A method and system for regulating the interface temperature in laser-arc hybrid welding with an externally applied magnetic field
Through the laser arc composite welding method with external magnetic field, the interface temperature distribution characteristic neural network and fuzzy rules are used to adjust the magnetic field strength, which solves the temperature control problem in the welding of titanium steel different metals and realizes high-quality welding.
Patent Information
- Application Number
- CN202510361737.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-26
- Publication Date
- 2025-07-29
- Estimated Expiration
- 2045-03-26
Smart Images

Figure CN119870715B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of welding processing technology, and in particular to a method and system for controlling the interface temperature of laser arc hybrid welding with an external magnetic field. Background Art
[0002] With the rapid development of new fourth-generation reactors, titanium alloys, with their enhanced radiation and corrosion resistance, while also offering high-temperature creep performance, will rapidly replace low-alloy stainless steel as a new structural material for thick-walled reactor pressure vessels. This will address the comprehensive performance requirements of thick-walled reactor pressure vessels for high-temperature strength, radiation resistance, and fatigue resistance. Titanium alloys and existing stainless steel materials form a new structural system, making dissimilar metal welding between titanium and steel essential.
[0003] Multi-field coupling technology based on an applied magnetic field, laser, and arc significantly enhances the controllability and stability of the welding heat source. Therefore, laser arc hybrid welding with an applied magnetic field can be applied to dissimilar metal welding, such as titanium-steel welding. The quality of titanium-steel dissimilar metal welds is primarily limited by the control of the weld interface temperature. Therefore, it is crucial to develop a method for controlling the interface temperature of laser arc hybrid welding with an applied magnetic field to achieve real-time temperature control of the dissimilar metal weld interface. Summary of the Invention
[0004] In view of this, an embodiment of the present application provides a method and system for controlling the interface temperature of laser arc hybrid welding with an external magnetic field.
[0005] According to the first aspect of the present application, an embodiment of the present application provides a method for controlling the interface temperature of laser arc hybrid welding with an external magnetic field, comprising:
[0006] Obtain the current welding parameters of the welding sample; welding parameters include welding current, welding voltage, laser power, welding sample surface temperature, arc deflection angle and arc width;
[0007] The interface temperature distribution characteristics neural network prediction model is used to process the welding parameters and predict the current temperature distribution characteristics of the welding interface;
[0008] Determining a deviation of the current temperature distribution characteristics of the welding interface based on the current temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface;
[0009] Based on the deviation, the adjustment amount of the magnetic field strength of the induction coil is determined by fuzzy rules;
[0010] Based on the adjustment amount of the magnetic field strength of the induction coil, the magnetic field strength of the induction coil is adjusted to adjust the current temperature of the welding interface.
[0011] Optionally, the steps of constructing the interface temperature distribution characteristic neural network prediction model include:
[0012] Obtain the training welding parameters of the pre-experiment welding specimen during the pre-experiment welding process; wherein the material and thickness of the pre-experiment welding specimen are the same as those of the welding specimen; the pre-experiment welding is a single-factor variable experiment, and during the single-factor variable experiment, the training welding voltage, training welding current, training laser power, training transverse magnetic field strength, and training longitudinal magnetic field strength are changed respectively; the training welding parameters include training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, the internal temperature of the pre-experiment welding specimen, and the surface temperature of the pre-experiment welding specimen;
[0013] Based on the internal temperature of the pre-experiment welding specimen, fit the laser-arc hybrid welding model of the welding specimen to obtain a welding simulation model;
[0014] Perform two-dimensional slicing on the welding simulation model to obtain the corresponding interface distribution cloud map; and perform feature extraction on the interface distribution cloud map to obtain the corresponding training interface temperature distribution features;
[0015] Use the training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, and the surface temperature of the pre-experiment welding specimen as training samples, and use the training interface temperature distribution features as labels to train the neural network to obtain an interface temperature distribution feature neural network prediction model.
[0016] Optionally, the steps for constructing the standard temperature distribution features of the welding interface include:
[0017] Obtain the compound analysis results of the pre-experiment welding specimen after the single-factor variable experiment, and the compound analysis results include the average thickness of the intermetallic compound and the content of the brittle intermetallic compound;
[0018] Select the training interface temperature distribution features of the pre-experiment welding specimen whose compound analysis results meet the requirements as the standard temperature distribution features of the welding interface.
[0019] Optionally, the deviation amount includes the average temperature difference and the difference change rate, the difference and the difference change rate of the X-direction component of the temperature gradient, and the difference and the difference change rate of the Y-direction component of the temperature gradient;
[0020] Based on the deviation amount, determine the adjustment amount of the magnetic field strength of the induction coil through fuzzy rules, including:
[0021] Through the interface temperature distribution fuzzy controller, process the average temperature difference and the difference change rate, the difference and the difference change rate of the X-direction component of the temperature gradient, and the difference and the difference change rate of the Y-direction component of the temperature gradient based on the fuzzy rules to obtain the fuzzy value of the adjustment amount of the transverse magnetic field strength of the induction coil and the fuzzy value of the adjustment amount of the longitudinal magnetic field strength of the induction coil;
[0022] Defuzzify the fuzzy value of the lateral magnetic field intensity adjustment of the induction coil and the fuzzy value of the lateral magnetic field intensity adjustment of the induction coil to obtain the lateral magnetic field intensity adjustment ΔH1 of the induction coil and the lateral magnetic field intensity adjustment ΔH2 of the induction coil.
