm, a melting point of 1668°C, and a thermal conductivity of 22 W / m·K. Using 3D modeling algorithms such as voxel modeling, these data were mapped into a digital model, resulting in an image showing a uniform thickness distribution in the outer layer and a slight gradient in the middle layer. The melting point distribution highlighted the lower thermal conductivity areas in the middle layer, while the thermal conductivity distribution emphasized the high thermal conductivity areas in the outer layer. This method is beneficial for identifying potential thermal inhomogeneities and optimizing pre-welding preparation.

[0046] For different scenarios, such as high thermal conductivity multilayer structures, the model can be extended to use parametric modeling algorithms. After inputting thickness data, a variable thickness distribution can be automatically generated. For example, the thickness increases from 0.3 mm at the edge to 0.7 mm towards the center. Simultaneously, the melting point distribution is linearly interpolated to obtain a gradient from 1400 degrees Celsius to 1600 degrees Celsius, and the thermal conductivity distribution is adjusted accordingly from 100 W / m·K to 300 W / m·K. This extension is beneficial for handling non-uniform materials and improves the model's adaptability.

[0047] Based on the three-dimensional digital model, the geometric features and material property distributions of each layer are determined. Geometric features, such as the curvature and boundary shape of each layer, are extracted from the model; for example, shallow layers are planar and deeper layers are curved. Property distributions are integrated, such as superimposing thickness distribution with melting point distribution, to obtain a comprehensive property map. These features and distributions serve as the basis for establishing a digital representation.

[0048] S3. Optimize the energy input parameters based on the heat-affected zone and the stress concentration region to generate an initial energy tuning parameter set.

[0049] In one specific embodiment, the step of optimizing the energy input parameters based on the heat-affected zone and the stress concentration region to generate an initial energy tuning parameter set includes:

[0050] (1) If the range of the heat-affected zone exceeds the preset threshold, the electron beam energy level parameters are optimized using a genetic algorithm;

[0051] (2) Generate differentiated energy input schemes based on the thickness and melting point of each layer of material;

[0052] (3) Determine the initial energy level tuning parameter set according to the differentiated energy input scheme.

[0053] Specifically, the heat-affected zone (HAZ) range data obtained from finite element analysis is compared with a preset threshold, for example, 2 mm. If the HAZ is exceeded, the optimization process is triggered. The genetic algorithm population is initialized, using electron beam energy level parameters such as voltage and current as chromosomes. An initial population of 50 individuals is randomly generated, each representing a set of energy level values. A fitness function is defined, and the fitness value is calculated based on minimizing the simulated HAZ and the uniformity of stress distribution. Through iterative selection, crossover, and mutation operations, optimization is performed until convergence, yielding the optimized electron beam energy level parameters. For example, in the welding of multilayer seals made of ultrahard materials, if the HAZ exceeds the threshold, the genetic algorithm can optimize from an initial energy level of 10 kW to 8 kW, reducing heat diffusion.

[0054] Thickness data for each layer is extracted from the 3D digital model, such as 0.5 mm for shallow layers and 1 mm for deep layers, as well as melting point data, such as 1500 degrees Celsius for shallow layers and 2000 degrees Celsius for deep layers. For thinner layers with lower melting points, lower energy input is allocated to avoid over-melting; for example, a gradual energy level transition is used for shallow layers. For thicker layers with higher melting points, energy is increased to ensure fusion. A combined generation scheme is used to ensure the total energy distribution matches the multilayer structure, avoiding uneven thermal distribution between layers.

[0055] The energy allocation in the scheme is converted into specific parameters, such as an initial voltage of 15 kV and a current of 200 mA, forming a parameter set.

[0056] Specifically, the energy input control during the welding process is initialized using the initial energy level tuning parameter set. This includes importing the parameter set into the control system and using these parameters to monitor the initial thermal field at the start of welding. For example, for welding multi-layer structures, the application of the initial parameter set can ensure that the heat-affected zone is controlled within a threshold from the initial stage, improving the overall fusion uniformity.

[0057] For the optimization of the genetic algorithm, different electron beam energy level scenarios are considered. For example, in structures dominated by high-melting-point layers, the algorithm iteration count is set to 100, and the fitness function emphasizes the melting point threshold to avoid overheating. In scenarios with many low-melting-point layers, the iteration focuses on minimizing the hot zone range. This multi-faceted approach supports the robustness of the optimization and is beneficial for adapting to various material combinations. (Refer to...) Figure 3 The two figures show the process of optimizing the electron beam energy level using a genetic algorithm in high-melting-point and low-melting-point material scenarios, respectively. The fitness value changes gradually with the number of generations, reflecting the optimization effect and convergence trend.

