A welding method for high-voltage quick-connect connector
Through real-time monitoring and dynamic adjustment of laser welding technology, the welding quality problems caused by the differences in material and thickness of conductive connecting sheets and conductive spring sheets are solved, and high-strength and high-reliability welding connections are achieved, which significantly improves the stability and quality of the welding process. .
Patent Information
- Application Number
- CN202410895554.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-07-05
AI Technical Summary
In the production and manufacturing of high-voltage fast plug connectors, due to the differences in materials and thicknesses of conductive connecting sheets and conductive shrapnels, the welding quality is difficult to ensure, and it is difficult to take into account the needs of different materials and thicknesses, resulting in insufficient welding strength or deformation.
The heat-affected area is monitored in real time through high-precision temperature sensors, combined with welding deformation and stress concentration analysis model, dynamically adjust the energy gradient and output of the laser beam, and adjust the laser beam focus parameters using the real-time monitoring system of the melt pool to ensure the thermal input and machinery of the welding area. performance.
The stability and reliability of welding quality are improved, welding defects and deformation are reduced, and the strength and reliability of welding joints are improved.
Smart Images

Figure CN118438039B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of information technology, and in particular to a welding method for a high-voltage quick-plug connector. Background Art
[0002] Background: In the manufacturing of high-voltage quick-plug connectors, the conductive connector of the elastic connector needs to be welded to the conductive springs at both ends through a double-sided penetration laser welding process. However, in actual operation, it was found that due to the differences in the material and thickness of the conductive connector and the conductive spring, the absorption and conduction capabilities of the laser energy are different, making it difficult to ensure the welding quality.
[0003] Specifically, the conductive connecting piece is usually made of stainless steel, which is thicker and has a stronger ability to absorb and conduct laser energy. However, the conductive spring is mostly made of copper alloy, which is thinner and has a relatively weaker ability to absorb and conduct laser energy. During double-sided penetration welding, if the laser energy is unevenly distributed, it is easy to cause the conductive connecting piece to overheat, while the conductive spring is underheated, which ultimately leads to insufficient welding strength or welding deformation.
[0004] At the same time, in order to ensure the mechanical and electrical properties of the welding area, the welding deformation needs to be strictly controlled. However, excessive control of welding deformation may limit the adjustment of laser energy gradient, making it difficult to take into account the requirements of different materials and thicknesses during the welding process.
[0005] Therefore, a welding process solution is urgently needed that can ensure welding quality while taking into account the material and thickness differences of the conductive connecting sheet and the conductive spring sheet to achieve precise control of the laser energy gradient. This solution should be able to dynamically adjust the distribution and output of laser energy according to the characteristics of different materials and thicknesses, ensure that the welding area obtains appropriate heat input, and minimize welding deformation and stress concentration, thereby obtaining a high-strength, high-reliability welding connection. Summary of the invention
[0006] The present invention provides a high-voltage quick-plug connector welding method, which mainly includes:
[0007] S1. The size and shape of the heat-affected zone are monitored in real time through high-precision temperature sensors. The collected temperature field data will be input into the welding deformation and stress concentration analysis model, which will predict the deformation trend and stress concentration at the weld according to the changes in the heat-affected zone;
[0008] S2. According to the prediction results of the welding deformation and stress concentration analysis model, the energy gradient of the laser beam is optimized by adjusting the laser power and welding speed, and dynamic adjustments are made to the deformation problems caused by the heat-affected zone;
[0009] S3. Use the real-time molten pool monitoring system to obtain the size and flow state information of the molten pool, and adjust the laser beam focusing parameters in combination with the laser beam quality evaluation data to stabilize the molten pool and reduce welding defects and deformation;
[0010] S4. If the molten pool monitoring results show instability, the real-time welding parameter adjustment system is used to adjust the laser output energy and scanning path to reduce stress concentration and avoid excessive welding deformation;
[0011] S5. After the laser welding process is completed, the mechanical properties of the weld are tested to evaluate the joint strength and reliability, obtain the specific impact of welding parameters on joint performance, and make necessary adjustments accordingly.
[0012] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:
[0013] The present invention discloses a high-precision temperature sensing technology and real-time monitoring system for optimizing the heat affected zone (HAZ), molten pool size and flow state, and welding parameters during laser welding, thereby improving the mechanical properties of the weld and reducing defects and deformation. Specifically, this is achieved through the following steps:
[0014] 1. The present invention deploys high-precision temperature sensors to monitor the size and shape of the heat-affected zone in real time, and the collected temperature field data is input into the welding deformation and stress concentration analysis model. The model predicts the deformation trend and stress concentration at the weld according to the changes in the heat-affected zone. This prediction capability enables the present invention to respond to problems that may arise during the welding process in a timely manner, thereby optimizing the welding process.
[0015] 2. According to the prediction results of the welding deformation and stress concentration analysis model, the present invention optimizes the energy gradient of the laser beam by adjusting the laser power and welding speed, and dynamically adjusts the deformation problem caused by the heat-affected zone. This step helps maintain the stability of the welding process and reduce material damage caused by inappropriate energy input.
