Adaptive control method and system for laser-arc hybrid welding process parameters of medium-thickness plate high-strength steel

Abstract

The invention belongs to the technical field of intelligent welding, and discloses an adaptive control method for laser-arc hybrid welding process parameters of high-strength steel of a medium-thickness plate, which comprises the following steps of: measuring by using a line structured light camera through a gap measuring module to obtain the butt joint gap size of a steel structural member, and sending the measured data to a process parameter calculating module; the data communication module is used for processing and sending system data, collecting data of the sensing module and sending the data to the calculation module, and meanwhile collecting data of the calculation module and sending the data to the execution module. And a process parameter calculation module is used for processing the gap size and obtaining a process parameter target value, establishing a relationship according to data obtained by a BPNN model, realizing conversion from the gap size to the process parameter, and sending the data to a data bus. The welding execution module is used for physically achieving the welding process and comprises a laser, a welding gun and a mechanical arm fixedly connected with the welding gun.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent welding, and in particular relates to a method and system for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel. Background Art

[0002] Laser-arc hybrid welding is a welding technology that uses a high-energy-density laser beam as a heat source coupled with an arc heat source. The energy synergy formed by the laser and the arc effectively enhances the thermal efficiency of welding, resulting in a large welding penetration and high welding efficiency. Therefore, laser-arc hybrid welding is widely used in welding medium and thick steel plates. Although laser-arc hybrid welding technology has obvious advantages, there are still some technical difficulties in practical applications. Among them, the more prominent problem is that welding requires high assembly precision. The welding system is more sensitive to the butt gap, and the actual welding process often causes butt gaps in the welded parts due to errors such as manufacturing errors and installation errors, which ultimately leads to a decline in welding quality.

[0003] Under defined process parameters, the butt gap during the welding of medium-thick steel plates directly impacts weld quality. Therefore, process exploration can be used to determine optimal parameters for varying butt gaps. However, in actual welding, the butt gap is often unstable, requiring adaptive, real-time control based on welding conditions to achieve optimal joint quality.

[0004] Currently, there is little research on the adaptive control of process parameters for laser-arc hybrid welding of medium and thick plate high-strength steel, and most of the research focuses on thin plate and gas laser welding. Single-pass full penetration welding of medium and thick plate steel structures is difficult to achieve, and the forming laws and defect suppression mechanisms are complex. At the same time, there are many process parameters for laser-arc hybrid welding, and the influence of the forming process under variable gap conditions requires more in-depth research. In terms of control methods, the control methods used in current research are mostly relatively simple, with a single controlled parameter, mostly linear fitting or manually set parameters, and there is a lack of a control method that can adaptively adjust and control in real time.

[0005] Through the above analysis, the problems and defects of the existing technology are as follows:

[0006] (1) Currently, there is little research on the adaptive control of process parameters for laser-arc hybrid welding of medium and thick plate high-strength steel, and most of the research focuses on thin plate and gas laser welding. Single-pass full penetration welding of medium and thick plate steel structures is difficult to achieve, and the forming rules and defect suppression mechanisms are complex.

[0007] (2) At the same time, there are many parameters in the laser-arc hybrid welding process, and the influence of the forming process under the variable gap working condition needs to be further explored. In terms of control methods, the control methods used in current research are mostly relatively simple, with a single controlled parameter, mostly linear fitting or manually set parameters, and lack a control method that can adaptively adjust in real time. Summary of the Invention

[0008] In view of the problems existing in the prior art, the present invention provides a method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel.

[0009] The present invention is implemented as follows: a method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel comprises:

[0010] S1, divide the gap into δ0, δ1, δ2, ..., δ M , gradually increasing from zero butt gap, and determining δ through process exploration i The welding process parameters with good formation under the gap of 100 mm / s are used to construct the "δ-process parameter" relationship table;

[0011] S2, constructing a BPNN model based on the process parameters obtained in S1, and deploying the BPNN model on a computer as a process parameter calculation module;

[0012] S3 uses a line structured light camera to detect the butt gap size in real time, and obtains the welding process parameters required under the current gap size in real time based on the BPNN model constructed in S2, so as to realize real-time control of the welding process parameters under variable gap conditions and obtain good weld formation.