[0058] For example, the generation of differentiated solutions can be extended to scenarios that include thermal conductivity data. If the shallow layer has high thermal conductivity, the energy input can be further reduced to 60% to ensure that heat is not excessively conducted to the deeper layers, forming a logically progressive solution and improving the accuracy and reliability of welding.

[0059] S4. Based on the initial energy tuning parameter set, collect temperature field data and stress field data during the welding process to obtain the dynamic heat-affected zone change trend.

[0060] In one specific embodiment, the step of acquiring temperature field data and stress field data during the welding process based on the initial energy tuning parameter set to obtain the dynamic heat-affected zone variation trend includes:

[0061] (1) Collect the temperature field data through the sensor, and calculate the range of the dynamic heat-affected zone based on the temperature field data;

[0062] (2) Collect the stress field data using sensors, and determine the changing trend of the dynamic heat-affected zone based on the stress field data and the temperature field data;

[0063] (3) Based on the range of the dynamic heat-affected zone, the influence of thermal cycling effect on the performance of welding materials during temperature change is further considered. Temperature fluctuation and thermal cycling simulation are introduced to obtain thermal cycling simulation results.

[0064] Specifically, by collecting temperature field data during the welding process using sensors, the temperature distribution in the welding area can be monitored in real time. This temperature field data is fed back to the system via temperature sensors, reflecting temperature changes during welding. Based on this temperature data, combined with the thermophysical properties of the welding material, such as thermal conductivity, specific heat capacity, and coefficient of thermal expansion, the system can calculate the temperature distribution and thermal gradient at different time points. Analyzing this temperature data allows identification of the heat-affected zone (HAZ), which is typically the region where the material undergoes significant structural changes due to high temperatures. When the temperature exceeds a certain threshold, the welding material undergoes a phase transition or structural change; this region is the HAZ. Based on the temperature change trend during welding, the size, shape, and expansion or contraction of the HAZ over time can be determined. Temperature field data can also be used to calculate the temperature peaks and cooling rates at different locations during welding, which is crucial for further optimizing the welding process and controlling the size of the HAZ. Precise analysis of the temperature field ensures that the welding quality is not affected by an excessively large HAZ.

[0065] A real-time feedback control algorithm is adopted, which collects temperature field data during the welding process through sensors and monitors the temperature distribution of each layer of the multi-layer structure in real time through sensors such as infrared thermal imagers to form temperature field data. The real-time feedback control algorithm is based on the proportional-integral-derivative control mechanism.

[0066] Specifically, the sensor sampling interval is set according to the initial energy tuning parameter set. For example, in a scenario where the electron beam energy level is at a medium level, the sampling interval is set to 0.1 seconds to capture rapid temperature changes in shallow materials.

[0067] For example, when welding a shallow layer of superhard material with a thickness of 2 mm, the sensor collected temperature field data showing that the local temperature peak was 1500 degrees Celsius. This helps in the subsequent calculation of the heat-affected zone range. The beneficial effect is that it improves the real-time performance of the data and avoids material damage caused by overheating.

[0068] Based on the temperature field data, the dynamic heat-affected zone (HAZ) is calculated. Specifically, this involves simulating the heat diffusion boundary using the heat conduction equation based on Fourier's law, and defining the region where the temperature gradient exceeds a threshold as the HAZ. The temperature field data is then meshed to generate a temperature distribution matrix; for example, a multi-layered structure is divided into a 100x100 grid, and the temperature value is calculated for each grid point. Based on the temperature distribution matrix, connected regions where the temperature exceeds the material's melting point threshold are identified, and their radii are calculated as the HAZ. For example, in a material with a melting point of 1800 degrees Celsius, a calculated range of 5 mm indicates that heat diffusion has affected adjacent layers.

[0069] For intermediate layer materials with high thermal conductivity, the calculated dynamic heat-affected zone (HAZ) range can be reduced to 3 mm. This is achieved through iterative solving of the heat conduction equation, where Fourier's law describes the heat flux density as proportional to the temperature gradient, simplified to q = -k∇T, where k is the thermal conductivity and ∇T is the temperature gradient. This calculation process ensures the accuracy of the range and optimizes energy input, reducing interlayer thermal stress accumulation. In another embodiment, if the shallow layer thickness is 1 mm and the melting point is low, the calculation range is extended to 7 mm. By adjusting the mesh resolution from 50x50 to 200x200, the calculation accuracy is improved, demonstrating adaptability under different material properties.