[0023] Optionally, based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil, including:
[0024] Determine the current adjustment amount of the induction coil based on the adjustment amount of the magnetic field intensity of the induction coil;
[0025] Adjust the current amount of the induction coil based on the current adjustment amount of the induction coil to adjust the magnetic field intensity of the induction coil.
[0026] Optionally, obtain the arc deflection angle and the arc width, including:
[0027] Obtain an arc image;
[0028] Based on the arc image, determine the arc deflection angle and the arc width.
[0029] Optionally, based on the arc image, determine the arc deflection angle and the arc width, including:
[0030] Identify the arc in the arc image to obtain the target area corresponding to the arc;
[0031] Extract the edge of the target area to obtain the arc contour;
[0032] Based on the arc contour, determine the arc deflection angle and the arc width.
[0033] Optionally, obtain the surface temperature of the welding specimen, including:
[0034] Obtain the temperature of the point on the surface of the welding specimen that is perpendicular to the weld seam and is 4 mm - 11 mm away from the welding groove to obtain the surface temperature of the welding specimen.
[0035] Optionally, the material of the welding specimen is a titanium-steel dissimilar material.
[0036] According to the second aspect of the present application, an interface temperature control system for laser-arc hybrid welding with an applied magnetic field provided by an embodiment of the present application includes:
[0037] A welding process monitoring subsystem and a computer control subsystem;
[0038] The welding process monitoring subsystem includes a welding parameter monitoring module, an arc monitoring module, and a temperature monitoring module; the welding parameter monitoring module is used to collect the current welding current, welding voltage, and laser power of the welding specimen; the arc monitoring module is used to collect the current arc image of the welding specimen; the temperature monitoring module is used to monitor the current surface temperature of the welding specimen;
[0039] The computer control subsystem includes an image feature processing module, an interface temperature distribution feature prediction module, and an interface temperature regulation module; the image feature processing module is used to determine the arc deflection angle and arc width based on the arc image; the interface temperature distribution feature prediction module is used to process the welding parameters using the interface temperature distribution feature neural network prediction model to predict the current temperature distribution feature of the welding interface; the welding parameters include welding current, welding voltage, laser power, the surface temperature of the welding specimen, arc deflection angle, and arc width; the interface temperature regulation module is used to determine the deviation of the current temperature distribution feature of the welding interface based on the current temperature distribution feature of the welding interface and the standard temperature distribution feature of the welding interface; based on the deviation, determine the adjustment amount of the magnetic field strength of the induction coil through fuzzy rules; based on the adjustment amount of the magnetic field strength of the induction coil, adjust the magnetic field strength of the induction coil to adjust the current temperature of the welding interface.
[0040] For the method and system for regulating the interface temperature of laser-arc hybrid welding with an applied magnetic field provided in the embodiments of the present application, since the magnetic field can act on the charged particles of the arc plasma, thereby dynamically changing the shape and position of the arc, the shape change can regulate the heat source action intensity, and the position change can affect the transmission direction of the heat source action. Therefore, by obtaining the current welding parameters of the welding specimen; the welding parameters include welding current, welding voltage, laser power, the surface temperature of the welding specimen, arc deflection angle, and arc width; using the interface temperature distribution feature neural network prediction model to process the welding parameters to predict the current temperature distribution feature of the welding interface; based on the current temperature distribution feature of the welding interface and the standard temperature distribution feature of the welding interface, determine the deviation of the current temperature distribution feature of the welding interface; based on the deviation, determine the adjustment amount of the magnetic field strength of the induction coil through fuzzy rules; based on the adjustment amount of the magnetic field strength of the induction coil, adjust the magnetic field strength of the induction coil to adjust the current temperature of the welding interface; thus, by regulating the magnetic field strength of the induction coil in real time, the shape and position of the arc can be changed, the heat source input position and intensity can be controlled, and the real-time regulation of the welding interface temperature can be realized to complete the high-quality welding of dissimilar metals.
[0041] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the following specifically illustrates the specific embodiments of the present application. Brief Description of the Drawings
[0042] Figure 1 It is a schematic flowchart of a method for regulating the interface temperature of laser-arc hybrid welding with an externally applied magnetic field in an embodiment of the present application;
[0043] Figure 2 It is a schematic diagram of the arrangement position of thermocouples in the preliminary experiment in an embodiment of the present application;
[0044] Figure 3 It is a schematic structural diagram of a system for regulating the interface temperature of laser-arc hybrid welding with an externally applied magnetic field in an embodiment of the present application. Detailed Embodiments
[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present application.
[0046] An embodiment of the present application provides a method for regulating the interface temperature of laser-arc hybrid welding with an externally applied magnetic field, as Figure 1 shown, including:
[0047] S101, obtaining the current welding parameters of the welding specimen; the welding parameters include welding current, welding voltage, laser power, surface temperature of the welding specimen, arc deflection angle, and arc width.
[0048] In this embodiment, laser-arc hybrid welding with an externally applied magnetic field is used to weld the welding specimen. When using laser-arc hybrid welding with an externally applied magnetic field to weld the welding specimen, the devices used may include an arc welding machine, a laser, a welding robot, an induction coil for generating a magnetic field, and a current controller for controlling the current flow of the induction coil; it may also include an arc monitoring module, such as a camera, for real-time acquisition of the arc image during the welding process to obtain the arc deflection angle and arc width; it may also include a temperature monitoring module, such as an infrared temperature sensor, for real-time monitoring of the surface temperature distribution of the welding specimen; it may also include a welding parameter monitoring module for real-time acquisition of the welding current and welding voltage of the arc welding machine and the laser power of the laser during the welding process.