[0013] Furthermore, the process parameters involved in S1 mainly include laser power p and arc length correction Ac.

[0014] Furthermore, the S1 includes the following sub-steps:

[0015] S11, the steel plates to be welded are spaced at a gap of δ i Install it on the welding test bench and preliminarily determine the gap δ based on the existing welding experience parameters. i The initial parameter p under i0 and Ac i0 ;

[0016] S12, after obtaining the test results, adjust the parameters according to the weld formation to obtain the gap δ i The next series of parameters p i1 、Ac i1 、p i2 、Ac i2 the formation of ...;

[0017] S13, by comparing the quality of each weld, select the best welding process parameters and complete the δ i Exploration of technology;

[0018] S14, repeating the process S11-S13, and finally obtaining the process parameters under each gap; thereby constructing a "δ-process parameter" relationship table.

[0019] Furthermore, in step S2, the BPNN structure input layer parameter is the docking gap δ i , the output layer can predict the laser power p i and arc length correction Ac i .

[0020] Furthermore, step S3 includes the following sub-steps:

[0021] S31, relying on the line structured light camera, detects the butt gap size in real time during the welding process and sends the measured data to the constructed BPNN model in real time;

[0022] S32, the BPNN model calculates the required process parameters of the corresponding welding process parameters in real time according to the measured gap size and transmits them to the welding execution equipment through the data communication module, thereby realizing real-time control of the welding process parameters and achieving good weld formation under variable gap conditions.

[0023] Another object of the present invention is to provide an adaptive control system for process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel, comprising:

[0024] The gap measurement module uses a line structured light camera to measure the gap size of the steel structure parts and sends the measurement data to the process parameter calculation module.

[0025] The data communication module is used to process and send system data, collect data from the perception module and send it to the calculation module, and can also collect data from the calculation module and send it to the execution module.

[0026] The process parameter calculation module is used to process the gap size and obtain the target value of the process parameter. It establishes a relationship based on the data obtained by the BPNN model, realizes the conversion of the gap size to the process parameter, and sends the data to the data bus.

[0027] The welding execution module is used to physically realize the welding process, including a laser, a welding gun and a robotic arm fixed to it.

[0028] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel.

[0029] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel.

[0030] Another object of the present invention is to provide an information data processing terminal, which is used to implement an adaptive control system for process parameters of the laser-arc hybrid welding of medium and thick plate high-strength steel.

[0031] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0032] The present invention uses a gap measurement module to measure the gap size of steel structure parts using a line structured light camera, and sends the measured data to the process parameter calculation module. The data communication module is used to process and send system data, collect perception module data and send it to the calculation module, and can also collect calculation module data and send it to the execution module. The process parameter calculation module is used to process the gap size and obtain the target value of the process parameter, establish a relationship based on the data obtained by the BPNN model, realize the conversion of gap size to process parameter, and send the data to the data bus. The welding execution module is used to physically realize the welding process, including a laser, a welding gun and a robotic arm fixedly connected thereto.

[0033] This paper addresses the challenge of achieving single-pass full penetration welding for medium-thick plate steel under variable-gap conditions with laser-arc hybrid welding. It proposes a method and system for adaptively controlling process parameters for laser-arc hybrid welding of medium-thick plate high-strength steel. The proposed process parameter adjustment strategy and adaptive control method achieve excellent single-pass full penetration welding for medium-thick plate steel under variable-gap conditions, providing a new technical approach for high-quality and efficient butt welding of medium-thick plate steel. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel provided by an embodiment of the present invention;

[0035] Figure 2 This is a structural block diagram of a process parameter adaptive control system for laser-arc hybrid welding of medium and thick plate high-strength steel provided by an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of a laser arc hybrid welding platform provided by an embodiment of the present invention;

[0037] Figure 4 is a schematic diagram of a neural network structure provided by an embodiment of the present invention;

[0038] Figure 5 This is a diagram of the BPNN model training results provided by an embodiment of the present invention;

[0039] Figure 6 This is a schematic diagram of a clamping example of welding provided by an embodiment of the present invention;