[0070] Stress field data is collected during the welding process using sensors. Specifically, strain gauges or fiber optic sensors are used to monitor the deformation and stress distribution of multi-layer structures during welding, forming stress field data. For example, in a scenario where the electron beam scanning path is linear, the stress field data shows a peak stress of 500 MPa, which is directly used for subsequent trend determination.

[0071] Stress and temperature field data are fused, and time series analysis is applied to predict the rate of change of the heat-affected zone (HAZ). The time series analysis is based on an autoregressive integral moving average (ARM) model to assess trends such as expansion or contraction. The fused data is normalized to generate a joint dataset, for example, by mapping temperature and stress values ​​to the 0-1 interval. The joint dataset is then fitted using the ARM model to calculate a trend coefficient. For example, a positive coefficient of 0.2 indicates that the HAZ expands by 0.2 mm per second, which guides the reduction of energy levels to control the trend.

[0072] For scenarios with uneven melting point distribution in multi-layered structures, the autoregressive integral moving average model predicts future trends through historical data sequences. After assuming the data is stable, the model uses differencing, such as performing first-order differencing on the temperature field data sequence to eliminate trends, and then estimates the parameters to predict the range of the thermally affected zone at the next moment.

[0073] Specifically, if the current stress field data indicates an interlayer stress concentration of 300 MPa and a temperature field peak of 1600 degrees Celsius, the model calculation trend is expansionary, leading to a 10% reduction in subsequent adjustment energy input. This has the beneficial effect of preventing insufficient fusion of deep materials and maintaining welding stability. In another embodiment, when deep materials with low thermal conductivity are involved, the fusion data may show a contractionary trend with a coefficient of negative -0.15. By smoothing noise through the model's moving average, the prediction accuracy is improved to 95%. This supports the effectiveness of the method from both real-time and accuracy perspectives, together forming a complete support for welding process optimization.

[0074] In one specific embodiment, based on the range of the dynamic heat-affected zone, the influence of thermal cycling on the performance of the welding material during temperature changes is further considered. Temperature fluctuation and thermal cycling simulation are introduced to obtain thermal cycling simulation results, including:

[0075] (1) Introduce a temperature fluctuation model to simulate the temperature fluctuation characteristics during the welding process;

[0076] (2) Apply thermal cycling simulation method to predict the cycle process of multiple heat loading and cooling during welding;

[0077] (3) Based on the results of thermal cycling simulation, evaluate the effect of temperature change on the performance of welding materials. The effect of temperature change on the performance of welding materials includes the strength, hardness and toughness of the materials.

[0078] Specifically, by introducing a temperature fluctuation model, the temperature fluctuation characteristics during the welding process were simulated, particularly the impact of rapid heating and cooling changes on the material. This model accurately captures temperature fluctuations in the welding zone, ensuring that temperature changes at different stages are fully simulated, thus providing detailed temperature data support for subsequent thermal cycling assessments. Further application of thermal cycling simulation methods allowed for detailed predictions of multiple heat loading and cooling cycles during the welding process. Each heat loading and cooling cycle leads to changes in the material's thermal stress. This repeated heat loading not only affects the physical state of the material's surface layer but also creates a cumulative thermal fatigue effect internally. Through thermal cycling simulation, the accumulation of thermal stress caused by each round of thermal cycling can be clearly predicted, thereby affecting the quality of the weld area. Combining the thermal cycling simulation results, the impact of temperature changes on the performance of the welding material was evaluated. During the simulation, the material's strength, hardness, and toughness were significantly affected by temperature changes. With continuous temperature fluctuations, the material's microstructure changes, especially at high temperatures, where the grain structure may undergo phase transitions, leading to a decrease in strength and hardness, and even potentially brittle damage. Furthermore, after repeated thermal cycling, the toughness of materials typically decreases, which can manifest as cracks or brittle fracture in the weld area. This process allows for a more comprehensive understanding of the thermal cycling effects during welding and provides a theoretical basis for subsequent optimization of welding techniques.

