Welding quality detection method and system

Abstract

The invention discloses a welding quality detection method and system. The method comprises the steps that temperature data and speed data are obtained; analyzing according to the temperature data to obtain a temperature gradient; analyzing according to the speed data and the temperature gradient to obtain a cooling rate; judging according to the cooling rate to obtain a cooling rate change trend; an abnormal area is obtained through analysis according to the temperature gradient and the cooling rate change trend; reading according to the abnormal region to obtain a first parameter set; analyzing the temperature gradient and the cooling rate according to the first parameter set to obtain a gradient weight and a rate weight; optimizing the first parameter set according to the gradient weight and the rate shadow weight to obtain a second parameter set; adjusting different workpieces according to the second parameter set to obtain a third process parameter set; training according to the third process parameter set to obtain an algorithm model; and inputting to-be-detected workpiece characteristics and shape information into the algorithm model for analysis to obtain an optimization scheme. According to the method, parameters can be optimized to construct a model, and the welding detection requirements of workpieces of different materials and shapes are met.

Description

Technical Field

[0001] The present invention relates to the technical field of welding detection, and in particular to a welding quality detection method and system. Background Art

[0002] In welding quality inspection, defects with similar characteristics can be classified by analyzing a large amount of welding image data, but in the actual production process, the stability of the welding process is a complex technical issue. The temperature field distribution and welding speed fluctuations during the welding process will cause changes in the temperature gradient and cooling rate inside the welding area, which will in turn affect the solidification process and organizational structure of the weld, and ultimately lead to welding defects.

[0003] In addition, factors such as welding position, welding angle, and wire feed speed during the welding process will also affect the welding quality. How to find the key factors affecting welding quality in the complex welding process parameters and improve the stability of the welding process by optimizing these key factors is a technical problem that needs to be solved urgently. In actual production, factors such as the material, size, and shape of the welding workpiece will also affect the welding quality. The welding performance of different materials varies greatly, and different welding process parameters and welding methods need to be adopted. Changes in the size and shape of the workpiece will also change the heat conduction and cooling conditions during the welding process, thereby affecting the welding quality. How to optimize the welding process parameters for different welding workpieces to ensure the consistency and stability of the welding quality is also a challenging technical problem.

[0004] In one existing technology, the specific implementation process includes: first collecting a large amount of welding image data, and using image processing and machine learning algorithms to extract and analyze the defect features in the image, such as identifying the defect edge through an edge detection algorithm, and using a clustering algorithm to classify defects with similar features. At the same time, the various process parameters in the welding process, such as temperature field distribution, welding speed, welding position, angle, wire feed speed, etc., as well as information such as the material, size, and shape of the welding workpiece are synchronously recorded. Through correlation analysis, the key factors that significantly affect the welding quality are found.

[0005] However, the prior art is unable to meet the welding inspection requirements of workpieces of different materials and shapes because the process parameters are not optimized and adjusted in combination with the material properties and shape information of the workpiece. Summary of the invention

[0006] The present invention provides a welding quality detection method and system to solve the problem in the prior art that the welding quality detection requirements of workpieces of different materials and shapes cannot be met due to the lack of optimization of process parameters and adjustment in combination with workpiece material properties and shape information.

[0007] In a first aspect, to solve the above technical problems, the present invention provides a welding quality detection method, including: Obtaining temperature field distribution data and welding speed fluctuation data; Performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient; Performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; Judging according to the distribution of the cooling rate to obtain a cooling rate change trend; Performing feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; Reading data according to the abnormal area to obtain a first set of process parameters; Performing correlation analysis on the temperature gradient and the cooling rate according to the first set of process parameters to obtain a gradient influence weight and a rate influence weight; Optimizing the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters; Adjusting according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; Using the third set of process parameters as training data, constructing an initial model based on the support vector machine algorithm and training it to obtain a welding parameter prediction model; Inputting the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

[0008] In a feasible implementation manner of the first aspect, the performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient includes: Performing preprocessing on the temperature field distribution data to obtain normalized temperature field distribution data; Using the normalized temperature field distribution data to construct a model for the welding process by the finite element analysis method to obtain a heat conduction model; Solving the heat conduction equation according to the heat conduction model to obtain a temperature gradient.

[0009] In a feasible implementation manner of the first aspect, the performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate includes: Performing preprocessing on the welding speed fluctuation data to obtain normalized speed fluctuation data; Extracting frequencies from the normalized speed fluctuation data to obtain frequency characteristics; Judging the frequency components of the speed fluctuation according to the frequency characteristics to obtain the dominant factor of the speed fluctuation; Perform correlation analysis based on the temperature gradient and the dominant factor of velocity fluctuation to obtain the first mapping relationship model; Map the rate according to the mapping relationship model to obtain the cooling rate.