[0040] Figure 7 This is a diagram showing scanning results and extraction results of a welding example provided by an embodiment of the present invention;

[0041] Figure 8 It is a welding forming result diagram of a welding example provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel, comprising the following steps:

[0044] S1, divide the gap into δ0, δ1, δ2, ..., δ M , gradually increasing from zero butt gap, and determining δ through process exploration i The welding process parameters with good formation under the gap of 100 mm / s are used to construct the "δ-process parameter" relationship table;

[0045] S2, constructing a BPNN model based on the process parameters obtained in S1, and deploying the BPNN model on a computer as a process parameter calculation module;

[0046] S3 uses a line structured light camera to detect the butt gap size in real time, and obtains the welding process parameters required under the current gap size in real time based on the BPNN model constructed in S2, so as to realize real-time control of the welding process parameters under variable gap conditions and obtain good weld formation.

[0047] The process parameters involved in S1 provided in the embodiment of the present invention mainly include laser power p and arc length correction Ac.

[0048] S1 provided in this embodiment of the present invention includes the following sub-steps:

[0049] S11, the steel plates to be welded are spaced at a gap of δ i Install it on the welding test bench and preliminarily determine the gap δ based on the existing welding experience parameters. i The initial parameter p under i0 and Ac i0 ;

[0050] S12, after obtaining the test results, adjust the parameters according to the weld formation to obtain the gap δ i The next series of parameters p i1 、Ac i1 、p i2 、Ac i2 the formation of ...;

[0051] S13, by comparing the quality of each weld, select the best welding process parameters and complete the δ i Exploration of technology;

[0052] S14, repeating the process S11-S13, and finally obtaining the process parameters under each gap; thereby constructing a "δ-process parameter" relationship table.

[0053] In step S2 provided by the embodiment of the present invention, the BPNN structure input layer parameter is the docking gap δ i , the output layer can predict the laser power p i and arc length correction Ac i .

[0054] Step S3 provided in this embodiment of the present invention includes the following sub-steps:

[0055] S31, relying on the line structured light camera, detects the butt gap size in real time during the welding process and sends the measured data to the constructed BPNN model in real time;

[0056] S32, the BPNN model calculates the required process parameters of the corresponding welding process parameters in real time according to the measured gap size and transmits them to the welding execution equipment through the data communication module, thereby realizing real-time control of the welding process parameters and achieving good weld formation under variable gap conditions.

[0057] like Figure 2 As shown, an embodiment of the present invention provides an adaptive control system for process parameters of medium and thick plate high-strength steel laser-arc hybrid welding, comprising:

[0058] The gap measurement module uses a line structured light camera to measure the gap size of the steel structure parts and sends the measurement data to the process parameter calculation module.

[0059] The data communication module is used to process and send system data, collect data from the perception module and send it to the calculation module, and can also collect data from the calculation module and send it to the execution module.

[0060] The process parameter calculation module is used to process the gap size and obtain the target value of the process parameter. It establishes a relationship based on the data obtained by the BPNN model, realizes the conversion of the gap size to the process parameter, and sends the data to the data bus.

[0061] The welding execution module is used to physically realize the welding process, including a laser, a welding gun and a robotic arm fixed to it.

[0062] Existing laser-arc hybrid welding technology for medium and thick plate high-strength steel often has difficulty achieving real-time and accurate matching of welding parameters when faced with complex weld geometry and changes in the mechanical properties of the base material, resulting in large fluctuations in weld formability and mechanical properties. Traditional parameter settings rely heavily on experience or offline testing, and are unable to adapt to rapid changes in factors such as weld gap, plate thickness, and alloy composition. As a result, defects such as insufficient penetration, weld offset, and slag inclusion often occur during the welding process, seriously restricting the reliability and efficiency of high-strength steel plates in the fields of bridges, ships, and large structures. With the increasing application of laser-arc hybrid welding technology, how to dynamically obtain key information at the welding site and implement adaptive control based on intelligent algorithms has become a technical bottleneck that needs to be overcome.