[0079] Taking multilayer welding of tungsten alloy and titanium alloy as an example, this study considers the impact of thermal cycling effects on the properties of the welding materials during temperature changes, based on the dynamic heat-affected zone. A temperature fluctuation model is introduced to simulate the temperature fluctuation characteristics of the contact area between the tungsten alloy and titanium alloy during welding, particularly the influence of electron beam heating and cooling processes on the contact layer. Simulation results show that in the initial stage of welding, the surface temperature of the tungsten alloy rises rapidly, while the temperature of the titanium alloy is relatively low, creating a temperature difference between the two materials, which is crucial to the weld quality. Thermal cycling simulation is applied to predict multiple cycles of heat loading and cooling, analyzing the thermal expansion and contraction of the materials during heating and cooling. In multiple thermal cycles, the difference in the coefficients of thermal expansion between the tungsten alloy and titanium alloy leads to significant stress at the interface, potentially causing microcrack formation. Analysis of the thermal cycling simulation results evaluates the impact of temperature changes on the strength, hardness, and toughness of the tungsten alloy and titanium alloy. With temperature fluctuations, the hardness of the tungsten alloy decreases significantly, while the strength of the titanium alloy weakens at high temperatures, especially during repeated heating and cooling processes, where toughness also declines significantly, potentially leading to cracks or delamination in the weld area.

[0080] S5. Adjust the energy input and scanning path according to the changing trend of the dynamic heat-affected zone to generate the corrected energy tuning parameters.

[0081] In one specific embodiment, adjusting the energy input and scanning path according to the changing trend of the dynamic heat-affected zone to generate corrected energy tuning parameters includes:

[0082] (1) If the temperature of the shallow material exceeds the melting point threshold, then reduce the electron beam energy level;

[0083] (2) Adjust the scanning path according to the changing trend of the dynamic heat-affected zone;

[0084] (3) Based on the reduced electron beam energy level and the adjusted scanning path, the corrected energy tuning parameters are generated.

[0085] Specifically, if the temperature of the shallow material exceeds the melting point threshold, the electron beam energy level is immediately reduced to avoid overheating that could damage the material or cause welding defects. By monitoring the dynamic heat-affected zone (HAZ) trend, the scanning path is adjusted in real time to ensure that heat does not concentrate in a localized area, reducing stress concentration or material damage caused by overheating. Simultaneously, by evaluating the welding material performance in conjunction with temperature changes, the energy distribution scheme is optimized to ensure that the energy input in different regions matches the thermal behavior of the material.

[0086] For example, for thicker or higher melting point areas, the energy input can be appropriately increased to ensure sufficient fusion, while for thinner or low-melting-point materials, the energy input can be reduced to avoid overheating. Through these dynamic adjustments, combined with the reduced electron beam energy level and optimized scanning path, a modified energy tuning parameter set is generated to ensure that the energy distribution of each layer of material reaches the optimal state during the welding process, thereby improving the overall welding quality and reducing the occurrence of welding defects.

[0087] Specifically, the temperature data of the shallow material is monitored in real time, and a reduction operation is triggered by comparing the current temperature with a preset melting point threshold. For example, when the temperature of the shallow material reaches 105% of the threshold, the electron beam energy level is immediately reduced by 10% from its initial value. The reduction magnitude is calculated based on the degree of temperature exceedance, and a linear interpolation method is used to determine the specific energy level value to ensure reduced heat input and avoid over-melting. The electron beam scanning path is adjusted according to the dynamic heat-affected zone (HAZ) change trend. The dynamic HAZ change trend is analyzed, and the expansion rate and direction of the HAZ are calculated using the finite difference method. For example, in a multilayer structure, if the HAZ expands into the shallow layer at a rate exceeding 0.5 mm / s, the path is preferentially adjusted to avoid the high-heat region.

[0088] Trend data, including the boundary coordinates and rate of change of the heat-affected zone, is collected to form a trend vector for path planning. Based on the trend vector, the A* algorithm is applied to generate a new scanning path, which minimizes heat accumulation by evaluating path costs. For example, the path may be changed from a straight line to a spiral to disperse heat. The effectiveness of the adjusted path is verified by iterative simulation to confirm its effectiveness.

[0089] The adjustment process can be applied to welding multilayer seals made of superhard materials. For example, when the shallow layer is a high-melting-point alloy, trend analysis shows that the heat-affected zone diffuses upwards. The adjustment path is to increase the scanning interval by 20%, thereby reducing local thermal stress and improving welding uniformity. Based on the reduced electron beam energy level and the adjusted scanning path, corrected energy tuning parameters are generated. The reduced energy level values ​​and path data are integrated, and the tuning parameters are calculated using a weighted average method. For example, the energy level weight accounts for 60%, and the path weight accounts for 40%, generating a comprehensive parameter set. The output parameters include energy level amplitude and path coordinates, ensuring that the parameter set supports real-time application. The energy input scheme during the welding process is optimized using the corrected energy tuning parameters. The corrected parameters are then applied to the welding control system.