[0016] 3. The present invention uses a real-time molten pool monitoring system to obtain information on the size and flow state of the molten pool, and combines it with the laser beam quality assessment data to adjust the laser beam focusing parameters to stabilize the molten pool and reduce welding defects and deformation. This monitoring and adjustment mechanism ensures that the molten pool remains in an ideal state during the welding process, thereby improving the quality of the welded joint.
[0017] 4. If the molten pool monitoring results show instability, the present invention uses a real-time welding parameter adjustment system to adjust the laser output energy and scanning path to reduce stress concentration and avoid excessive welding deformation. This feedback adjustment mechanism further enhances the adaptability and controllability of the welding process.
[0018] 5. After the laser welding process is completed, the mechanical properties test of the weld is used to evaluate the joint strength and reliability, obtain the specific impact of welding parameters on joint performance, and make necessary adjustments accordingly. This evaluation ensures that the welding quality meets the design requirements and provides a reference for future welding operations.
[0019] Overall, the present invention provides a comprehensive optimization method for laser welding technology, which significantly improves the stability of the welding process and the quality of the weld by combining real-time monitoring with dynamic adjustment, reduces production costs and time, and enhances product reliability and performance. These improvements are particularly important for modern manufacturing industries that require high-precision and high-reliability welding. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 The present invention is a flow chart of a high-voltage quick-plug connector welding method.
[0021] Figure 2 It is a schematic diagram of a welding method of a high-voltage quick-plug connector of the present invention.
[0022] Figure 3 It is another schematic diagram of a welding method for a high-voltage quick-plug connector according to the present invention. DETAILED DESCRIPTION
[0023] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.
[0024] like Figure 1-3 In this embodiment, a high-voltage quick-connector welding method may specifically include:
[0025] S1. The size and shape of the heat-affected zone are monitored in real time through high-precision temperature sensors. The collected temperature field data will be input into the welding deformation and stress concentration analysis model, which will predict the deformation trend and stress concentration at the weld according to the changes in the heat-affected zone.
[0026] By arranging a high-precision temperature sensor array during the welding process, the temperature field distribution data of the heat-affected zone is collected in real time. The collected temperature field distribution data is input into the welding thermal-mechanical coupling model based on finite element analysis. The model comprehensively considers the influence of factors such as welding heat input, material thermophysical properties, and welding process parameters on the welding process. The thermoelastic-plastic constitutive equation is used in the model to describe the mechanical behavior of the material at high temperature. By solving the heat conduction equation and the force balance equation, the transient temperature field and displacement field distribution during the welding process are calculated. According to the calculated temperature field distribution, the size and morphological change trend of the heat-affected zone are judged. If the size of the heat-affected zone exceeds the preset threshold or the morphology is abnormal, the early warning mechanism is triggered, indicating that there may be a risk of welding deformation. While obtaining the temperature field distribution, the residual stress distribution at the weld is calculated using the welding stress analysis model. Based on the incremental theory and the small deformation hypothesis, the model uses the thermoelastic-plastic mechanics theory to establish the control equation of welding stress, and the stress distribution cloud map at the weld is obtained by finite element discretization. The stress concentration area is identified by post-processing analysis of the stress distribution cloud map. The stress concentration factor method is used to determine the degree of stress concentration. When the stress concentration factor exceeds the allowable value of the material, it indicates that welding cracks or fatigue failure may occur in the area. The results of heat affected zone analysis and stress concentration analysis are coupled to construct a welding defect risk assessment model. The model comprehensively considers process parameters such as heat input, cooling rate, constraint conditions and material characteristic parameters, and quantitatively evaluates the defect risks caused by welding deformation and stress concentration through fuzzy comprehensive evaluation method to generate a risk level report. According to the risk level report, the welding process parameters are adaptively adjusted, including adjusting the heat input, optimizing the weld bead layout, changing the welding sequence, etc., to reduce the risk of welding deformation and stress concentration. The adjusted process parameters are fed back to the aforementioned thermal-mechanical coupling model and stress analysis model, and the temperature field and stress distribution are recalculated, and the defect risk assessment is performed until the risk level drops to an acceptable range. Through closed-loop feedback control, intelligent early warning and process optimization of welding defects based on temperature field monitoring and stress analysis are realized, which effectively improves the reliability and stability of welding quality.
[0027] Specifically, for example, 25 high-precision temperature sensors are arranged during the welding process, with a sampling frequency of 100Hz, to collect the temperature field distribution data of the welding heat-affected zone. The collected temperature data is input into the finite element welding thermal-mechanical coupling model for calculation. The model contains 1,500,000 units and 1,800,000 nodes. The calculation results show that the maximum temperature at the center of the weld reaches 1,550℃, and the width of the heat-affected zone is 2mm. While obtaining the temperature field distribution, the welding stress analysis model is used to calculate the residual stress distribution at the weld. For example, for low-alloy high-strength steel welding, the maximum vonMises equivalent stress at the weld is 650MPa, and the stress concentration factor is 8, which exceeds the material yield strength (550MPa), and there is a risk of welding cracks and fatigue failure. The results of heat-affected zone analysis and stress concentration analysis are input into the welding defect risk assessment model, and the fuzzy comprehensive evaluation method is used to calculate the welding defect risk level by combining 10 key process parameters and material characteristic parameters that affect welding quality. For example, the defect risk degree under the current process parameters is 85, which belongs to the high risk level. According to the risk assessment results, the welding process parameters were adaptively adjusted, such as reducing the welding current from 150A to 120A and increasing the welding speed from 25cm / min to 30cm / min. After the adjustment, the temperature field and stress distribution were recalculated, the maximum temperature at the center of the weld was reduced to 1,280℃, and the width of the heat-affected zone was reduced to 5mm; the maximum equivalent stress at the weld was reduced to 480MPa, the stress concentration factor was reduced to 1, and the defect risk was reduced to 32, which is a low risk level.