[0049] In this embodiment, the welding specimen includes, but is not limited to, dissimilar metals, and the dissimilar metals include, but are not limited to, titanium alloys and stainless steels.
[0050] In some embodiments, obtaining the surface temperature of the welded specimen includes: obtaining the temperature at a point on the surface of the welded specimen that is perpendicular to the weld seam and is 4 mm - 11 mm away from the welding groove, so as to obtain the surface temperature of the welded specimen. Specifically, in implementation, the temperatures at points on the surface of the welded specimen that are perpendicular to the weld seam and are 5 mm and 10 mm away from the welding groove can be obtained as the surface temperature of the welded specimen.
[0051] S102. Process the welding parameters using the interface temperature distribution characteristic neural network prediction model to predict the current temperature distribution characteristics of the welding interface.
[0052] In this embodiment, the current temperature distribution characteristics of the welding interface can characterize the temperature characteristics corresponding to multiple different regions of the welding interface. The current temperature distribution characteristics of the welding interface can include the current average temperatures of multiple regions of the welding interface, the X - component of the temperature gradient, and the Y - component of the temperature gradient.
[0053] In some embodiments, the steps for constructing the interface temperature distribution characteristic neural network prediction model include:
[0054] Obtain the training welding parameters during the welding pre - experiment of the pre - experiment welded specimen; wherein the material and thickness of the pre - experiment welded specimen are the same as those of the welded specimen; the welding pre - experiment is a single - factor variable experiment, and during the single - factor variable experiment, the training welding voltage, training welding current, training laser power, training transverse magnetic field intensity, and training longitudinal magnetic field intensity are changed respectively; the training welding parameters include the training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, the internal temperature of the pre - experiment welded specimen, and the surface temperature of the pre - experiment welded specimen; based on the internal temperature of the pre - experiment welded specimen, fit the laser - arc hybrid welding model of the welded specimen to obtain a welding simulation model; perform two - dimensional slicing on the welding simulation model to obtain the corresponding interface distribution cloud map; and perform feature extraction on the interface distribution cloud map to obtain the corresponding training interface temperature distribution characteristics; use the training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, and the surface temperature of the pre - experiment welded specimen as training samples, and use the training interface temperature distribution characteristics as labels to train the neural network to obtain the interface temperature distribution characteristic neural network prediction model.
[0055] Specifically, the method for establishing the interface temperature distribution characteristic neural network prediction model is as follows:
[0056] S11. Prepare the pre - experiment welded specimen: Prepare multiple pre - experiment welded specimens with the same material and thickness as the welded specimen to be welded, and embed thermocouples that can collect the interface temperature distribution of the pre - experiment welded specimen inside the pre - experiment welded specimen.
[0057] In this example, the thermocouples are arranged as follows: three thermocouples are embedded at equal intervals in the butt joint plates, parallel to the welding groove direction, 5 mm away from the welding groove, and at the same time, the corresponding thermocouples are embedded at equal intervals of 10 mm along the welding direction, as Figure 2 shown, Figure 2 which shows that 4 groups of thermocouples numbered 1-4 are embedded in Ti and Fe respectively at equal intervals of 10 mm along the welding direction, and each group of thermocouples includes 3 thermocouples.
[0058] S12. Single-factor variable welding pre-experiment: Conduct a laser-arc hybrid welding test with an externally applied magnetic field with single-factor variables on the pre-experiment welding specimens. For different pre-experiment welding specimens, change the welding voltage, welding current, laser power, transverse magnetic field intensity H1, and longitudinal magnetic field intensity H2 of the arc welding machine during the welding process for welding; during the welding process, collect the welding current, welding voltage of the arc welding machine, laser power of the laser, and arc images in real time to obtain the corresponding training welding current, training welding voltage, training laser power, and training arc images. Collect the internal temperature of the pre-experiment welding specimens through thermocouples and the surface temperature distribution of the pre-experiment welding specimens through infrared temperature sensors, and then obtain the training arc width and training arc deflection angle by processing the training arc images.
[0059] During the welding pre-experiment in this example, the collection frequencies of the welding current, welding voltage of the arc welding machine, laser power of the laser, and arc images are all 100 Hz, the collection frequency of the internal temperature of the pre-experiment welding specimens by thermocouples is 10 KHz, and the collection frequency of the surface temperature distribution of the pre-experiment welding specimens by infrared temperature sensors is 100 Hz.
[0060] S13. Extract the temperature distribution characteristics of the training interface: First, establish a laser-arc hybrid welding model of the pre-experiment welding specimens, and correct the laser-arc hybrid welding model of the pre-experiment welding specimens through the temperature change values collected by thermocouples to fit the welding simulation model. Perform two-dimensional slicing on the fitted welding simulation model to obtain the interface temperature distribution contour map, and divide the interface temperature distribution contour map into different regions; then extract the characteristics of the interface temperature distribution contour map. The specific method is as follows: Fit the temperature inverse scale formula through the temperature values within the temperature scale output by the contour map and the red intensity value R, green intensity value G, and blue intensity value B of the corresponding points' pixels, and then calculate the temperature values of each pixel point within the interface temperature distribution contour map through the temperature inverse scale formula, and calculate the regional average temperature e, X-component tx of the temperature gradient, and Y-component ty of the temperature gradient within each divided region to obtain the corresponding training interface temperature distribution characteristics.