[0090] For example, tuning parameters are input into a PID controller to achieve closed-loop regulation of energy input, simulating the optimization effect. The optimized heat distribution is predicted using the finite element method to ensure that the heat-affected zone is controlled within a threshold. The optimization scheme is iterated, and if the simulation shows that the residual stress is too high, the parameters are fine-tuned to further distribute the energy. The final input scheme is generated, including hierarchical energy distribution.

[0091] For example, low-energy inputs can be assigned to shallow layers with a thickness of 0.5 mm, while gradually increasing inputs can be assigned to deeper layers. This optimization can be extended to scenarios with different material properties. For instance, in multilayer structures with uneven melting point distribution, the energy input scheme can be adjusted to a pulse mode using modified parameters, reducing the energy level from the initial 5 kW to 3 kW and changing the path to multi-circle scanning. This helps to minimize interlayer stress concentration and improve the overall durability of the seal.

[0092] For example, in the welding of superhard material seals, if the shallow layer temperature exceeds the melting point threshold of 1500℃, the energy level is reduced to 80% of the initial value, and the path is adjusted to a Z-shape according to the trend of the heat-affected zone, reducing heat accumulation by 20%. A parameter set is generated, such as an energy level of 3.5kW and a path spacing of 1mm. The optimized scheme ensures uniform fusion depth and brings beneficial effects such as reducing the defect rate by 15%.

[0093] For multilayer structures with high thermal conductivity, the thermal diffusion equation is used to predict changes. The path is adjusted to a deceleration scan, and combined with optimization, the energy input is changed from continuous to intermittent, enhancing the accuracy of thermal control and helping to prevent shallow overheating. For example, assuming the shallow thermal conductivity is 200 W / m·K, the dynamic trend shows that the hot zone expands rapidly. After reducing the energy level, the path is adjusted to increase the dwell time by 0.2 s. The resulting tuning parameters optimize the input scheme, reducing the heat-affected zone by 30% and improving welding quality.

[0094] S6. Control the depth distribution of energy in the multilayer structure according to the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution.

[0095] In one specific embodiment, controlling the depth distribution of energy in the multilayer structure according to the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution includes:

[0096] (1) Using a layered scanning algorithm, the depth distribution of the electron beam in the multilayer structure is controlled according to the modified energy tuning parameters, and the fusion depth of each layer of material is calculated according to the depth distribution;

[0097] (2) Determine the distribution of the heat-affected zone of each layer of material based on the fusion depth.

[0098] Specifically, the initial power and focal depth of the electron beam are determined based on the corrected energy tuning parameters, which include energy level values ​​for the thickness and melting point of each material layer. For example, in a multilayer superhard material seal, the energy level is set to 80kV when the first layer is 0.5mm thick and has a melting point of 1500℃, and adjusted to 100kV when the second layer is 1.0mm thick and has a melting point of 1800℃, to achieve layer-by-layer energy matching. The multilayer structure is divided into multiple scanning layers, each corresponding to one material layer, and the energy penetration depth of each layer is calculated based on parameters. For example, the energy attenuation from the surface to the deeper layers is estimated using the heat conduction equation to ensure that the shallow layers do not over-melt. The electron beam is controlled to scan each layer along a preset path, and the beam intensity is dynamically adjusted to ensure uniform depth distribution. For example, during welding, if the sensor detects that the temperature of the shallow layer is close to the melting point, the beam current is reduced to 90% of the initial value, thereby avoiding interlayer thermal stress concentration. This control method can improve welding accuracy, reduce material deformation caused by the expansion of the heat-affected zone, and benefit the overall stability of the seal.

[0099] Based on the depth allocation, the fusion depth of each material layer is calculated. Specifically, this includes calculating the fusion depth of each material layer based on the depth allocation. Using the depth allocation data, a thermal equilibrium model is applied to calculate the fusion depth. For example, the energy input is multiplied by the material's thermal conductivity to obtain the melting front position, such as the fusion depth of the first layer being 0.4 mm and the second layer being 0.8 mm.

[0100] Based on the fusion depth, the distribution of the heat-affected zone (HAZ) of each material layer is determined. Specifically, this includes determining the HAZ distribution of each material layer based on the fusion depth. Based on the fusion depth, a temperature gradient is simulated to delineate the boundaries of the HAZ. For example, if the fusion depth is 0.6 mm, the HAZ extends to a region where the temperature drops to 50% of the melting point at a depth of 1.2 mm.