[0010] In an implementable manner of the first aspect, the feature analysis based on the temperature gradient and the change trend of the cooling rate to obtain the abnormal area includes: Perform feature analysis based on the change trends of the temperature gradient and the cooling rate to obtain dynamic features; wherein, the dynamic features include the temperature gradient change rate and the cooling rate fluctuation amplitude; Make a judgment based on the temperature gradient change rate and the cooling rate fluctuation amplitude; if the temperature gradient change rate exceeds the preset temperature gradient change rate threshold or the cooling rate fluctuation amplitude is greater than the predetermined fluctuation amplitude range, determine that the welding area is an abnormal area at this time.

[0011] In an implementable manner of the first aspect, the correlation analysis of the temperature gradient and the cooling rate according to the first process parameter set to obtain the gradient influence weight and the rate influence weight includes: Perform parameter correlation analysis according to the first process parameter set to obtain a multi-parameter correlation model; Calculate the parameter correlation degree for each parameter in the first process parameter set according to the multi-parameter correlation model to obtain the parameter correlation degree; Perform quantitative analysis according to the parameter correlation degree to obtain the gradient influence weight and the rate influence weight.

[0012] In an implementable manner of the first aspect, the optimization of the first process parameter set according to the gradient influence weight and the rate influence weight to obtain the second process parameter set includes: Construct a model according to the gradient influence weight and the rate influence weight to obtain a second mapping relationship model; According to the second mapping relationship model, with the minimization of the temperature gradient change rate and the minimization of the cooling rate fluctuation amplitude as the optimization objectives, optimize the first process parameter set to obtain the second process parameter set.

[0013] In an implementable manner of the first aspect, the optimization scheme refers to, for a new welding workpiece, predicting the optimal welding parameter combination of the welding position, welding angle and wire feeding speed for different workpieces according to their material, size and shape attributes.

[0014] In a second aspect, the present invention provides a welding quality detection system, including: A data acquisition module for acquiring temperature field distribution data and welding speed fluctuation data; Gradient analysis module, configured to perform gradient analysis based on the temperature field distribution data to obtain a temperature gradient; Rate analysis module, configured to perform rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; Trend judgment module, configured to make a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; Abnormal area determination module, configured to perform feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; Parameter reading module, configured to read data based on the abnormal area to obtain a first set of process parameters; wherein, the first set of process parameters includes: welding position, welding angle, and wire feeding speed; Parameter analysis module, configured to perform correlation analysis on the temperature gradient and the cooling rate based on the welding position, the welding angle, and the wire feeding speed to obtain a gradient influence weight and a rate influence weight; Parameter optimization module, configured to optimize the first set of process parameters based on the gradient influence weight and the rate influence weight to obtain a second set of process parameters; Parameter adjustment module, configured to adjust according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; Model construction module, configured to use the third set of process parameters as training data, construct an initial model based on the support vector machine algorithm and perform training to obtain a welding parameter prediction model; Output module, configured to input the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

[0015] In an implementable manner of the second aspect, the performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient includes: Performing preprocessing on the temperature field distribution data to obtain normalized temperature field distribution data; Constructing a model for the welding process using the finite element analysis method based on the normalized temperature field distribution data to obtain a heat conduction model; Solving the heat conduction equation based on the heat conduction model to obtain a temperature gradient.

[0016] In an implementable manner of the second aspect, the performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate includes: Performing preprocessing on the welding speed fluctuation data to obtain normalized speed fluctuation data; Performing frequency extraction based on the normalized speed fluctuation data to obtain frequency characteristics; Judge the frequency components of the speed fluctuation according to the said frequency characteristics to obtain the dominant factor of the speed fluctuation; Conduct correlation analysis based on the said temperature gradient and the dominant factor of the speed fluctuation to obtain the first mapping relationship model; Map the rate according to the said mapping relationship model to obtain the cooling rate.

[0017] In an implementable manner of the second aspect, the feature analysis based on the said temperature gradient and the change trend of the cooling rate to obtain the abnormal area includes: Conduct feature analysis based on the change trends of the temperature gradient and the cooling rate to obtain dynamic features; wherein, the said dynamic features include the temperature gradient change rate and the cooling rate fluctuation amplitude; Judge according to the said temperature gradient change rate and the cooling rate fluctuation amplitude; if the said temperature gradient change rate exceeds the preset temperature gradient change rate threshold or the cooling rate fluctuation amplitude is greater than the predetermined fluctuation amplitude range, determine that the welding area is an abnormal area at this time.

[0018] In an implementable manner of the second aspect, the correlation analysis of the said temperature gradient and the cooling rate based on the said first process parameter set to obtain the gradient influence weight and the rate influence weight includes: Conduct parameter correlation analysis based on the said first process parameter set to obtain a multi-parameter correlation model; Calculate the correlation degree of each parameter in the said first process parameter set according to the said multi-parameter correlation model to obtain the parameter correlation degree; Conduct quantitative analysis according to the said parameter correlation degree to obtain the gradient influence weight and the rate influence weight.