[0061] It should be noted here that in actual welding, the role of the magnetic field is to change the arc position, thereby changing the heat distribution of the welding heat source. Therefore, in the simulation model established in this application, a welding heat source model is selected to replace the thermal effects of the arc and laser on the material during the welding process. Further, by modifying the welding heat source model, the influence of the change in the heat distribution of the welding heat source under different magnetic field actions on the temperature distribution of the welding interface is replaced.
[0062] S14. Training of the neural network prediction model for the interface temperature distribution characteristics: Using the training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle of the pre-experiment arc welding machine and the surface temperature distribution of the pre-experiment welding specimen as input variables, and using the training interface temperature distribution characteristics of the pre-experiment welding specimen as the output variable, a neural network prediction model for the interface temperature distribution characteristics is obtained through neural network training.
[0063] The method for establishing the neural network prediction model for the interface temperature distribution characteristics in this example is specifically operated as follows:
[0064] 1) Preparation of the sample set: The acquisition frequency of the arc image is 100Hz, the acquisition frequencies of the welding current, welding voltage, and laser power values are 100Hz, and the acquisition frequency of the surface temperature of the pre-experiment welding specimen is 100Hz. Three groups of arc widths, arc deflection angles, welding voltages, welding currents, laser powers within n×0.03s and the surface temperature of the pre-experiment welding specimen and the welding interface temperature distribution characteristics at the n×0.03s form a sample, where n is a positive integer. All samples are combined into a sample set.
[0065] 2) Division of the sample set: Normalize the eigenvalue data in the sample set to make the numerical sizes of different types of eigenvalues consistent and eliminate the numerical value differences between different eigenvalue data. The sample set is divided into a training set and a test set according to a certain ratio.
[0066] 3) Optimization of the topological structure design of the neural network prediction model for the interface temperature distribution characteristics: Adopt 10-fold cross-validation, divide the sample set into 10 parts, randomly select 9 parts of the samples as the training set, and the remaining 1 part of the samples as the test set. Set the selection range of the number of hidden layers and the number of nodes, and select the number of hidden layers and the number of nodes through the grid search algorithm. Determine the optimal number of hidden layers and the number of nodes according to the prediction accuracy of the test set.
[0067] 4) Optimization of the hyperparameters of the neural network prediction model for the interface temperature distribution characteristics: Take the training set, test set, optimal number of hidden layers, and number of nodes divided in the previous step as the preconditions for this step. Set the parameter selection ranges of the weight, threshold, and learning rate, and select the hyperparameters for model training through the improved sparrow algorithm. Determine the optimal weight, threshold, and learning rate according to the prediction accuracy of the test set;
[0068] 5) Establishment of neural network prediction model for interface temperature distribution characteristics: Use the selected number of hidden layers, number of nodes, weights, thresholds, and learning rate to train the entire training set to obtain a neural network prediction model for interface temperature distribution characteristics. Use the sample test set to test the trained neural network prediction model for interface temperature distribution characteristics. After the test is completed, generate a neural network prediction model for interface temperature distribution characteristics.
[0069] S103. Based on the current temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface, determine the deviation of the current temperature distribution characteristics of the welding interface.
[0070] In this embodiment, the standard temperature distribution characteristics of the welding interface characterize the temperature distribution characteristics of the welding interface under standard temperature conditions.
[0071] In this embodiment, the standard temperature distribution characteristics of the welding interface may include the standard average temperature of multiple regions of the welding interface, the X-component of the standard temperature gradient, and the Y-component of the standard temperature gradient.
[0072] In some embodiments, the current temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface can be compared to obtain a deviation. The deviation may include the average temperature difference and the difference change rate, the X-component difference and the difference change rate of the temperature gradient, and the Y-component difference and the difference change rate of the temperature gradient.
[0073] Specifically, when implementing, the region with the largest average temperature difference in the current temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface can be selected as the comparison region; based on the average temperature, the X-component of the temperature gradient, and the Y-component of the temperature gradient corresponding to the comparison region, calculate the deviation of the comparison region. The deviation includes the average temperature difference and the difference change rate Δec of the comparison region, the X-component difference and the difference change rate Δtxc of the temperature gradient, and the Y-component difference and the difference change rate Δtyc of the temperature gradient.
[0074] Among them, the difference change rate Δec = (Δe1 - Δe2) / Δt, the difference change rate Δtxc = (Δtx1 - Δtx2) / Δt, the difference change rate Δtyc = (Δty1 - Δty2) / Δt. Δe2, Δtxc2, and Δty2 are the previous average temperature difference, the previous X-component difference, and the previous Y-component difference respectively. Δe1, Δtxc1, and Δty1 are the current average temperature difference, the current X-component difference, and the current Y-component difference respectively. Δt is the interval time.
[0075] S104. Based on the deviation, determine the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules.
[0076] In this embodiment, the adjustment amount of the magnetic field intensity of the induction coil includes a first adjustment amount corresponding to the transverse magnetic field intensity of the induction coil and a second adjustment amount corresponding to the longitudinal magnetic field intensity of the induction coil.
[0077] In this embodiment, since the magnetic field can act on the charged particles of the arc plasma, thereby dynamically changing the shape and position of the arc, the shape change can regulate the heat source action intensity, and the position change can affect the transmission direction of the heat source action. Therefore, the magnetic field intensity is related to the current temperature distribution characteristics of the welding interface. Thus, based on the deviation amount of the current temperature distribution characteristics of the welding interface, the adjustment amount of the magnetic field intensity of the induction coil can be determined through fuzzy rules.