[0101] Analyzing the overlap of the heat-affected zones (HAZs) between layers ensures there are no stress zones caused by excessive overlap. This method helps identify potential defects and improves welding reliability. The distribution of the HAZ is used to verify the stability of the welding process. The HAZ distribution is compared with a preset threshold. For example, if the distribution range is less than 2.0 mm and there are no concentrated peaks, it is considered stable. If the distribution is uneven, the parameters are adjusted and the simulation is repeated. For example, in the welding of a sealing body, if the deep HAZ is found to be too small, the energy level is increased by 5%, and the stability is confirmed to be improved after verification.

[0102] In the welding scenario of a superhard material seal, assuming a multi-layer structure with three layers: the first layer is a diamond composite material, 0.3 mm thick, with a melting point of 2000℃; the second layer is a tungsten alloy, 0.7 mm thick, with a melting point of 3400℃; and the third layer is a titanium alloy, 1.0 mm thick, with a melting point of 1600℃. When using a layer-by-layer scanning algorithm, the electron beam power is first set to 90 kV based on the correction parameters, with the focal depth increasing layer by layer, from 0.2 mm at the surface to 1.5 mm at the depth. This ensures that the energy depth distribution controls the fusion depth of the shallow layer to 0.25 mm, avoiding overheating. After calculating the fusion depth, it was found that the second layer reached 0.6 mm, meeting the requirements. Then, the heat-affected zone was determined: the first layer is 0.4 mm wide, the second layer is 0.8 mm wide, and the third layer is 1.2 mm wide, with no overlap. When verifying stability, the uniformity of distribution is checked. If the width of the third layer exceeds the threshold of 1.0 mm, it is considered unstable. The scanning speed is adjusted to 80% of the original value. After re-verification, the distribution narrows to 0.9 mm, which improves the stability of the welding process and reduces the risk of interlayer cracks.

[0103] In another scenario, for multi-layered structures with uneven thickness, such as a first layer of 0.4 mm, a second layer of 0.6 mm, and a third layer of 0.8 mm, when controlling the depth distribution, parameters are used to input the electron beam energy in layers: a low energy level of 70 kV for the shallow layer and a high energy level of 110 kV for the deep layer. The calculated fusion depths are 0.3 mm for the first layer, 0.5 mm for the second layer, and 0.7 mm for the third layer. After determining the heat-affected zone distribution, the first layer has a depth of 0.5 mm, the second layer 0.7 mm, and the third layer 0.9 mm. During verification, if the distribution variance is less than 0.2, the structure is considered stable; otherwise, the path is optimized. The results show enhanced stability and a 10% increase in welding efficiency.

[0104] For the verification steps, real-time data is considered. For example, the temperature field is collected during welding. If the heat-affected zone distribution shows that the shallow layer has expanded to 0.6 mm, exceeding the threshold of 0.5 mm, it is determined to be unstable, triggering parameter correction, and ultimately ensuring process stability, which is beneficial to the manufacturing of high-precision seals.

[0105] S7. Adjust the energy input and scanning speed according to the fusion depth and the heat-affected zone distribution to generate the final energy input parameter set.

[0106] In one specific embodiment, the step of adjusting the energy input and scanning speed according to the fusion depth and the heat-affected zone distribution to generate the final energy input parameter set includes:

[0107] (1) If the fusion depth of the deep material is lower than the second preset threshold, an adaptive adjustment algorithm is used to increase the electron beam energy level;

[0108] (2) Optimize the electron beam scanning speed based on the fusion depth;

[0109] (3) Generate the final energy input parameter set based on the increased electron beam energy level and the optimized scanning speed.

[0110] Specifically, if the fusion depth of the deep materials is lower than a second preset threshold, an adaptive adjustment algorithm is needed to increase the electron beam energy level to ensure sufficient melting of the material. This process dynamically adjusts the energy input of the electron beam to meet the requirements of deep melting and prevents poor fusion due to insufficient energy. As the electron beam energy increases, the melting depth gradually improves, ensuring a strong connection between each layer and avoiding poor interlayer bonding. Simultaneously, as a critical factor, the fusion depth needs to be monitored continuously during the welding process for timely adjustments.

[0111] Specifically, after increasing the electron beam energy level, the scanning speed needs to be optimized based on the specific fusion depth. The fusion depth directly affects the scanning path design; therefore, the adjustment of the scanning speed must be closely linked to the fusion depth. Generally, a deeper fusion depth requires a lower scanning speed to ensure sufficient time for heat transfer to the deeper material, achieving a good welding effect. The optimized scanning speed ensures a uniform energy distribution of the electron beam, avoiding uneven energy distribution due to excessive speed, which could cause welding defects or interlayer stress concentration.