[0019] In an implementable manner of the second aspect, the optimization of the said first process parameter set according to the said gradient influence weight and the rate influence weight to obtain the second process parameter set includes: Construct a model according to the said gradient influence weight and the rate influence weight to obtain the second mapping relationship model; According to the said second mapping relationship model, with the minimization of the temperature gradient change rate and the minimization of the cooling rate fluctuation amplitude as the optimization objectives, optimize the said first process parameter set to obtain the second process parameter set.

[0020] In an implementable manner of the second aspect, the said optimization scheme refers to, for new welding workpieces, predicting the optimal welding parameter combination of the welding position, welding angle and wire feeding speed for different workpieces according to their material, size and shape attributes.

[0021] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the welding quality detection method described in any one of the above is implemented.

[0022] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the welding quality detection method described in any one of the above.

[0023] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a welding quality detection method, including obtaining temperature field distribution data and welding speed fluctuation data; performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient; performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; making a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; performing feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; reading data according to the abnormal area to obtain a first set of process parameters; performing correlation analysis on the temperature gradient and the cooling rate based on the first set of process parameters to obtain a gradient influence weight and a rate influence weight; optimizing the first set of process parameters based on the gradient influence weight and the rate influence weight to obtain a second set of process parameters; adjusting according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; training according to the third set of process parameters to obtain a welding parameter prediction model; inputting the characteristics and shape information of the workpiece to be detected into the welding parameter prediction model for data analysis to obtain an optimization plan. The present invention obtains the temperature field distribution and speed fluctuation data during the welding process, calculates the temperature gradient and cooling rate by using finite element analysis, and analyzes the grain growth and solute redistribution characteristics during the solidification process. For the abnormal area, the present invention obtains the real-time welding parameter data, establishes a multi-parameter correlation model, and determines the influence weights of each parameter on the temperature gradient and the cooling rate. Based on this, the present invention uses a genetic algorithm to optimize the welding process parameters and adjusts them in combination with the characteristics and shape information of the workpiece material. Finally, the present invention uses a support vector machine algorithm to predict the optimal welding parameter combination for different workpieces, generates an optimization plan and outputs it to the welding control system. The present invention can effectively improve the welding quality, reduce defects, meet the welding requirements of different materials and shaped workpieces, and improve the intelligence and adaptability of the welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic flowchart of the welding quality detection method provided by the first embodiment of the present invention; Figure 2 It is a schematic structural diagram of a welding quality detection system provided by the second embodiment of the present invention. Specific embodiments

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0026] Refer to Figure 1 , the first embodiment of the present invention provides a welding quality detection method, including the following steps: S1. Obtain temperature field distribution data and welding speed fluctuation data; S2. Perform gradient analysis based on the temperature field distribution data to obtain a temperature gradient; S3. Perform rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; S4. Make a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; S5. Perform feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; S6. Read data according to the abnormal area to obtain a first set of process parameters; S7. Perform correlation analysis on the temperature gradient and the cooling rate based on the first set of process parameters to obtain a gradient influence weight and a rate influence weight; S8. Optimize the first set of process parameters based on the gradient influence weight and the rate influence weight to obtain a second set of process parameters; S9. Adjust according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; S10. Use the third set of process parameters as training data, construct an initial model based on the support vector machine algorithm and perform training to obtain a welding parameter prediction model; S11. Input the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

[0027] In step S1, it is necessary to obtain temperature field distribution data and welding speed fluctuation data.

[0028] Optionally, the temperature field distribution and velocity fluctuations during the welding process are crucial for welding quality. By acquiring this data with sensors, in-depth understanding of the dynamic changes in the welding process can be achieved.

[0029] In step S2, gradient analysis is performed based on the temperature field distribution data to obtain the temperature gradient.

[0030] In the above step S2, the gradient analysis based on the temperature field distribution data to obtain the temperature gradient specifically includes the following steps: S21, preprocess the temperature field distribution data to obtain normalized temperature field distribution data; S22, use the finite element analysis method to construct a model for the welding process based on the normalized temperature field distribution data to obtain a heat conduction model; S23, solve the heat conduction equation based on the heat conduction model to obtain the temperature gradient.

[0031] It should be noted that in the above steps S21 to S23, the temperature field distribution data and welding speed fluctuation data during the welding process are acquired through sensors, and the acquired data is preprocessed to remove outliers and noise interference, obtaining normalized temperature field distribution data and welding speed fluctuation data. Based on the preprocessed temperature field distribution data, a heat conduction model of the welding process is established using the finite element analysis method, and by solving the heat conduction equation, the temperature gradient during the welding process is calculated. Exemplarily, data preprocessing is a key step. Median filtering can be used to remove outliers, and wavelet transform can be used to eliminate noise interference, thereby obtaining reliable temperature field and velocity fluctuation data. The finite element analysis method is an effective tool for establishing a welding heat conduction model. By dividing the welded part into a finite number of elements and considering factors such as material properties and boundary conditions, the heat conduction equation can be solved. For example, for the welding process of a steel plate, it can be divided into thousands of tetrahedral elements, set the heat source movement path, and calculate the temperature gradient around the weld.