[0028] S2. According to the prediction results of the welding deformation and stress concentration analysis model, the energy gradient of the laser beam is optimized by adjusting the laser power and welding speed, and dynamic adjustments are made to the deformation problems caused by the heat-affected zone.
[0029] Specifically including: S21. According to the prediction results of the welding deformation and stress concentration analysis model, the finite element analysis method is used to perform numerical simulation on the temperature field and stress field in the welding process to obtain the temperature gradient distribution and residual stress distribution in the welding heat affected zone; S22. By analyzing the temperature gradient distribution and residual stress distribution, the high temperature zone and high stress concentration zone in the welding heat affected zone are determined, and it is judged that the high temperature zone and high stress concentration zone are the main influencing factors of welding deformation and stress concentration; S23. According to the position distribution of the high temperature zone and the high stress concentration zone, the particle swarm optimization algorithm is used to optimize the laser power and welding speed, and the optimal laser beam energy gradient distribution is obtained by adjusting the laser power and welding speed, so that the laser beam energy gradient distribution matches the position distribution of the high temperature zone and the high stress concentration zone; S24. The optimized The laser power and welding speed obtained by solving the problem are used as control parameters of the laser welding equipment to dynamically adjust the energy of the laser beam, and to compensate and correct the deformation problem caused by the welding heat affected zone in real time; S25. During the laser welding process, the welding temperature field data is collected in real time by sensors to obtain the actual temperature distribution in the welding heat affected zone; S26. The real-time collected temperature distribution data is compared with the temperature gradient distribution predicted by the finite element analysis method, and the Kalman filter algorithm is used to perform online correction on the finite element analysis model to continuously improve the prediction accuracy of the welding deformation and stress concentration analysis model; S27. According to the prediction results of the corrected welding deformation and stress concentration analysis model, steps S23-S26 are repeated to optimize and adjust the energy gradient of the laser beam again until the welding deformation and stress concentration are effectively controlled.
[0030] Specifically, the finite element analysis method is used to numerically simulate the temperature field and stress field during the welding process. For example, the simulation results show that the maximum temperature of the welding heat affected zone is 1500℃ and the maximum residual stress is 300MPa. By analyzing the temperature gradient distribution and residual stress distribution, it is determined that the temperature of the high temperature zone in the welding heat affected zone exceeds 1000℃, and the stress value of the high stress concentration zone is greater than 200MPa. It is judged that they are the main influencing factors of welding deformation and stress concentration. The particle swarm optimization algorithm is used to optimize the laser power and welding speed. For example, through 100 iterations, the optimal laser power is 2000W and the optimal welding speed is 20mm / s, so that the laser beam energy gradient distribution matches the position distribution of the high temperature zone and the high stress concentration zone. The laser power of 2000W and the welding speed of 20mm / s obtained by optimization are used as the control parameters of the laser welding equipment, and the energy of the laser beam is dynamically adjusted to compensate and correct the deformation caused by the welding heat affected zone in real time. During the laser welding process, the welding temperature field data is collected every 1 second by the sensor to obtain the actual temperature distribution in the welding heat affected zone. The collected temperature distribution data was compared with the temperature gradient distribution predicted by finite element analysis, and the temperature error was within 5%. The Kalman filter algorithm was used to correct the finite element analysis model online. After 10 iterations of correction, the prediction accuracy of the welding deformation and stress concentration analysis model was improved to more than 95%. According to the prediction results of the corrected welding deformation and stress concentration analysis model, steps 3-6 were repeated to optimize the energy gradient of the laser beam again. For example, in the second optimization, the laser power was adjusted to 1800W and the welding speed was adjusted to 22mm / s. The final welding deformation was controlled within 2mm, and the residual stress peak was reduced to below 100MPa, achieving the goal of effectively controlling welding deformation and stress concentration.
[0031] Specifically, it also includes: using the laser power and welding speed obtained through optimization as control parameters of the laser welding equipment, dynamically adjusting the energy of the laser beam, and performing real-time compensation and correction for deformation problems caused by the heat-affected zone.