[0112] Specifically, the current fusion depth data is collected. By comparing the current fusion depth with a preset threshold (e.g., a threshold of 2.5 mm), if the current depth is 1.8 mm, an adjustment mechanism is triggered. The adaptive adjustment algorithm is based on the proportional-integral-derivative (PID) control principle, calculating the energy level increment. The increment formula is: Increment = Kp * Error + Ki * Integral Error + Kd * Error Change Rate, where Kp, Ki, and Kd are preset coefficients. Iterative adjustments ensure the depth gradually approaches the threshold. (Reference) Figure 4 The figure illustrates the structure of an adaptive PID controller.

[0113] Based on the collected temperature field data, the error value is calculated, which is the threshold minus the current depth. This value is then input into the proportional-integral-derivative controller to generate an initial incremental signal. Combined with the analysis of the thermal conductivity distribution of the material, for example, for deep materials with a thermal conductivity of 150 W / m·K, the increment is adjusted to avoid overheating, generating the final energy level increase value, such as increasing from the initial 10 kW to 12 kW.

[0114] For example, the application of this adaptive adjustment algorithm can respond to the problem of insufficient depth in the welding of multilayer superhard materials in real time. Through the closed-loop feedback of proportional-integral-derivative control, the energy level adjustment process is stabilized, manual intervention is avoided, and welding accuracy is improved. For example, in the case of deep material with a thickness of 3mm, the depth is increased from 1.8mm to 2.6mm after adjustment, reducing the expansion of the heat-affected zone and benefiting the uniform distribution of interlayer stress.

[0115] Based on the fusion depth, the electron beam scanning speed is optimized. Specifically, this includes calculating a speed adjustment value using a linear interpolation method based on the fusion depth data. For example, if the depth is 2.0 mm and the target depth is 2.5 mm, the speed is reduced from the initial 5 mm / s to 4 mm / s to increase the thermal input time. This optimization ensures that the scanning speed matches the depth, enhancing the fusion effect.

[0116] Based on the increased electron beam energy level and the optimized scanning speed, a final energy input parameter set is generated. Specifically, the increased energy level value and the optimized speed are combined into a parameter vector. For example, the parameter set is {energy level: 12kW, speed: 4mm / s}. The total energy input is calculated by weighted summation to ensure that the parameter set covers the requirements of the multilayer structure.

[0117] For example, this generation method integrates the aforementioned adjustment results to form a complete input scheme, which is beneficial to the stability of subsequent welding. For instance, in multilayer materials with large differences in thermal conductivity, the parameter set can control the heat-affected zone within 1.5 mm, thereby improving the overall structural integrity.

[0118] S8. Adjust the energy input and scanning path again according to the final energy input parameter set to obtain the welding results and interlayer stress distribution of the multi-layer structure.

[0119] In one specific embodiment, based on the final energy input parameter set, welding process parameters are further determined to achieve complete welding of multi-layer structures. Optimized parameters such as energy input, scanning speed, and path planning are mapped to specific welding processes. This includes precisely setting welding time and scanning path to ensure that heat input and heat distribution during welding meet material and structural requirements. Through reasonable welding time control, the fusion depth of each layer is guaranteed to reach the expected value, while avoiding overheating or an excessively large heat-affected zone.

[0120] In terms of path planning, the direction and sequence of the scanning path are adjusted based on the characteristics of different layers of materials to ensure that the weld joints between each layer are strong and uniform. The optimized process parameters can also effectively reduce interlayer stress concentration and avoid structural cracks or deformations caused by uneven stress. The adjustments ensure the stability and quality of the welding process, so that the welding results of multi-layer structures meet the design requirements, achieving high strength, good connection and uniform interlayer stress distribution.

[0121] The above are only some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and modifications can be made. Any improvements and modifications made based on the basic principles of the present invention should be considered to fall within the protection scope of the present invention.

[0122] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A welding method for a multi-layer structure of a superhard material sealing body, characterized in that, include: Acquire geometric shape data and material property data of multi-layer structures, and generate three-dimensional digital models of multi-layer structures; Based on the three-dimensional digital model, the heat conduction and stress distribution of the multi-layer structure under different energy inputs are simulated to obtain the heat-affected zone and stress concentration area. Based on the heat-affected zone and the stress concentration region, optimize the energy input parameters to generate an initial energy tuning parameter set; Based on the initial energy tuning parameter set, temperature field data and stress field data during the welding process are collected to obtain the dynamic heat-affected zone variation trend. The energy input and scanning path are adjusted according to the changing trend of the dynamic heat-affected zone to generate the corrected energy tuning parameters. The energy depth distribution in the multilayer structure is controlled by the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution. The energy input and scanning speed are adjusted according to the fusion depth and the heat-affected zone distribution to generate the final energy input parameter set; Based on the final energy input parameter set, the energy input and scanning path are adjusted again to obtain the welding results and interlayer stress distribution of the multi-layer structure.