[0063] In this system, the gap measurement module uses a line structured light camera as the core sensing unit, utilizing light fringe decomposition and triangulation principles to perform high-precision measurements of weld butt gaps. The structured light camera, combined with a high-speed image acquisition card, enables online acquisition and sub-pixel fitting of the weld geometry, maintaining millimeter-level and even sub-millimeter measurement accuracy even under complex lighting conditions. During measurement, the camera uses image preprocessing algorithms to remove noise and distortion, and incorporates a deep learning model to extract weld boundaries. The final output is numerical gap information that can be directly used in subsequent calculations, providing a reliable perception foundation for the entire adaptive control process.

[0064] The data communication module, using real-time Ethernet or an industrial bus, receives and forwards data from each submodule of the system, ensuring efficient exchange of measurement data and calculation results. The module integrates a data acquisition driver and transmission buffer, ensuring data integrity and timing consistency through a ring buffer and timestamp synchronization mechanism. Furthermore, the data communication module features fault self-detection and retransmission strategies, providing real-time responses to potential network jitter and communication interruptions. This ensures that the welding execution module can obtain the latest process parameters in a timely manner, enabling closed-loop feedback control.

[0065] The process parameter calculation module constructs a multivariate input-output mapping based on a BP neural network. It uses geometric features derived from gap measurements as network inputs and predicts key parameters such as laser power, arc current, and welding speed through a multilayer perceptron. This module pre-trains the network using extensive welding test data to extract the coupling relationship between weld heat input and penetration distribution within different gap ranges. During online operation, the module rapidly calculates the optimal welding parameters through forward propagation. When necessary, the neural network predictions are fine-tuned using genetic algorithms or particle swarm optimization to further reduce the risk of welding defects. The target values ​​output by the process parameter calculation module are transmitted to the execution layer via a data bus, achieving zero-latency parameter delivery.

[0066] The welding execution module, serving as the execution end of the system, integrates a high-power fiber laser and a CNC arc welder, and uses a robotic arm to achieve dynamic path tracking in three-dimensional space. The coordinated drive of the laser beam and arc welding wire achieves an energy superposition effect through a dedicated hybrid molten pool control strategy, ensuring high welding speeds while taking into account the weld depth-to-width ratio and microstructure performance. The robotic arm adjusts the welding gun posture in real time based on the welding speed and trajectory information issued by the process parameter calculation module to ensure precise alignment of the weld direction with the plate surface contour. The module also has built-in multiple temperature sensors and an optical monitoring unit, which uses molten pool image recognition to perform online monitoring of weld width and depth, and returns feedback data to the parameter calculation module to form a closed-loop precision control.

[0067] In order to facilitate system deployment and maintenance, the present invention also proposes a computer device and a readable storage medium solution to encapsulate the above-mentioned software algorithms and data processing logic into an executable program. The computer device adopts an industrial-grade embedded processor and a large-capacity Flash memory, which can run for a long time in a harsh workshop environment; the stored program files cover image preprocessing, neural network reasoning and optimization algorithms, and the underlying communication interface protocol with the hardware driver. When the processor loads the stored program, it can drive the entire adaptive control process to achieve closed-loop management of the entire process from sensor acquisition to execution feedback. In addition, the information data processing terminal can adjust the parameters of the welding task, perform online monitoring and status alarms through the touch screen interface to meet the diverse needs of different projects for welding quality and production rhythm.

[0068] The medium and thick plates of the steel structure of the present invention can be high-strength steel or other types of steel. The following description is based on the material and plate thickness of 10mm high-strength steel S700MC as an example, without limiting the application range of the material and plate thickness. The welding joint form of the present application is flat plate butt joint. In the following embodiments, the pre-processing of the plate adopts wire cutting. The purpose of the pre-processing of the plate of the present application is to ensure that the butt joint surface maintains the same roughness level and reduce the influence of roughness on weld formation. The material cutting method is not limited and can be wire cutting, laser cutting or other methods.

[0069] Example 1

[0070] like Figure 1 As shown, the present invention provides a method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel under variable butt gap, comprising:

[0071] Step S1, divide the gap into δ0, δ1, δ2, ..., δ M , gradually increasing from zero butt gap, and determining δ through process exploration i The welding process parameters with good forming under the gap of 100 mm / s are used to construct the "δ-process parameter" relationship table.