[0032] According to the laser welding process parameter optimization model, the optimal laser power and welding speed parameters are obtained and used as the control command input of the laser welding equipment. The Kalman filter algorithm is used to filter the laser beam energy to obtain a smooth energy output curve. The motion trajectory of the laser beam deflection mirror is calculated by the inverse kinematics algorithm, and the energy distribution of the laser beam is adjusted in real time. The temperature field distribution data during the welding process is obtained, and the deformation of the heat-affected zone is calculated by the finite element analysis method. According to the deformation amount, the deformation compensation surface equation is fitted by the least squares method. The compensation surface equation is used as input, and the motion compensation amount of the laser beam deflection mirror is solved by the numerical integration algorithm. The motion compensation amount is superimposed with the motion trajectory of the deflection mirror calculated by the inverse kinematics algorithm to obtain the final laser beam deflection mirror motion control instruction. The laser beam deflection mirror motion control instruction is input into the motion control system of the laser welding equipment to dynamically adjust the energy distribution of the laser beam and realize deformation compensation control during the welding process.
[0033] Specifically, - For example, the optimal laser power is calculated to be 1500W and the optimal welding speed is 30mm / s through the laser welding process parameter optimization model. - For example, the Kalman filter algorithm is used to filter a set of laser beam energy data [2J, 5J, 9J, 8J, 3J], and the smoothed energy output is [1J, 3J, 2J, 4J, 3J]. - For example, the inverse kinematics algorithm is used to calculate the motion coordinates of the laser beam deflection mirror at t=1s to be (15mm, 28mm, 13mm). - For example, the finite element analysis shows that the maximum deformation of a welded workpiece in the heat affected zone is 32mm. - For example, the least squares method is used to fit the deformation compensation surface equation z=12x^2+09y^2+03xy-18x-14y+05. - For example, when t=2s, the numerical integration algorithm is used to solve the motion compensation amount of the laser beam deflection mirror to be (28mm,-37mm,15mm). -For example, the above motion compensation amount is superimposed on the deflection mirror motion trajectory (15mm, 28mm, 13mm) calculated by the inverse kinematics algorithm, and the final motion control instruction at t=2s is obtained.
[0034] S3. Use the real-time molten pool monitoring system to obtain the size and flow state information of the molten pool, and adjust the laser beam focusing parameters in combination with the laser beam quality evaluation data to stabilize the molten pool and reduce welding defects and deformation.
[0035] A high-speed camera is used to collect the molten pool image in real time to obtain the size information of the molten pool, including parameters such as the length, width and depth of the molten pool. At the same time, the molten pool image is analyzed by an image processing algorithm, including one of the threshold segmentation algorithm and the edge detection algorithm, to obtain the flow state information of the molten pool, including the turbulence degree and flow velocity distribution of the molten pool. The quality parameters of the laser beam, including the divergence angle, spot diameter and energy distribution of the laser beam, are obtained through the laser beam quality evaluation system. According to the laser beam quality evaluation data, it is judged whether the laser beam meets the welding process requirements. If the laser beam quality does not meet the requirements, the laser needs to be adjusted, including optimizing the working parameters and cooling conditions of the laser, to improve the laser beam quality. The molten pool size and flow state information and the laser beam quality evaluation data are input into the intelligent control algorithm, including the fuzzy control algorithm or the neural network algorithm. The intelligent control algorithm adaptively adjusts the laser beam focusing parameters, including the focus position, spot diameter and energy density, according to the input data, to achieve stable control of the molten pool size and flow state. In the process of adjusting the laser beam focusing parameters, the changes in the molten pool size and flow state are continuously monitored. If the size of the molten pool fluctuates or the flow becomes turbulent, the intelligent control algorithm needs to correct the laser beam focusing parameters in time to suppress the fluctuation and turbulence of the molten pool, thereby reducing the occurrence of welding defects and deformation. During the welding process, the real-time monitoring system continuously collects molten pool images and laser beam quality data, and feeds the data back to the intelligent control algorithm. Based on the feedback data, the intelligent control algorithm dynamically optimizes the laser beam focusing parameters to adapt to changes in the welding process, ensuring that the molten pool is always in a stable state, thereby obtaining high-quality welded joints.
[0036] Specifically, for example, a high-speed camera is used to collect the molten pool image, and the analysis results show that the molten pool length is 8mm, the width is 5mm, and the depth is 2mm. For example, the molten pool image is processed by the threshold segmentation algorithm. When the threshold is set to 150, the molten pool area can be accurately segmented; the molten pool edge is extracted by the Canny edge detection algorithm to obtain the pixel coordinates of the molten pool edge contour. For example, the laser beam quality evaluation system measures that the laser beam divergence angle is 5mrad, the spot diameter is 3mm, and the energy distribution uniformity is 95%. If the welding process requires the divergence angle to be less than 1mrad, the spot diameter to be less than 4mm, and the energy distribution uniformity to be greater than 90%, the current laser beam quality meets the requirements. For example, the molten pool size parameters (length 8mm, width 5mm, depth 2mm) and the laser beam quality parameters (divergence angle 5mrad, spot diameter 3mm) are input into the fuzzy control algorithm. After inference calculation, the output optimized laser beam focus position is 2mm, the spot diameter is 35mm, and the energy density is 106W / cm2. For example, after adjusting the laser beam focusing parameters, 100 molten pool images were continuously monitored, and it was found that the average fluctuation rate of the molten pool size was 5%, and the maximum fluctuation rate was 8%, indicating that the molten pool size was basically stable. By analyzing the flow state of the molten pool, the average turbulence intensity was 8, the maximum value was 2, and the coefficient of variation of the flow velocity distribution was 15, indicating that the molten pool flow was relatively stable. For example, the real-time monitoring system collects molten pool images and laser beam quality data every 50ms, and transmits the data to the intelligent control algorithm. The intelligent algorithm dynamically adjusts the focal position, spot diameter and energy density of the laser every 100ms according to the latest feedback of the molten pool size, flow state and laser beam parameters, and continuously optimizes 100 times to reduce the fluctuation rate of the molten pool size to less than 2% and the turbulence intensity to less than 5, thereby obtaining a high-quality welded joint with good weld formation, dense structure and excellent mechanical properties.