2. The method as described in claim 1, characterized in that, The process of acquiring geometric shape data and material property data of the multi-layer structure and generating a three-dimensional digital model of the multi-layer structure includes: Obtain data on the thickness, melting point, and thermal conductivity of each layer of material; Using a three-dimensional modeling algorithm, a three-dimensional digital model is generated based on the thickness, melting point, and thermal conductivity data to obtain the thickness distribution, melting point distribution, and thermal conductivity distribution of each layer of material; Based on the three-dimensional digital model, determine the geometric features and material property distribution of each layer of material; A digital representation of the multilayer structure is established based on the geometric features and the distribution of material properties.

3. The method as described in claim 1, characterized in that, The process involves simulating the heat conduction and stress distribution of a multilayer structure under different energy inputs based on the three-dimensional digital model, resulting in the heat-affected zone and stress concentration region, including: The finite element method is used to mesh the three-dimensional digital model; Based on the grid division, the heat conduction process under different electron beam energy levels is simulated to obtain the temperature field distribution of each layer of material; Based on the temperature field distribution, the range of the heat-affected zone of each layer of material is calculated, and the heat-affected zone is determined. Based on the temperature field distribution and material properties, the interlayer stress distribution is simulated to determine the stress concentration region.

4. The method as described in claim 2, characterized in that, The step of optimizing the energy input parameters based on the heat-affected zone and the stress concentration region to generate an initial energy tuning parameter set includes: If the range of the heat-affected zone exceeds the first preset threshold, a genetic algorithm is used to optimize the electron beam energy level parameters. Based on the thickness and melting point of each layer of material, a differentiated energy input scheme is generated; Based on the differentiated energy input scheme, determine the initial energy level tuning parameter set; Based on the initial energy tuning parameter set, temperature field data and stress field data during the welding process are collected to obtain the dynamic heat-affected zone variation trend.

5. The method as described in claim 1, characterized in that, The step of acquiring temperature field data and stress field data during the welding process based on the initial energy tuning parameter set to obtain the dynamic heat-affected zone variation trend includes: The temperature field data is collected by sensors, and the range of the dynamic heat-affected zone is calculated based on the temperature field data. The stress field data is collected by sensors, and the changing trend of the dynamic heat-affected zone is determined based on the stress field data and the temperature field data.

6. The method as described in claim 5, characterized in that, Based on the dynamic heat-affected zone range, and further considering the impact of thermal cycling effects on the welding material properties during temperature changes, temperature fluctuation and thermal cycling simulations are introduced to obtain thermal cycling simulation results, including: The temperature fluctuation characteristics during the welding process are simulated using a preset temperature fluctuation model; The thermal cycling simulation method is used to predict the cycle of multiple heat loading and cooling during welding. The impact of temperature changes on the properties of welding materials is evaluated based on the results of thermal cycling simulations. The properties of the welding materials include the strength, hardness, and toughness of the materials.

7. The method as described in claim 2, characterized in that, The step of adjusting the energy input and scanning path according to the changing trend of the dynamic heat-affected zone to generate corrected energy tuning parameters includes: If the temperature of the shallow material exceeds the melting point threshold, the electron beam energy level is reduced; The scanning path is adjusted according to the changing trend of the dynamic heat-affected zone; Based on the reduced electron beam energy level and the adjusted scanning path, the corrected energy tuning parameters are generated; The energy input scheme during the welding process is optimized using the modified energy tuning parameters.

8. The method as described in claim 3, characterized in that, The step of controlling the depth distribution of energy in the multilayer structure according to the modified energy tuning parameters to obtain the fusion depth and heat-affected zone distribution includes: A layered scanning algorithm is used to control the depth distribution of the electron beam in the multilayer structure according to the corrected energy tuning parameters, and the fusion depth of each layer of material is calculated according to the depth distribution. The distribution of the heat-affected zone of each layer of material is determined based on the fusion depth.

9. The method as described in claim 8, characterized in that, The step of adjusting the energy input and scanning speed according to the fusion depth and the heat-affected zone distribution to generate the final energy input parameter set includes: If the fusion depth of the deep material is lower than the second preset threshold, an adaptive adjustment algorithm is used to increase the electron beam energy level; Optimize the electron beam scanning speed based on the fusion depth; The final energy input parameter set is generated based on the increased electron beam energy level and the optimized scanning speed.

Images