[0072] Step S2: construct a BPNN model based on the process parameters obtained in S1.

[0073] In step S3, the gap size-process parameter relationship is obtained based on the constructed BPNN model, and a curve showing the process parameter and gap variation is generated. By detecting the gap size in real time and obtaining the required process parameters based on the relationship, real-time control of the welding process parameters is achieved.

[0074] Specifically, a laser-arc hybrid heat source is used to weld a pre-set plate. The BPNN model serves as the process parameter calculation module, relying on a data communication module to transmit the calculated process parameters to the welding execution equipment (laser, welding robot). The BPNN model is constructed based on the acquired process parameters for different gap sizes. After training, it can achieve adaptive control of welding process parameters under varying gap sizes.

[0075] It should be noted that before welding, the surface of the pre-made plate needs to be polished with a wire brush and cleaned with denatured alcohol or acetone. Ensure that the floating dust or oxide layer generated during the plate cutting and pre-processing steps does not affect the welding results.

[0076] Example 2

[0077] The welding and process parameter test platform used in this application is as follows Figure 3As shown in the figure, the gap measurement module obtains the docking gap dimensions in real time and sends them to the process parameter calculation module on the computer. The process parameter calculation module establishes a connection with the KUKA robot's control cabinet through the data communication module, enabling two-way data transmission.

[0078] The KUKA robot is connected to the laser and arc welding machine via a field bus, and the laser power and arc parameters are controlled uniformly on the KUKA robot.

[0079] Specifically, the gap measurement module transmits the gap dimensions obtained in real time to the process parameter calculation module. This module generates XML-formatted characters on a computer and transmits them to the KUKA robot via the data communication module. Based on the I / O mapping table, the laser power and arc length correction values ​​are directly assigned to the KUKA robot's control signal locations for these parameters. Once the KUKA robot's laser power and arc length correction parameters are modified, they are reflected in real time to the laser and arc welder, enabling real-time control of the welding process parameters.

[0080] Example 3

[0081] Step S1 includes S11: It is necessary to determine how to adjust the parameter value according to the weld formation and the forming defects. i Install it on the welding test bench and preliminarily determine the gap δ based on the existing welding experience parameters. i The initial parameter p under i0 and Ac i0 ; S12: After obtaining the test results, adjust the parameters according to the weld formation to obtain a series of parameters p under the gap δi i1 、Ac i1 、p i2 、Ac i2 ...formation; S13: By comparing the formation quality of each weld, the optimal welding process parameters are selected, completing the process exploration of δi; S14: Repeating processes S11-S13, ultimately obtaining the process parameters that achieve good formation at each gap, and thus constructing a "δ-process parameter" relationship table. The weld formation and data results for the optimal welding parameters are shown in Table 1 below.

[0082] Table 1 Welding seam formation

[0083]

[0084]

[0085] Table 2 Welding process parameters

[0086]

[0087] Specifically, the butt joint gap is achieved by clamping a feeler gauge of a specific size between the butt joint surfaces of the two plates at the welding light on and off points. In actual welding, the plates are first fixed by spot welding before welding, and the feeler gauge is not removed during spot welding. Fixed process parameters are used in fixed gap welding to obtain different gap sizes δ i Under different welds, the weld with good shape and no obvious defects is selected as the current gap size δ i The good process parameters are recorded in Table 2 as a gap δ-process parameter relationship table. It should be noted that this application does not limit the implementation method of the docking gap.

[0088] Furthermore, in the experiment, this application determined the key process parameters that need to be regulated in laser-arc hybrid welding of medium and thick plates at different gap sizes - laser power and arc length correction parameters, among which laser power regulation is the main means to suppress collapse and root concave defects, and arc length correction is the main means to suppress upper surface undercut defects. It is clear that as the gap size increases, the laser power and arc length correction need to be appropriately reduced. It should be noted that there are many parameters in laser-arc hybrid welding, and the influence relationship between the parameters is complex. Parameters other than laser power and arc length correction are not within the scope of discussion of this application.

[0089] Example 4

[0090] This embodiment mainly describes step S2 in detail. Step S2 mainly includes building a gap size-process parameter correlation model based on BPNN, using the welding process parameters in Table 2 as the input data of the BPNN model and the gap size as the output data of the BPNN model, such as Figure 4 shown.