[0037] S4. If the molten pool monitoring results show instability, the real-time welding parameter adjustment system is used to adjust the laser output energy and scanning path to reduce stress concentration and avoid excessive welding deformation.
[0038] The molten pool image signal is collected in real time by the photoelectric sensor, and the image signal is de-noised by the wavelet analysis algorithm to obtain the de-noised molten pool image. According to the de-noised molten pool image, the fuzzy C-means clustering algorithm is used to extract the molten pool morphological features and obtain the geometric parameters such as the width, length and height of the molten pool. The extracted molten pool geometric parameters are input into the support vector machine model for classification to determine whether the current molten pool state is stable. If the classification result shows that the molten pool state is unstable, the real-time welding parameter adjustment of the next step is triggered. According to the classification result of the support vector machine model, the welding parameters are adjusted in real time by the adaptive neural fuzzy inference system. The system takes the molten pool geometric parameters as input and dynamically adjusts the laser output energy and scanning path parameters through fuzzy inference rules. The adjusted laser parameters output by the adaptive neural fuzzy inference system are transmitted to the laser control unit to control the laser output energy and scanning path in real time, reduce welding stress concentration, and avoid excessive welding deformation. The adjusted laser parameters are applied to the subsequent welding process, and the continuous optimization of the molten pool stability is achieved through real-time feedback control. At the same time, the optimized laser parameters are recorded as the initial parameters of the next welding layer to form a closed-loop control.
[0039] Specifically, no English will appear in the label. Here is the generated result: <result>The molten pool image signal is collected in real time by the photoelectric sensor, and the image signal is denoised by the wavelet analysis algorithm to obtain the denoised molten pool image. For example, the signal-to-noise ratio of the original image is 20dB. After wavelet denoising, the signal-to-noise ratio is increased to 35dB, effectively removing the noise interference in the image. According to the denoised molten pool image, the fuzzy C-means clustering algorithm is used to extract the molten pool morphological features and obtain the geometric parameters such as the width, length, and height of the molten pool. For example, through clustering analysis, it can be obtained that the average width of the molten pool is 5mm, the average length is 8mm, and the average height is 2mm. The extracted molten pool geometric parameters are input into the support vector machine model for classification to determine whether the current molten pool state is stable. For example, after training, the support vector machine model has a classification accuracy of more than 95% for the molten pool state. When a set of molten pool parameters (width 7mm, length 1mm, height 5mm) is input, the model determines that the molten pool state is unstable and the parameters need to be adjusted. According to the classification results of the support vector machine model, the adaptive neural fuzzy inference system is used to adjust the welding parameters in real time. For example, when the molten pool is judged to be unstable, the inference system adjusts the laser power from 1000W to 1200W and the scanning speed from 20mm / s to 18mm / s according to the molten pool parameters and the preset fuzzy rules to achieve the purpose of stabilizing the molten pool. The adjusted laser parameters are applied to the subsequent welding process, and the continuous optimization of the molten pool stability is achieved through real-time feedback control. For example, before optimization, the deformation of the weldment is 5mm and the residual stress is 200MPa; after 3 cycles of parameter optimization, the deformation of the weldment is reduced to 2mm, the residual stress is reduced to 120MPa, and the welding quality is significantly improved. At the same time, the optimized laser power (1150W) and scanning speed (19mm / s) are used as the initial parameters of the next welding layer to form a closed-loop control.
[0040] S5. After the laser welding process is completed, the mechanical properties of the weld are tested to evaluate the joint strength and reliability, and the specific impact of welding parameters on joint performance is obtained, and necessary adjustments are made accordingly.