[0032] S3, perform rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain the cooling rate.

[0033] In the above step S3, the rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain the cooling rate specifically further includes the following steps: S31, preprocess the welding speed fluctuation data to obtain normalized speed fluctuation data; S32, extract the frequency based on the normalized speed fluctuation data to obtain the frequency characteristics; S33, judge the frequency components of the speed fluctuation based on the frequency characteristics to obtain the dominant factor of the speed fluctuation; S34. Perform correlation analysis based on the temperature gradient and the dominant factor of velocity fluctuation to obtain a first mapping relationship model; S35. Map the rate according to the mapping relationship model to obtain a cooling rate.

[0034] It should be noted that in the above steps S31 to S35, based on the preprocessed welding speed fluctuation data, the frequency characteristics of the speed fluctuation are extracted by the time-frequency analysis method, the main frequency components of the speed fluctuation are judged, and the dominant factor of the speed fluctuation is determined. The calculated temperature gradient distribution is correlated with the dominant factor of the speed fluctuation, and a mapping relationship model between the temperature gradient and the speed fluctuation is established through the support vector machine algorithm to realize the mutual prediction between the temperature gradient and the speed fluctuation. Based on the established heat conduction model, by solving the inverse problem of the heat conduction equation, the cooling rate of the welding process is calculated. Exemplarily, the short-time Fourier transform can be used for the time-frequency analysis of the welding speed fluctuation. By performing time-frequency transformation on the speed data, the main frequency components can be extracted. For example, the analysis results may show that there are two main frequencies of 0.5 Hz and 2 Hz, corresponding to the effects of mechanical vibration and power supply fluctuation respectively. The support vector machine algorithm can be used to establish the mapping relationship between the temperature gradient and the speed fluctuation. By selecting an appropriate kernel function, such as the radial basis function, the non-linear relationship between the temperature gradient and the speed fluctuation can be captured. This mapping model can predict the influence of the speed fluctuation on the temperature field or infer the speed fluctuation situation according to the change of the temperature gradient. Solving the inverse problem of the heat conduction equation can obtain the cooling rate distribution. This requires the use of an iterative algorithm, such as the conjugate gradient method, to inversely deduce the heat source parameters and boundary conditions through the known temperature field distribution. For example, for a thick plate welding process, the change curve of the cooling rate along the weld center line with time can be calculated to judge whether there is a hardening risk caused by rapid cooling. Finally, by fusing the information of the temperature gradient, the speed fluctuation and the cooling rate, a comprehensive prediction model can be constructed. The convolutional neural network is an ideal choice for processing this multi-dimensional data. By designing multiple convolutional layers and pooling layers, features can be automatically extracted to realize the dynamic prediction of the welding process. For example, by inputting the temperature field image and speed data at the current moment, the model can predict the temperature distribution and cooling rate at the next moment. This comprehensive analysis method can provide a comprehensive view of the welding process, help optimize the welding parameters, improve the welding quality and efficiency. Through real-time monitoring and prediction, the welding process can be adjusted in time to avoid defect formation and ensure the mechanical properties and reliability of the welded joint.

[0035] S4. Make a judgment based on the distribution of the cooling rate to obtain the change trend of the cooling rate.

[0036] Exemplarily, the cooling rate is calculated by solving the inverse problem of the heat conduction equation, which generally requires the aid of iterative algorithms such as the conjugate gradient method. By using the known temperature field distribution to inversely deduce the heat source parameters and boundary conditions, the change trend of the cooling rate can be judged.

[0037] S5. Perform feature analysis based on the temperature gradient and the change trend of the cooling rate to obtain an abnormal area.

[0038] It should be noted that in the above step S5, the step of performing feature analysis based on the temperature gradient and the change trend of the cooling rate to obtain an abnormal area specifically further includes the following steps: S51. Perform feature analysis based on the temperature gradient and the change trend of the cooling rate to obtain dynamic features; wherein, the dynamic features include the temperature gradient change rate and the cooling rate fluctuation amplitude. S52. Make a judgment based on the temperature gradient change rate and the cooling rate fluctuation amplitude; if the temperature gradient change rate exceeds the preset temperature gradient change rate threshold or the cooling rate fluctuation amplitude is greater than the predetermined fluctuation amplitude range, it is determined that the welding area at this time is an abnormal area.

[0039] Optionally, obtain the real-time temperature data and cooling rate data during the solidification process of the weld, and perform data acquisition and recording according to the preset sampling frequency. Filter the collected temperature data to remove noise interference and obtain a smooth temperature curve. Calculate the temperature gradient change rate based on the temperature curve and judge whether the temperature gradient exceeds the preset threshold range. Perform statistical analysis on the collected cooling rate data, and calculate the fluctuation amplitude and frequency of the cooling rate. If the fluctuation amplitude exceeds the preset range or the frequency is abnormal, it is determined that there is an abnormal fluctuation in the cooling rate. Establish an association model between the temperature gradient, the cooling rate, the grain growth, and the solute redistribution through a machine learning algorithm. Adopt a support vector machine or decision tree algorithm to train the historical data to obtain model parameters. Input the real-time collected temperature gradient change rate and cooling rate fluctuation data into the trained machine learning model to predict the dynamic change trend of the grain size and solute distribution. Judge whether there is an abnormality in the solidification process according to the prediction results of the grain growth and solute redistribution. If the predicted grain size or solute distribution exceeds the preset range, it is determined that the area is a solidification abnormal area.