[0091] Specifically, regarding the parameter setting and training of the BPNN model, the laser power and arc length correction are used as the two nodes of the BPNN input layer, the number of hidden layer nodes of the BPNN is set to 4, and the gap size is used as the only node of the output layer. The gradient descent method is used for training, with a learning rate of 0.3, a momentum of 0.8, and a maximum number of iterations of 20,000. The prediction curve and training error curve of the constructed BPNN model are shown in Figure 2. Figure 5 shown.

[0092] Furthermore, the BPNN model is deployed on a computer using C# and .NET environments as a process parameter calculation module. This module connects to the measurement module and welding execution equipment via a data communication module, analyzing the measurement data to generate corresponding welding process parameters. The process parameters generated by the BPNN model are converted into XML format for transmission. It should be noted that this application does not restrict the deployment method of the BPNN model.

[0093] Example 5

[0094] Step S3 includes sub-step S31: The gap measurement module uses a line structured light camera to obtain the gap size. The line structured light camera used is SmartRay ECCO 95.040, which has a horizontal resolution of 0.013mm and a vertical resolution of 0.00135mm. Specifically, communication with the computer is established through C# and Socket Ethernet communication protocol. The image data obtained during the scanning process can be directly converted into the coordinate data of the laser surface according to the calibration method inside the camera. The gap width can be obtained by calculating the difference in the horizontal coordinate data at both ends of the gap, such as Figure 6 As shown. The acquired gap width data is transmitted to the process parameter calculation module in real time for subsequent steps. It should be noted that the line structured light camera and its related communication system used in this application are only used as a method demonstration and do not limit the brand, performance and data type of the line structured light camera.

[0095] Example 6

[0096] Step S3 includes sub-step S32: the process parameter calculation module calculates the process parameters required for the corresponding welding process parameters in real time according to the measured gap size and transmits them to the welding execution equipment through the data communication module, thereby realizing real-time control of the welding process parameters and achieving good weld formation under variable gap conditions.

[0097] Specifically, the gap size model obtained by the measurement module is input into the process parameter calculation module, and the BPNN model obtains the process parameter prediction value under the gap size, as well as the specific laser power and arc length correction value. The obtained laser power and arc length correction are converted into xml file format and sent to the KUKA robot through the data communication module. The laser power and arc length correction have corresponding parameter bits in the KUKA robot, and data transmission can be achieved based on the I / O mapping table. Finally, the 0-0.6mm variable gap is used as a verification method, and the plate clamping is as follows Figure 7 As shown. The light-on point is located at the 0mm gap, and the light-off point is on the other side, and welding is carried out from a small gap to a large gap. Finally, under the 0-0.6mm variable gap working condition, the welding process parameters can be changed in real time with the change of the gap size, and a good formation of a single-pass full penetration welding of a 10mm thick medium and thick plate can be achieved. The weld formation result is shown as follows Figure 8 shown.

[0098] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0099] Example 7

[0100] According to another aspect of the present invention, there is provided an adaptive control device for process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel with a variable butt gap, comprising:

[0101] The welding robot used for the welding process was a KUKA robot. The arc welder was a Fronius 500i welder, which needed to be set to a special two-step mode to enable real-time communication between the KUKA robot and the welder. The laser used was an IPGYLS-30000 fiber laser, and the laser power was controlled via a signal variable specified in the KUKA robot.

[0102] The gap measurement module uses a SmartRay ECCO 95.040 line structured light camera to collect the gap size of the plate joints in real time and send it to the process parameter calculation module in real time;

[0103] The process parameter calculation module uses the BPNN model to build a gap-process parameter correlation model to achieve real-time calculation of welding process parameters within a certain range and send the calculated welding process parameters to the KUKA robot in real time;

[0104] The data communication module uses C# and Sockets to establish data communication between the KUKA robot, the line structured light camera, and the computer. Specifically, the KUKA robot uses the Ethernet KRL communication protocol as a client for welding process parameters. The process parameter calculation module, acting as a server, converts the generated welding process parameters into an XML file that the KUKA robot can understand and transfers them to designated signal type variables within the KUKA robot, enabling real-time assignment of values ​​to the variables controlling the welding execution equipment within the KUKA robot. The KUKA robot communicates with the laser and arc welder using fieldbus.