[0041] Specifically including: S51. Use Gaussian process regression algorithm to model the relationship between welding parameters and joint mechanical properties in the laser welding process, and obtain the mapping relationship model between welding parameters and joint mechanical properties through training. S52. According to the mapping relationship between welding parameters and joint mechanical properties obtained by Gaussian process regression model, use Monte Carlo simulation method to generate a large number of welding parameter combinations, and predict the joint mechanical properties corresponding to each parameter combination. S53. Perform statistical analysis on the welding parameter combinations generated by Monte Carlo simulation and the predicted joint mechanical properties to obtain the distribution characteristics of the joint mechanical properties, and determine the mean, variance, quantile and other statistics of the joint mechanical properties. S54. According to the statistical analysis results of the joint mechanical properties, use the Six Sigma design method to determine the reasonable range of welding parameters to ensure that the probability that the joint mechanical properties meet the design requirements reaches the preset confidence level. S55. Within the determined welding parameter range, select representative welding parameter combinations through orthogonal experimental design method, conduct laser welding tests, and obtain weld mechanical properties test data under different parameter combinations. S56. Perform variance analysis on the weld mechanical properties test data obtained from the orthogonal test to obtain the significance of the influence of each welding parameter on the mechanical properties of the joint and determine the main influencing factors. S57. Based on the results of the variance analysis, the response surface method is used to construct a polynomial regression model between the welding parameters and the mechanical properties of the joint to obtain the optimal combination of welding parameters so that the mechanical properties of the joint are optimal. S58. Based on the optimal welding parameter combination, perform laser welding process test verification, test the mechanical properties of the weld, evaluate the strength and reliability of the joint, and compare them with the design requirements. If the weld mechanical properties test results meet the design requirements, the optimal welding parameter combination is determined as the process parameter and applied to actual production; if the test results do not meet the requirements, return to step S57, adjust the polynomial regression model, and re-optimize the welding parameters until the requirements are met.
[0042] Specifically, when Gaussian process regression modeling is performed, it is assumed that the input variables (such as welding speed and power) obey Gaussian distribution, for example, the mean of welding speed is 30 mm / s and the variance is 5. The model hyperparameters are calculated by maximum likelihood estimation and other methods to obtain the mapping relationship model between welding parameters and joint mechanical properties. During Monte Carlo simulation, for example, 10,000 sets of welding parameter combinations are generated, each set of parameters includes welding speed (obeying Gaussian distribution with mean 30 mm / s and variance 5), power (obeying Gaussian distribution with mean 2000 W and variance 50), etc., and the corresponding joint tensile strength is predicted by the Gaussian process regression model. The tensile strength of the joints of 10,000 sets of simulation data is statistically analyzed, for example, the mean is 550 MPa, the standard deviation is 20 MPa, and the 90% quantile is 535 MPa. In Six Sigma design, for example, the lower limit of the tensile strength of the joint is 500MPa, and the confidence level is 973%. According to the statistical results, the welding speed range is 25~35mm / s and the power range is 1900~2100W. In orthogonal experimental design, for example, 3 factors are selected, each with 3 levels, L9 (3^4) orthogonal table, a total of 9 groups of experiments, and 9 groups of weld tensile strength data are measured. In variance analysis, for example, the F value of the welding speed factor is 12, which is greater than the critical value, indicating that the welding speed has a significant effect on the tensile strength. The polynomial regression model constructed by the response surface method, for example, tensile strength = 505 + 5*speed + 18*power-14*speed^2-23*power^2, thus the optimal welding parameter combination is 32mm / s speed and 2050W power, and the predicted tensile strength can reach a maximum value of 580MPa. Five verification tests were carried out under the optimal parameter combination, and the average tensile strength was measured to be 585MPa and the standard deviation was 8MPa, which met the design requirements. Finally, this parameter combination was determined as the process parameter.
[0043] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.< / result>
Claims
1. A high-voltage quick-connector welding method, characterized in that: The method comprises: S1. The size and shape of the heat-affected zone are monitored in real time through high-precision temperature sensors. The collected temperature field data will be input into the welding deformation and stress concentration analysis model, which will predict the deformation trend and stress concentration at the weld according to the changes in the heat-affected zone; By arranging a high-precision temperature sensor array during the welding process, the temperature field distribution data of the heat-affected zone is collected in real time; The collected temperature field distribution data is input into the welding thermal-mechanical coupling model based on finite element analysis. This model comprehensively considers the influence of welding heat input, material thermal properties and welding process parameters on the welding process. The model uses the thermoelastic-plastic constitutive equation to describe the mechanical behavior of the material at high temperature. By solving the heat conduction equation and the force balance equation, the transient temperature field and displacement field distribution during the welding process are calculated. According to the calculated temperature field distribution, the size and morphological change trend of the heat-affected zone are determined; If the size of the heat-affected zone exceeds the preset threshold or the shape is abnormal, the early warning mechanism will be triggered to indicate the possible risk of welding deformation; While obtaining the temperature field distribution, the residual stress distribution at the weld is calculated using the welding stress analysis model; The model is based on the incremental theory and small deformation assumption, and uses the thermoelastic-plastic mechanics theory to establish the control equation of welding stress. The stress distribution cloud diagram at the weld is obtained by finite element discretization. Identify stress concentration areas by post-processing and analyzing stress distribution cloud maps; The stress concentration factor method is used to determine the degree of stress concentration. When the stress concentration factor exceeds the allowable value of the material, it indicates that welding cracks or fatigue failure may occur in the area. The results of heat-affected zone analysis and stress concentration analysis are coupled to construct a welding defect risk assessment model; The model comprehensively considers heat input, cooling rate, constraints and material property parameters, and quantitatively evaluates the defect risks caused by welding deformation and stress concentration through fuzzy comprehensive evaluation method to generate a risk level report; Adaptively adjust welding process parameters according to the risk level report, including adjusting heat input, optimizing weld bead layout and changing at least one process parameter in the welding sequence to reduce the risk of welding deformation and stress concentration; The adjusted process parameters are fed back to the aforementioned thermal-mechanical coupling model and stress analysis model to recalculate the temperature field and stress distribution, and perform defect risk assessment until the risk level is reduced to an acceptable range; Through closed-loop feedback control, intelligent early warning of welding defects and process optimization based on temperature field monitoring and stress analysis are realized, effectively improving the reliability and stability of welding quality; S2. According to the prediction results of the welding deformation and stress concentration analysis model, the energy gradient of the laser beam is optimized by adjusting the laser power and welding speed, and dynamic adjustments are made to the deformation problems caused by the heat-affected zone; S3. Use the real-time molten pool monitoring system to obtain the size and flow state information of the molten pool, and adjust the laser beam focusing parameters in combination with the laser beam quality evaluation data to stabilize the molten pool and reduce welding defects and deformation; S4. If the molten pool monitoring results show instability, the real-time welding parameter adjustment system is used to adjust the laser output energy and scanning path to reduce stress concentration and avoid excessive welding deformation; S5. After the laser welding process is completed, the mechanical properties of the weld are tested to evaluate the joint strength and reliability, obtain the specific impact of welding parameters on joint performance, and make necessary adjustments accordingly.