[0040] Exemplarily, the real-time data acquisition system can record temperature changes at a preset sampling frequency (e.g., 10 times per second). The raw data collected often contains noise and needs to be processed by filtering to obtain a smooth temperature curve. Common filtering methods include the moving average method and low-pass filters, which can effectively remove high-frequency noise. The temperature gradient change rate reflects the severity of weld cooling and has an important impact on grain growth. For example, if the temperature gradient change rate exceeds 100 °C / s·mm, it will lead to coarse grains. By setting a threshold range (e.g., 50 - 80 °C / s·mm), abnormal situations can be detected in a timely manner. The fluctuation of the cooling rate affects the uniformity of the solidification microstructure. Statistical analysis can calculate the standard deviation and main frequency components of the cooling rate. If the standard deviation exceeds a preset value (e.g., 20%) or there are abnormal high-frequency fluctuations (e.g., greater than 1 Hz), it is determined that there is an abnormality. This abnormality stems from unstable cooling medium flow or external interference. Machine learning algorithms can establish an association model between the temperature gradient, cooling rate, and microstructure. Support vector machines are suitable for dealing with non-linear relationships and can map input features to a high-dimensional space through a kernel function (such as the radial basis function). The decision tree algorithm can intuitively reflect the influence weights of various factors. Model training requires a large amount of historical data, including temperature curves, cooling rates, and corresponding metallographic analysis results under different process parameters. Using the trained model, the grain size and solute distribution can be predicted in real time. For example, when the temperature gradient change rate is 60 °C / s·mm and the cooling rate is 15 °C / s, the model may predict an average grain size of 50 μm and a primary α-phase volume fraction of 60%. If the prediction result deviates from the target value (e.g., grain size > 100 μm or primary α-phase < 40%), it is determined as the solidification abnormal area.

[0041] S6. Read data according to the abnormal area to obtain a first set of process parameters.

[0042] Exemplarily, the first set of process parameters may include real-time data during the welding process such as real-time data of the welding position, welding angle, and wire feeding speed.

[0043] S7. Conduct a correlation analysis on the temperature gradient and cooling rate according to the first set of process parameters to obtain the gradient influence weight and the rate influence weight.

[0044] In the above step S7, the step of conducting a correlation analysis on the temperature gradient and cooling rate according to the first set of process parameters to obtain the gradient influence weight and the rate influence weight specifically further includes the following steps: S71. Conduct a parameter correlation analysis according to the first set of process parameters to obtain a multi-parameter correlation model; S72. Calculate the parameter correlation degree for each parameter in the first set of process parameters according to the multi-parameter correlation model to obtain the parameter correlation degree; S73. Perform a quantitative analysis based on the parameter correlation degree to obtain the gradient influence weight and the rate influence weight.

[0045] It should be noted that in the above steps S71 to S73, real-time data of the welding position, welding angle, and wire feeding speed are obtained through sensors and transmitted to the data processing module. Preprocessing is performed on the obtained real-time data, including operations such as data cleaning and normalization, to improve the data quality and usability. Based on the preprocessed data, a multi-parameter correlation model is constructed using the grey correlation analysis method to quantify the correlation degree between each parameter and the temperature gradient and cooling rate. Based on the calculation results of the correlation degree, determine the gradient influence weight and the rate influence weight of the welding position, welding angle, and wire feeding speed.

[0046] Exemplarily, during the welding process, the acquisition and processing of real-time data are crucial for controlling the weld quality. The sensor system may include a laser displacement sensor, an angle sensor, an encoder, etc., which are respectively used to measure the welding position, welding angle, and wire feeding speed. These sensors convert the signals into digital quantities through a high-speed data acquisition card and then transmit them to the data processing module through the industrial Ethernet. Data preprocessing is a key step in improving data quality. Data cleaning can remove outliers, such as sudden data jumps caused by electromagnetic interference. For example, the welding angle data may show instantaneous large fluctuations, which may be caused by sensor failures or external interference. By setting reasonable thresholds, these outliers can be identified and removed. The normalization process can unify parameters with different dimensions to the same scale for subsequent analysis. For example, the welding position may be in millimeters, while the wire feeding speed may be in meters per minute. Through normalization, they can be mapped to the range of 0-1. The grey correlation analysis method is an effective multi-parameter correlation modeling method. It quantifies the correlation strength between parameters by calculating the grey correlation degree between the reference sequence (such as the temperature gradient) and the comparison sequence (such as the welding position). For example, assuming there is a set of welding position data and corresponding temperature gradient data, first perform a dimensionless processing on the two sets of data, and then calculate the difference between the two sequences at each time point. The smaller the difference, the higher the correlation degree. By setting the resolution coefficient (usually taking 0.5), the final correlation degree can be obtained. The determination of the influence weight is of great significance for optimizing welding parameters. For example, if the analysis result shows that the correlation degree between the welding angle and the temperature gradient is 0.85, while the correlation degree with the cooling rate is 0.65, this means that adjusting the welding angle has a more significant effect on controlling the temperature gradient.