[0105] It should be noted that this application does not limit the welding execution equipment, and other brands of welding heat sources can be used.

[0106] Example 8

[0107] According to another aspect of the present invention, an adaptive control system for process parameters of medium-thick plate high-strength steel laser-arc hybrid welding with a variable butt gap is provided, comprising a memory, a processor, and all of the aforementioned modules. The memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned method.

[0108] The present invention can be applied to butt welding of steel components in fields such as new energy vehicles, ships, and rail transit. Specifically, for the butt gap condition that is prone to occur in the welding of medium and thick plate steel, a laser-arc hybrid welding process and a method for adaptive control of process parameters are used to achieve a well-formed single-pass full penetration weld.

[0109] The present invention uses 0-0.6mm variable gap as a method for verification, and the plate is clamped as follows Figure 7 As shown. The light-on point is located at the 0mm gap, and the light-off point is on the other side, and welding is performed from a small gap to a large gap. Under the 0-0.6mm variable gap working condition, the welding process parameters can be adaptively controlled in real time as the gap size changes. The parameter changes are as follows: Figure 5 The present invention can finally achieve good forming of single-pass full penetration welding of 10mm thick medium and thick plates, and the weld formation result is as shown in FIG. Figure 8 shown.

[0110] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for adaptively controlling process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel, characterized in that: Including steps: The butt joint gap is divided into δ0, δ1, δ2…δM, and gradually increases from zero gap. For each δi, process exploration is carried out and the laser power and arc length correction that can achieve good weld formation are recorded, thereby establishing a gap-process parameter relationship table; Using the relationship table as a sample to train a back propagation neural network (BPNN), and deploying the trained BPNN in a computer as a process parameter calculation module; A line structured light camera is used to measure the butt gap in real time during the welding process. The measured gap data is input into the BPNN, and the corresponding laser power and arc length correction are immediately output and sent to the welding execution equipment to maintain good weld formation under variable gap conditions.

2. The method according to claim 1, characterized in that The process parameters only include the laser power p and the arc length correction Ac.

3. The method according to claim 1, characterized in that The process exploration includes: Install the steel plates to be welded under the gap δi, and set the initial parameters pi0 and Aci0 according to the empirical values; After welding, the parameters are adjusted in sequence according to the weld formation to obtain multiple groups of test data pi1 Aci1, pi2 Aci2…; Compare the quality of each group of welds to select the best parameters and complete the process determination of δi; Repeat the above steps for all gaps to form a complete gap-process parameter relationship table.

4. The method according to claim 1, wherein The input layer of the BPNN is a single node δi, the output layer is two nodes p and Ac, and at least one hidden layer is set in the middle.

5. The method according to claim 1, characterized in that The parameters output by BPNN are sent to the laser and welding gun control units in real time through the data communication module, and the robotic arm performs welding synchronously.

6. An adaptive control system for process parameters of laser-arc hybrid welding of medium and thick plate high-strength steel, characterized in that: include: Gap measurement module, used to collect the gap size of steel structure joints and output measurement data; Data communication module, used to forward data between various functional modules of the system; Process parameter calculation module, with built-in trained BPNN, calculates laser power and arc length correction based on gap size; The welding execution module includes a laser, a welding gun and a robotic arm, and is used to perform welding operations.

7. The system according to claim 6, characterized in that The gap measurement module is a line structured light camera, and the welding execution module can simultaneously receive the laser power signal and the arc length correction signal and output them according to the set value.

8. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 5.

9. A computer device comprising a processor and a memory, wherein the memory stores the computer-readable storage medium according to claim 8, and the processor implements the method according to any one of claims 1 to 5 when executing a program on the medium.

10. An information data processing terminal, characterized in that: It is equipped with the gap measurement module, data communication module and process parameter calculation module as described in claim 6, and can communicate with the welding execution module to complete the adaptive control of the laser-arc hybrid welding process parameters of medium and thick plate high-strength steel.

Images