2. The high-voltage quick-connector welding method according to claim 1, characterized in that: in, In step S2, according to the prediction results of the welding deformation and stress concentration analysis model, the energy gradient of the laser beam is optimized by adjusting the laser power and welding speed, and the deformation problem caused by the heat affected area is dynamically adjusted, including: S21. According to the prediction results of the welding deformation and stress concentration analysis model, the temperature field and stress field in the welding process are numerically simulated by the finite element analysis method to obtain the temperature gradient distribution and residual stress distribution in the welding heat affected zone; S22. By analyzing the temperature gradient distribution and residual stress distribution, the high temperature zone and high stress concentration zone in the welding heat affected zone are determined, and it is determined that the high temperature zone and high stress concentration zone are the main influencing factors of welding deformation and stress concentration; S23. According to the position distribution of the high temperature zone and the high stress concentration zone, the particle swarm optimization algorithm is used to optimize the laser power and the welding speed, and the optimal laser beam energy gradient distribution is obtained by adjusting the laser power and the welding speed, so that the laser beam energy gradient distribution matches the position distribution of the high temperature zone and the high stress concentration zone; S24. The laser power and welding speed obtained by optimization are used as control parameters of the laser welding equipment, the energy of the laser beam is dynamically adjusted, and the deformation problem caused by the welding heat affected zone is compensated and corrected in real time; S25. During the laser welding process, the welding temperature field data is collected in real time by the sensor to obtain the actual temperature distribution in the welding heat affected area; S26. Compare the temperature distribution data collected in real time with the temperature gradient distribution predicted by the finite element analysis method, and use the Kalman filter algorithm to perform online correction on the finite element analysis model to continuously improve the prediction accuracy of the welding deformation and stress concentration analysis model; S27. According to the prediction results of the modified welding deformation and stress concentration analysis model, repeat steps S23-S26 to optimize and adjust the energy gradient of the laser beam again until the welding deformation and stress concentration are effectively controlled; It also includes: using the laser power and welding speed obtained through optimization as control parameters of the laser welding equipment, dynamically adjusting the energy of the laser beam, and performing real-time compensation and correction for deformation problems caused by the heat-affected zone.
3. The high-voltage quick-connector welding method according to claim 1, characterized in that: in, The laser power and welding speed obtained by optimization are used as control parameters of the laser welding equipment to dynamically adjust the energy of the laser beam and compensate and correct the deformation caused by the heat-affected zone in real time, including: The optimal laser power and welding speed parameters are obtained according to the laser welding process parameter optimization model and used as the control command input of the laser welding equipment; The Kalman filter algorithm is used to filter the laser beam energy to obtain a smooth energy output curve; The motion trajectory of the laser beam deflection mirror is calculated through the inverse kinematics algorithm, and the energy distribution of the laser beam is adjusted in real time; Obtain the temperature field distribution data during welding and use the finite element analysis method to calculate the deformation of the heat affected zone; According to the deformation amount, the deformation compensation surface equation is fitted by the least square method; The compensation surface equation is used as input, and the motion compensation of the laser beam deflection mirror is solved by numerical integration algorithm. The motion compensation amount is superimposed on the motion trajectory of the deflection mirror calculated by the inverse kinematics algorithm to obtain the final motion control instruction of the laser beam deflection mirror; The laser beam deflection mirror motion control instructions are input into the motion control system of the laser welding equipment to dynamically adjust the energy distribution of the laser beam and realize deformation compensation control during the welding process.