[0078] S105. Based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface.
[0079] In some embodiments, adjusting the magnetic field intensity of the induction coil based on the adjustment amount of the magnetic field intensity of the induction coil includes:
[0080] Determine the current adjustment amount of the induction coil based on the adjustment amount of the magnetic field intensity of the induction coil; based on the current adjustment amount of the induction coil, adjust the current amount of the induction coil to adjust the magnetic field intensity of the induction coil.
[0081] In this embodiment, the current adjustment amount of the induction coil includes a first current adjustment amount corresponding to the transverse magnetic field of the induction coil and a second current adjustment amount corresponding to the longitudinal magnetic field of the induction coil.
[0082] In this embodiment, the magnetic field intensity of the induction coil is related to the current amount of the induction coil. Therefore, based on the adjustment amount of the magnetic field intensity of the induction coil, the current adjustment amount of the induction coil is determined. Thus, the transverse magnetic field intensity can be adjusted by the current amount when the induction coil generates a transverse magnetic field, and the longitudinal magnetic field intensity can be adjusted by the current amount when the induction coil generates a longitudinal magnetic field.
[0083] In this embodiment, the current adjustment amount of the induction coil can be obtained through the magnetic field intensity formula.
[0084] In this embodiment, the current adjustment amount of the induction coil can be sent to the current controller of the current amount of the induction coil, so that the current controller adjusts the current amount of the induction coil based on the current adjustment amount.
[0085] In this embodiment, based on the current adjustment amount of the induction coil, the current amount of the induction coil is adjusted, so that the transverse magnetic field intensity and the longitudinal magnetic field intensity of the induction coil can be adjusted, and thus the shape and position of the arc can be changed to adjust the current temperature of the welding interface to the standard temperature in real time.
[0086] In this embodiment, the above steps S101 - S105 are repeated throughout the welding process until the welding is completed, achieving real - time regulation of the welding interface temperature.
[0087] The method for regulating the interface temperature of laser - arc hybrid welding with an externally applied magnetic field provided by the embodiment of the present application. Since the magnetic field can act on the charged particles of the arc plasma, the shape and position of the arc can be dynamically changed. The shape change can regulate the heat source action intensity, and the position change can affect the transmission direction of the heat source action. Thus, by obtaining the current welding parameters of the welding specimen; the welding parameters include welding current, welding voltage, laser power, the surface temperature of the welding specimen, arc deflection angle, and arc width; using the interface temperature distribution characteristic neural network prediction model to process the welding parameters, predicting the current temperature distribution characteristics of the welding interface; based on the current temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface, determining the deviation of the current temperature distribution characteristics of the welding interface; based on the deviation, determining the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules; based on the adjustment amount of the magnetic field intensity of the induction coil, adjusting the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface; in this way, by regulating the magnetic field intensity of the induction coil in real - time, the shape and position of the arc can be changed, the heat source input position and intensity can be controlled, the real - time regulation of the welding interface temperature can be achieved, and the high - quality welding of dissimilar metals can be completed.
[0088] In some embodiments, in step S101, obtaining the arc deflection angle and arc width includes:
[0089] Obtaining an arc image; based on the arc image, determining the arc deflection angle and arc width.
[0090] In this embodiment, by analyzing the arc image, the arc deflection angle and arc width can be determined.
[0091] In some embodiments, based on the arc image, determining the arc deflection angle and arc width includes:
[0092] Identifying the arc in the arc image to obtain the target area corresponding to the arc; performing edge extraction on the target area to obtain the arc contour; based on the arc contour, determining the arc deflection angle and arc width.
[0093] Specifically, during implementation, the arc in the arc image can be identified, the region of interest in the arc image can be extracted to obtain the target area corresponding to the arc. Then, the target area is subjected to binary processing, then the arc contour is extracted from the binary - processed target area, and then the characteristic values of the arc contour are extracted to obtain the arc deflection angle and arc width.
[0094] In this embodiment, by extracting the arc profile, the arc deflection angle and the arc width can be quickly and accurately determined.
[0095] In an alternative embodiment, in step S103, the steps for constructing the standard temperature distribution characteristics of the welding interface include:
[0096] Obtain the compound analysis results of the pre-experiment welding specimens after the single-factor variable experiment. The compound analysis results include the average thickness of the intermetallic compound and the content of the brittle intermetallic compound; select the training interface temperature distribution characteristics of the pre-experiment welding specimens whose compound analysis results meet the requirements as the standard temperature distribution characteristics of the welding interface.
[0097] In this embodiment, the standard temperature distribution characteristics of the welding interface can be determined by the single-factor variable welding pre-experiment of the above embodiment. The specific implementation process can be: sampling and analyzing the pre-experiment welding specimens obtained from different pre-experiments, and selecting the interface temperature distribution characteristics corresponding to the pre-experiment welding specimen with the thinnest average intermetallic compound layer thickness and the lowest content of the brittle intermetallic compound as the standard temperature distribution characteristics of the interface; among them, the average thickness of the intermetallic compound layer can be measured by an optical microscope, and the content of the brittle intermetallic compound can be analyzed by X-ray diffractometer technology.
[0098] In some embodiments, the deviation amount includes the average temperature difference and the difference change rate, the difference and the difference change rate of the X-direction component of the temperature gradient, and the difference and the difference change rate of the Y-direction component of the temperature gradient.