[0047] S8. Optimize the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters.

[0048] In the above step S8, optimizing the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters specifically further includes the following steps: S81. Construct a model according to the gradient influence weight and the rate influence weight to obtain a second mapping relationship model; S82. Optimize the first set of process parameters according to the second mapping relationship model with the minimization of the temperature gradient change rate and the minimization of the cooling rate fluctuation amplitude as the optimization objectives to obtain a second set of process parameters.

[0049] It should be noted that in the above steps S81 to S82, a mapping relationship model between welding process parameters and the temperature gradient change rate and the cooling rate fluctuation amplitude is established according to the influence weight. The genetic algorithm is used to encode the first set of process parameters such as the welding position, welding angle, and wire feeding speed with the minimization of the temperature gradient change rate and the minimization of the cooling rate fluctuation amplitude as the optimization objectives to generate an initial population. According to the fitness function, the fitness value of each individual is calculated. The higher the fitness value, the smaller the temperature gradient change rate and the smaller the cooling rate fluctuation amplitude under the combination of the process parameters. Through genetic operations such as selection, crossover, and mutation, the process parameter combination is continuously optimized until the termination condition is met to obtain the optimal set of welding process parameters. Decode the optimal set of welding process parameters to obtain a second set of process parameters composed of the corresponding welding position, welding angle, and wire feeding speed parameter values.

[0050] S9. Adjust according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters.

[0051] Exemplarily, according to the type of the workpiece material, the thermal physical parameters such as the thermal conductivity and specific heat capacity of the material are obtained from the material thermophysical property database. Using three-dimensional modeling software, a three-dimensional geometric model of the workpiece is established according to the size and shape information of the workpiece. Input the obtained material thermophysical parameters into the finite element analysis software and import the three-dimensional geometric model of the workpiece. In the finite element analysis software, set the welding heat source model and boundary conditions, and initialize the second process parameters. Through finite element analysis, simulate the transient temperature field distribution during the welding process to obtain the temperature field nephogram under different welding process parameters. According to the temperature field distribution range and the peak temperature, judge whether the current welding process parameters meet the requirements. If not, adjust the welding speed and heat input, and re-perform the finite element analysis. Through repeated iterative optimization, finally determine the appropriate welding speed range and temperature field distribution range to obtain an optimized third set of process parameters.

[0052] S10, using the third process parameter set as training data, constructing an initial model based on a support vector machine algorithm and performing training to obtain a welding parameter prediction model; inputting the material characteristics and shape information of the workpiece to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

[0053] In one achievable manner, the optimization scheme refers to predicting the optimal welding parameter combination of welding position, welding angle and wire feeding speed for different workpieces according to the material, size and shape properties of the new welding workpiece.

[0054] Exemplarily, a third set of process parameters of workpieces of different materials, sizes and shapes is obtained as a training data set for the support vector machine algorithm. The training data set is preprocessed, key parameters related to welding quality are extracted as feature vectors, and the feature vectors are normalized to normalize features of different dimensions to the [0,1] interval. For example, the wire feed speed (unit: m / min) and thermal conductivity (unit: W / m·K) need to be scaled uniformly. Then, importance analysis is performed using the Pearson correlation coefficient or random forest features, features with low correlation with the output label are eliminated, and the normalized data is divided into a training set and a test set in proportion (e.g., 7:3).

[0055] In the process of constructing the support vector machine model, the kernel function is selected according to the nonlinear degree of the data features. For example, the radial basis function (RBF) is suitable for nonlinear relationships, and the hyperparameters C (penalty coefficient) and gamma (kernel function bandwidth) need to be optimized. Grid Search or Bayesian optimization is used to traverse the hyperparameter combinations, cross-validate (such as 5-fold cross-validation) to evaluate the model performance, and select the parameter combination with the smallest mean square error (MSE) of the validation set. When training the model, the preprocessed training set feature matrix (X_train) and label vector (y_train) are input. The trained SVM regression model can predict welding parameters based on the input features. Through the SVM model, the multi-dimensional process parameters in the third process parameter set are effectively mapped to the material and shape features to achieve accurate prediction of welding parameters. This method significantly improves the adaptability of different workpieces, solves the problem of disconnection between process parameters and workpiece characteristics in the background technology, and ultimately improves the stability of welding quality.