4. The high-voltage quick-connector welding method according to claim 1, characterized in that: in, In step S3, the size and flow state information of the molten pool are obtained by using the molten pool real-time monitoring system, and the laser beam focusing parameters are adjusted in combination with the laser beam quality evaluation data to stabilize the molten pool and reduce welding defects and deformation, including: Use a high-speed camera to collect molten pool images in real time to obtain molten pool size information, including molten pool length, width and depth; At the same time, an image processing algorithm, including one of a threshold segmentation algorithm and an edge detection algorithm, is used to analyze the molten pool image to obtain flow state information of the molten pool, including the turbulence degree and flow velocity distribution of the molten pool; The laser beam quality evaluation system is used to obtain the quality parameters of the laser beam, including the divergence angle, spot diameter and energy distribution of the laser beam; According to the laser beam quality evaluation data, determine whether the laser beam meets the welding process requirements; If the laser beam quality does not meet the requirements, the laser needs to be adjusted, including optimizing the laser's operating parameters and cooling conditions to improve the laser beam quality; Inputting the molten pool size and flow state information and laser beam quality assessment data into an intelligent control algorithm, including a fuzzy control algorithm or a neural network algorithm; The intelligent control algorithm adaptively adjusts the laser beam focusing parameters, including focal position, spot diameter and energy density, based on the input data to achieve stable control of the molten pool size and flow state; In the process of adjusting the laser beam focusing parameters, the changes in the molten pool size and flow state are continuously monitored; If the size of the molten pool fluctuates or the flow becomes turbulent, the intelligent control algorithm needs to correct the laser beam focusing parameters in time to suppress the fluctuation and turbulence of the molten pool, thereby reducing the occurrence of welding defects and deformation; During the welding process, the real-time monitoring system continuously collects molten pool images and laser beam quality data, and feeds the data back to the intelligent control algorithm; The intelligent control algorithm dynamically optimizes the laser beam focusing parameters based on feedback data to adapt to changes in the welding process and ensure that the molten pool is always in a stable state, thereby obtaining high-quality welded joints.
5. The high-voltage quick-connector welding method according to claim 1, characterized in that: in, In step S4, if the molten pool monitoring result shows that there is instability, the output energy and scanning path of the laser are adjusted by using a real-time welding parameter adjustment system to reduce stress concentration and avoid excessive welding deformation, including: The molten pool image signal is collected in real time by a photoelectric sensor, and the image signal is subjected to noise reduction processing by a wavelet analysis algorithm to obtain a molten pool image after noise reduction; According to the denoised molten pool image, the fuzzy C-means clustering algorithm is used to extract the morphological features of the molten pool and obtain the geometric parameters of the width, length and height of the molten pool. The extracted molten pool geometric parameters are input into the support vector machine model for classification to determine whether the current molten pool state is stable; If the classification result shows that the molten pool state is unstable, the next step of real-time welding parameter adjustment is triggered; According to the classification results of the support vector machine model, an adaptive neuro-fuzzy inference system is used to adjust the welding parameters in real time; The system takes the molten pool geometry parameters as input and dynamically adjusts the laser output energy and scanning path parameters through fuzzy inference rules; The adjusted laser parameters output by the adaptive neural fuzzy inference system are transmitted to the laser control unit to control the output energy and scanning path of the laser in real time, reduce welding stress concentration, and avoid excessive welding deformation; The adjusted laser parameters are applied in the subsequent welding process to achieve continuous optimization of the molten pool stability through real-time feedback control; At the same time, the optimized laser parameters are recorded as the initial parameters of the next welding layer to form a closed-loop control.
6. The high-voltage quick-connector welding method according to claim 1, characterized in that: in, In step S5, after the laser welding process is completed, the strength and reliability of the joint are evaluated by testing the mechanical properties of the weld, the specific influence of the welding parameters on the joint performance is obtained, and necessary adjustments are made accordingly, including: S51. Gaussian process regression algorithm is used to model the relationship between welding parameters and joint mechanical properties during laser welding, and a mapping relationship model between welding parameters and joint mechanical properties is obtained through training; S52. Based on the mapping relationship between welding parameters and joint mechanical properties obtained by Gaussian process regression model, a large number of welding parameter combinations are generated by Monte Carlo simulation method, and the joint mechanical properties corresponding to each parameter combination are predicted; S53. Statistically analyze the welding parameter combinations generated by the Monte Carlo simulation and the predicted joint mechanical properties to obtain the distribution characteristics of the joint mechanical properties and determine the mean, variance and quantile of the joint mechanical properties; S54. Based on the statistical analysis results of the joint mechanical properties, the Six Sigma Design method is used to determine the reasonable range of welding parameters to ensure that the probability that the joint mechanical properties meet the design requirements reaches the preset confidence level; S55. Within the determined welding parameter range, a representative welding parameter combination is selected by orthogonal test design method, and laser welding test is performed to obtain weld mechanical property test data under different parameter combinations; S56. Perform variance analysis on the weld mechanical properties test data obtained from the orthogonal test to obtain the significance of the influence of each welding parameter on the joint mechanical properties and determine the main influencing factors; S57. Based on the results of variance analysis, a polynomial regression model between welding parameters and joint mechanical properties was constructed using response surface methodology to obtain the optimal combination of welding parameters so that the joint mechanical properties can be optimized; S58. Based on the optimal welding parameter combination, conduct laser welding process test verification, test the mechanical properties of the weld, evaluate the joint strength and reliability, and compare with the design requirements; If the weld mechanical properties test results meet the design requirements, the optimal welding parameter combination is determined as the process parameters and applied to actual production; If the test result does not meet the requirements, return to step S57, adjust the polynomial regression model, and re-optimize the welding parameters until the requirements are met.