[0099] Step S104, based on the deviation amount, determine the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules, including:
[0100] Through the interface temperature distribution fuzzy controller, process the average temperature difference and the difference change rate, the difference and the difference change rate of the X-direction component of the temperature gradient, and the difference and the difference change rate of the Y-direction component of the temperature gradient based on the fuzzy rules to obtain the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil and the fuzzy value of the adjustment amount of the longitudinal magnetic field intensity of the induction coil;
[0101] Perform defuzzification processing on the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil and the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil to obtain the adjustment amount ΔH1 of the transverse magnetic field intensity of the induction coil and the adjustment amount ΔH2 of the transverse magnetic field intensity of the induction coil.
[0102] In this embodiment, the fuzzy rules can be as follows: when the difference is less than the first threshold and the rate of change of the difference is less than the second threshold, ΔH1 and ΔH2 take smaller values; when the difference is less than the first threshold and the rate of change of the difference is greater than the third threshold, ΔH1 takes a larger value and ΔH2 takes a smaller value; when the difference is greater than the first threshold and less than the fourth threshold, and the rate of change of the difference is greater than the second threshold and less than the fifth threshold, ΔH1 and ΔH2 take moderate values; when the difference is greater than the fourth threshold and the rate of change of the difference is less than the second threshold, ΔH1 takes a smaller value and ΔH2 takes a larger value; when the difference is greater than the fourth threshold and the rate of change of the difference is greater than the fifth threshold, ΔH1 and ΔH2 take larger values.
[0103] In this embodiment, the deviation amount can be used as an input quantity and transmitted to the interface temperature distribution fuzzy controller, and a fuzzy value can be obtained through the fuzzy rules; then, after defuzzifying the fuzzy value, the adjustment amounts ΔH1 and ΔH2 of the transverse magnetic field strength H1 and the longitudinal magnetic field strength H2 can be obtained.
[0104] The specific establishment method of the interface temperature distribution fuzzy controller in this example is as follows:
[0105] S31. Establish the fuzzy subsets and universes of discourse for the input and output quantities respectively: Calculate the change magnitude of the change in the temperature distribution characteristics of the welding interface and the standard temperature distribution characteristics of the welding interface caused by the change in the magnetic field strength through preliminary pre-tests. Set the minimum values of the average temperature difference and its change rate, the difference and its change rate of the X-component of the temperature gradient, and the difference and its change rate of the Y-component of the temperature gradient as Δemin, Δecmin, Δtxmin, Δtxcmin, Δtymin, Δtycmin, and set the maximum values of the average temperature difference and its change rate, the difference and its change rate of the X-component of the temperature gradient, and the difference and its change rate of the Y-component of the temperature gradient as Δemax, Δecmax, Δtxmax, Δtxcmax, Δtymax, Δtycmax. Set the universe of discourse of the average temperature difference Δe as {Δemin, Δemin / 2, Δemin / 3, 0, Δemax / 3, Δemax / 2, Δemax}, the universe of discourse of the average temperature difference change rate Δec as {Δecmin, Δecmin / 2, Δecmin / 3, 0, Δecmax / 3, Δecmax / 2, Δecmax}, the universe of discourse of the difference Δtx of the X-component of the temperature gradient as {Δtxmin, Δtxmin / 2, Δtxmin / 3, 0, Δtxmax / 3, Δtxmax / 2, Δtxmax}, the universe of discourse of the difference change rate Δtxc of the X-component of the temperature gradient as {Δtxcmin, Δtxcmin / 2, Δtxcmin / 3, 0, Δtxcmax / 3, Δtxcmax / 2, Δtxcmax}, the universe of discourse of the difference Δty of the Y-component of the temperature gradient as {Δtymin, Δtxmin / 2, Δtymin / 3, 0, Δtymax / 3, Δtymax / 2, Δtymax}, and the universe of discourse of the difference change rate Δtyc of the Y-component of the temperature gradient as {Δtycmin, Δtxmin / 2, Δtycmin / 3, 0, Δtycmax / 3, Δtycmax / 2, Δtycmax}. Establish the corresponding fuzzy subsets {NB, NM, NS, Z0, PS, PM, PB} for the input quantities respectively. Set eH1 and eH2 as the differences between the maximum and minimum values of the transverse magnetic field strength and the longitudinal magnetic field strength in the preliminary experiment. Set the universe of discourse of the adjustment amount ΔH1 of the transverse magnetic field strength as {-eH1, -eH1 / 2, -eH1 / 3, 0, eH1 / 3, eH1 / 2, eH1}, and the universe of discourse of the adjustment amount ΔH2 of the longitudinal magnetic field strength as {-eH2, -eH2 / 2, -eH2 / 3, 0, eH2 / 3, eH2 / 2, eH2}. Establish the corresponding fuzzy subsets {NB, NM, NS, Z0, PS, PM, PB} for the output quantities respectively;
[0106] S32. Set the fuzzy rule criteria and establish a fuzzy rule base: Establish the fuzzy rules for the average temperature, the X-component of the temperature gradient, and the Y-component of the temperature gradient according to the following criteria:
[0107] 1) When the deviation and the rate of change of the deviation are small, take small values for ΔH1 and ΔH2;
[0108] 2) When the deviation is small and the rate of change of the deviation is large, take a large value for ΔH1 and a small value for ΔH2;
[0109] 3) When the deviation and the rate of change of the deviation are medium-sized, take moderate values for ΔH1 and ΔH2;
[0110] 4) When the deviation is large and the rate of change of the deviation is small, take a small value for ΔH1 and a large value for ΔH2;
[0111] 5) When the deviation and the rate of change of the deviation are large, take large values for ΔH1 and ΔH2;
[0112] The above-mentioned "small" means a small negative deviation or a small positive deviation, and "large" means a large negative deviation or a large positive deviation.