[0056] Exemplarily, for a new welding workpiece, according to its material, size, and shape attributes, it is classified into the corresponding workpiece attribute combination category. The key welding parameters of the new workpiece are extracted and used as a feature vector to be input into the welding parameter prediction model of the corresponding category to predict the optimal welding parameter combination. For example, the physical properties of the workpiece are obtained from a material database or a sensor, including: thermal conductivity (unit: W / m·K), specific heat capacity (J / kg·K), melting point (°C), material type code (e.g., low-carbon steel = 1, aluminum alloy = 2); the shape parameters are extracted through three-dimensional modeling or image processing techniques, including geometric complexity: calculated based on the surface area to volume ratio, and the larger the ratio, the more complex the shape; key dimensions: such as thickness, radius of curvature, weld length; symmetry: the degree of shape symmetry is judged through principal component analysis (PCA) (0-1 normalization).

[0057] Taking a certain welding workpiece as an example, which is an aluminum alloy with a thickness of 8 mm (material code = 2) and has a special-shaped part with a curved surface (geometric complexity = 0.85, radius of curvature = 50 mm), the parameters with different dimensions are normalized to the [0, 1] interval, and the material attributes and shape parameters are combined into an input vector. The input vector is [0.2, 0.8, 0.6, 0.85, 0.3], and the features in the vector respectively correspond to the normalized thermal conductivity, melting point, thickness, geometric complexity, and radius of curvature. Then, the preprocessed feature vector is input into the trained welding parameter prediction model. The model maps the input to a high-dimensional space through a kernel function (such as RBF), finds the optimal hyperplane for regression prediction, and the output of the model is a multi-objective welding parameter combination, including: welding position coordinates (x, y, z), welding angle (°), wire feeding speed (m / min).

[0058] In summary, the present invention discloses a welding quality detection method, including obtaining temperature field distribution data and welding speed fluctuation data; performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient; performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; making a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; performing feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; reading data according to the abnormal area to obtain a first set of process parameters; performing correlation analysis on the temperature gradient and the cooling rate according to the first set of process parameters to obtain a gradient influence weight and a rate influence weight; optimizing the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters; adjusting according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; training according to the third set of process parameters to obtain a welding parameter prediction model; inputting the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimization scheme. The present invention obtains the temperature field distribution and speed fluctuation data during the welding process, calculates the temperature gradient and cooling rate by using finite element analysis, and analyzes the grain growth and solute redistribution characteristics during the solidification process. For the abnormal area, the present invention obtains real-time welding parameter data, establishes a multi-parameter correlation model, and determines the influence weights of each parameter on the temperature gradient and the cooling rate. Based on this, the present invention uses a genetic algorithm to optimize the welding process parameters and adjusts them in combination with the workpiece material characteristics and shape information. Finally, the present invention uses a support vector machine algorithm to predict the optimal welding parameter combination for different workpieces, generates an optimization scheme and outputs it to the welding control system. The present invention can effectively improve the welding quality, reduce defects, meet the welding requirements of different materials and shape workpieces, and improve the intelligence and adaptability of the welding process.

[0059] Referring to Figure 2 , the second embodiment of the present invention provides a welding quality detection system, including: A data acquisition module 101, configured to obtain temperature field distribution data and welding speed fluctuation data; A gradient analysis module 102, configured to perform gradient analysis based on the temperature field distribution data to obtain a temperature gradient; A rate analysis module 103, configured to perform rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; A trend judgment module 104, configured to make a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; An abnormal area determination module 105, configured to perform feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; A parameter reading module 106 is configured to read data according to the abnormal area to obtain a first set of process parameters. The first set of process parameters includes: welding position, welding angle, and wire feeding speed. A parameter analysis module 107 is configured to perform a correlation analysis on the temperature gradient and the cooling rate according to the welding position, the welding angle, and the wire feeding speed to obtain a gradient influence weight and a rate influence weight. A parameter optimization module 108 is configured to optimize the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters. A parameter adjustment module 109 is configured to adjust according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters. A model construction module 110 is configured to use the third set of process parameters as training data, construct an initial model based on the support vector machine algorithm and perform training to obtain a welding parameter prediction model. An output module 111 is configured to input the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

[0060] It should be noted that a welding quality detection system provided in an embodiment of the present invention is used to execute all the process steps of a welding quality detection method in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be described in detail.

[0061] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a welding quality detection program. When the processor executes the computer program, the steps in the above embodiments of various welding quality detection methods are implemented, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above embodiments of each device are implemented, such as the model construction module.

[0062] Exemplarily, the computer program can be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0063] The electronic device can be a computing device such as a desktop computer, notebook, palm computer, and smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0064] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0065] The memory can be used to store the computer program and / or module. The processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.), etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0066] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0067] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative work.