[0113] The established fuzzy control rule table is shown in the following table:
[0114]
[0115] Above, NB is the abbreviation of Negative Big (large negative), representing a large negative deviation, NM is the abbreviation of Negative Middle (medium negative), representing a medium negative deviation, NS is the abbreviation of Negative Small (small negative), representing a small negative deviation, Z0 represents no deviation, and PS, PM, PB are the abbreviations of Positive Small (small positive), Positive Middle (medium positive), and Positive Big (large positive) respectively, representing small, medium, and large positive deviations.
[0116] S33. Select the membership function for the fuzzification process of the input and output quantities and the defuzzification method for the output quantity: Use the triangular membership function to determine the membership degree of the elements in the domain of the input and output quantities to the fuzzy variables, and use the weighted average method to achieve the defuzzification of the output quantity.
[0117] The embodiment of the present application also provides a laser-arc hybrid welding interface temperature control system with an external magnetic field, as Figure 3 shown, including:
[0118] A welding process monitoring subsystem 31 and a computer control subsystem 32.
[0119] The welding process monitoring subsystem 31 includes a welding parameter monitoring module 311, an arc monitoring module 312, and a temperature monitoring module 313; the welding parameter monitoring module 311 is used to collect the current welding parameters of the welding specimen, and the welding parameters include welding current, welding voltage, laser power, and the current of the induction coil; the arc monitoring module 312 is used to collect the current arc image of the welding specimen; the temperature monitoring module 313 is used to monitor the current surface temperature of the welding specimen.
[0120] The computer control subsystem 32 includes an image feature processing module 321, an interface temperature distribution feature prediction module 322, and an interface temperature regulation module 323; the image feature processing module 321 is used to determine the arc deflection angle and arc width based on the arc image; the interface temperature distribution feature prediction module 322 is used to process the welding parameters using the interface temperature distribution feature neural network prediction model to predict the current temperature distribution feature of the welding interface; the welding parameters include welding current, welding voltage, laser power, the surface temperature of the welding specimen, arc deflection angle, and arc width; the interface temperature regulation module 323 is used to determine the deviation of the current temperature distribution feature of the welding interface based on the current temperature distribution feature of the welding interface and the standard temperature distribution feature of the welding interface; based on the deviation, determine the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules; based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface.
[0121] In the external magnetic field laser-arc hybrid welding interface temperature regulation system provided by the embodiment of the present application, since the magnetic field can act on the charged particles of the arc plasma to dynamically change the arc shape and position, the shape change can regulate the heat source action intensity, and the position change can affect the heat source action transmission direction. Therefore, by obtaining the current welding parameters of the welding specimen; the welding parameters include welding current, welding voltage, laser power, the surface temperature of the welding specimen, arc deflection angle, and arc width; using the interface temperature distribution feature neural network prediction model to process the welding parameters to predict the current temperature distribution feature of the welding interface; based on the current temperature distribution feature of the welding interface and the standard temperature distribution feature of the welding interface, determine the deviation of the current temperature distribution feature of the welding interface; based on the deviation, determine the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules; the first adjustment amount corresponding to the transverse magnetic field intensity of the induction coil and the second adjustment amount corresponding to the longitudinal magnetic field intensity of the induction coil; based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface; thus, by regulating the magnetic field intensity of the induction coil in real time, the arc shape and position can be changed, the heat source input position and intensity can be controlled, and the real-time regulation of the welding interface temperature can be achieved to complete the high-quality welding of dissimilar metals.
[0122] In an optional embodiment, the image feature processing module is configured to identify an arc in an arc image to obtain a target area corresponding to the arc; extract edges from the target area to obtain an arc contour; and determine an arc deflection angle and an arc width based on the arc contour.
[0123] The program code for implementing the method of the present application can be written in any combination of one or more programming languages. These program codes can be provided to a processor or a controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, so that when the program codes are executed by the processor or the controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.
[0124] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or an LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and the input from the user can be received in any form (including acoustic input, voice input, or tactile input).
[0125] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected with each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.
[0126] A computer system can include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The client-server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, or a server of a distributed system, or a server combined with a blockchain.
[0127] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this application can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved, and no limitations are imposed herein.
[0128] In addition, the terms "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of this application, "a plurality" means two or more, unless otherwise specifically defined.
[0129] As described above, the above are only specific embodiments of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claimed rights.
Claims
1. A method for regulating the interface temperature in laser-arc hybrid welding with an externally applied magnetic field, characterized in that, Comprising: Obtaining the current welding parameters of the welded specimen; the welding parameters include welding current, welding voltage, laser power, surface temperature of the welded specimen, arc deflection angle, and arc width; Processing the welding parameters using the interface temperature distribution characteristic neural network prediction model to predict the current temperature distribution characteristic of the welding interface; Determining the deviation amount of the current temperature distribution characteristic of the welding interface based on the current temperature distribution characteristic of the welding interface and the standard temperature distribution characteristic of the welding interface; Determining the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules based on the deviation amount; Adjusting the magnetic field intensity of the induction coil based on the adjustment amount of the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface.
2. The method according to claim 1, wherein The construction steps of the interface temperature distribution characteristic neural network prediction model include: Obtaining the training welding parameters during the welding pre-experiment of the pre-experiment welded specimen; wherein the material and thickness of the pre-experiment welded specimen are the same as those of the welded specimen; the welding pre-experiment is a single-factor variable experiment, and during the single-factor variable experiment, the training welding voltage, training welding current, training laser power, training transverse magnetic field intensity, and training longitudinal magnetic field intensity are changed respectively; the training welding parameters include training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, the internal temperature of the pre-experiment welded specimen, and the surface temperature of the pre-experiment welded specimen; Fitting the laser-arc hybrid welding model of the welded specimen based on the internal temperature of the pre-experiment welded specimen to obtain a welding simulation model; Performing two-dimensional slicing on the welding simulation model to obtain a corresponding interface distribution cloud map; and extracting features from the interface distribution cloud map to obtain the corresponding training interface temperature distribution characteristic; Using the training welding voltage, training welding current, training laser power, training arc width, training arc deflection angle, and the surface temperature of the pre-experiment welded specimen as training samples, and using the training interface temperature distribution characteristic as a label to train the neural network to obtain the interface temperature distribution characteristic neural network prediction model.
3. The method according to claim 2, wherein The construction steps of the standard temperature distribution characteristic of the welding interface include: Obtaining the compound analysis results of the pre-experiment welded specimen after the single-factor variable experiment, and the compound analysis results include the average thickness of the intermetallic compound and the content of the brittle intermetallic compound; Selecting the training interface temperature distribution characteristic of the pre-experiment welded specimen whose compound analysis results meet the requirements as the standard temperature distribution characteristic of the welding interface.
4. The method according to claim 1, wherein The deviation amount includes the average temperature difference and the difference change rate, the difference and difference change rate of the X-direction component of the temperature gradient, and the difference and difference change rate of the Y-direction component of the temperature gradient; Determining the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules based on the deviation amount, including: Through the interface temperature distribution fuzzy controller, based on fuzzy rules, process the average temperature difference and its rate of change, the difference and its rate of change of the X - component of the temperature gradient, and the difference and its rate of change of the Y - component of the temperature gradient to obtain the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil and the fuzzy value of the adjustment amount of the longitudinal magnetic field intensity of the induction coil; Perform defuzzification processing on the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil and the fuzzy value of the adjustment amount of the transverse magnetic field intensity of the induction coil to obtain the adjustment amount ΔH1 of the transverse magnetic field intensity of the induction coil and the adjustment amount ΔH2 of the transverse magnetic field intensity of the induction coil.
5. The method according to claim 1, wherein Based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil, including: Based on the adjustment amount of the magnetic field intensity of the induction coil, determine the adjustment amount of the current of the induction coil; Based on the adjustment amount of the current of the induction coil, adjust the current amount of the induction coil to adjust the magnetic field intensity of the induction coil.
6. The method according to claim 1, characterized in that, Obtain the arc deflection angle and arc width, including: Obtain the arc image; Based on the arc image, determine the arc deflection angle and arc width.
7. The method according to claim 6, characterized in that, Based on the arc image, determine the arc deflection angle and arc width, including: Identify the arc in the arc image to obtain the target area corresponding to the arc; Extract the edge of the target area to obtain the arc contour; Based on the arc contour, determine the arc deflection angle and arc width.
8. The method according to claim 1, characterized in that Obtain the surface temperature of the welded specimen, including: Obtain the temperature of the point on the surface of the welded specimen perpendicular to the weld seam and at a distance of 4 mm - 11 mm from the welding groove to obtain the surface temperature of the welded specimen.
9. The method according to claim 1, wherein The material of the welded specimen is a titanium - steel dissimilar material.
10. A laser-arc hybrid welding interface temperature control system with an externally applied magnetic field, characterized in that, Including: A welding process monitoring subsystem and a computer control subsystem; The welding process monitoring subsystem includes a welding parameter monitoring module, an arc monitoring module, and a temperature monitoring module; The welding parameter monitoring module is used to collect the current welding current, welding voltage, and laser power of the welded specimen; the arc monitoring module is used to collect the current arc image of the welded specimen; the temperature monitoring module is used to monitor the current surface temperature of the welded specimen; The computer control subsystem includes an image feature processing module, an interface temperature distribution feature prediction module, and an interface temperature regulation module; the image feature processing module is used to determine the arc deflection angle and arc width based on the arc image; the interface temperature distribution feature prediction module is used to process the welding parameters using an interface temperature distribution feature neural network prediction model to predict the current temperature distribution feature of the welding interface; the welding parameters include the welding current, welding voltage, laser power, surface temperature of the welded specimen, arc deflection angle, and arc width; the interface temperature regulation module is used to determine the deviation amount of the current temperature distribution feature of the welding interface based on the current temperature distribution feature of the welding interface and the standard temperature distribution feature of the welding interface; based on the deviation amount, determine the adjustment amount of the magnetic field intensity of the induction coil through fuzzy rules; based on the adjustment amount of the magnetic field intensity of the induction coil, adjust the magnetic field intensity of the induction coil to adjust the current temperature of the welding interface.
Citation Information
Patent Citations
Laser-arc hybrid welding on-line monitoring method
CN106583927A
Intelligent monitoring method for microwave heating temperature field based on on-line learning
CN109287021A