[0068] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A welding quality inspection method, characterized in that, Executed by a computer, including: Obtaining temperature field distribution data and welding speed fluctuation data; Performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient; Performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; Making a judgment based on the distribution of the cooling rate to obtain a cooling rate change trend; Performing feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area; Reading data according to the abnormal area to obtain a first set of process parameters; Performing correlation analysis on the temperature gradient and the cooling rate according to the first set of process parameters to obtain a gradient influence weight and a rate influence weight; Optimizing the first set of process parameters according to the gradient influence weight and the rate influence weight to obtain a second set of process parameters; Adjusting according to the second set of process parameters for different workpiece materials to obtain a third set of process parameters; Using the third set of process parameters as training data, constructing an initial model based on the support vector machine algorithm and training it to obtain a welding parameter prediction model; Inputting the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain an optimal welding parameter combination, thereby generating an optimization plan.

2. The welding quality detection method according to claim 1, characterized in that, The performing gradient analysis based on the temperature field distribution data to obtain a temperature gradient includes: Performing preprocessing on the temperature field distribution data to obtain normalized temperature field distribution data; Using the finite element analysis method to construct a model for the welding process based on the normalized temperature field distribution data to obtain a heat conduction model; Solving the heat conduction equation according to the heat conduction model to obtain a temperature gradient.

3. The welding quality detection method according to claim 1, wherein The performing rate analysis based on the welding speed fluctuation data and the temperature gradient to obtain a cooling rate includes: Performing preprocessing on the welding speed fluctuation data to obtain normalized speed fluctuation data; Performing frequency extraction on the normalized speed fluctuation data to obtain frequency characteristics; Judging the frequency components of the speed fluctuation according to the frequency characteristics to obtain the dominant factor of the speed fluctuation; Performing correlation analysis on the temperature gradient and the dominant factor of the speed fluctuation to obtain a first mapping relationship model; Mapping the rate according to the mapping relationship model to obtain a cooling rate.

4. The welding quality detection method according to claim 1, wherein, The performing feature analysis based on the temperature gradient and the cooling rate change trend to obtain an abnormal area includes: Performing feature analysis on the temperature gradient and the cooling rate change trend to obtain dynamic features; wherein, the dynamic features include the temperature gradient change rate and the cooling rate fluctuation amplitude; Making a judgment according to the temperature gradient change rate and the cooling rate fluctuation amplitude; if the temperature gradient change rate exceeds a preset temperature gradient change rate threshold or the cooling rate fluctuation amplitude is greater than a predetermined fluctuation amplitude range, it is determined that the welding area at this time is an abnormal area.

5. The welding quality detection method according to claim 1, wherein, The performing correlation analysis on the temperature gradient and the cooling rate according to the first set of process parameters to obtain a gradient influence weight and a rate influence weight includes: Performing parameter correlation analysis according to the first set of process parameters to obtain a multi-parameter correlation model; Calculate the correlation degree of each parameter in the first process parameter set according to the multi-parameter correlation model to obtain the parameter correlation degree; Conduct quantitative analysis according to the parameter correlation degree to obtain the gradient influence weight and the rate influence weight.

6. The welding quality detection method according to claim 1, characterized in that Optimizing the first process parameter set according to the gradient influence weight and the rate influence weight to obtain a second process parameter set, including: Construct a model according to the gradient influence weight and the rate influence weight to obtain a second mapping relationship model; According to the second mapping relationship model, taking the minimization of the temperature gradient change rate and the minimization of the cooling rate fluctuation amplitude as the optimization objectives, optimize the first process parameter set to obtain a second process parameter set.

7. The welding quality detection method according to claim 1, characterized in that The optimization scheme refers to, for a new welding workpiece, predicting the optimal welding parameter combination of the welding position, welding angle, and wire feeding speed for different workpieces according to their material, size, and shape attributes.

8. A welding quality detection system, characterized in that, Including: A data acquisition module for acquiring temperature field distribution data and welding speed fluctuation data; A gradient analysis module for performing gradient analysis according to the temperature field distribution data to obtain a temperature gradient; A rate analysis module for performing rate analysis according to the welding speed fluctuation data and the temperature gradient to obtain a cooling rate; A trend judgment module for judging according to the distribution of the cooling rate to obtain the cooling rate change trend; An abnormal area determination module for performing feature analysis according to the temperature gradient and the cooling rate change trend to obtain an abnormal area; A parameter reading module for reading data according to the abnormal area to obtain a first process parameter set; wherein, the first process parameter set includes: welding position, welding angle, and wire feeding speed; A parameter analysis module for performing correlation analysis on the temperature gradient and the cooling rate according to the welding position, the welding angle, and the wire feeding speed to obtain a gradient influence weight and a rate influence weight; A parameter optimization module for optimizing the first process parameter set according to the gradient influence weight and the rate influence weight to obtain a second process parameter set; A parameter adjustment module for adjusting according to the second process parameter set for different workpiece materials to obtain a third process parameter set; A model construction module for using the third process parameter set as training data, constructing an initial model based on the support vector machine algorithm and training it to obtain a welding parameter prediction model; An output module for inputting the workpiece material characteristics and shape information to be detected into the welding parameter prediction model for data analysis to obtain the optimal welding parameter combination, thereby generating an optimization scheme.

9. An electronic device, characterized in that, Including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the welding quality detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the welding quality detection method according to any one of claims 1 to 7.

Images