A box-shaped steel member processing method

Abstract

The application discloses a box type steel component machining method, which comprises the following steps: generating machining data based on a component BIM model, driving a laser cutting device to complete one-time precise cutting of a plate and a bevel; obtaining actual point clouds of the cut plate through three-dimensional scanning; performing welding under the planning of a welding digital twin model, and simultaneously collecting infrared thermal field distribution data, restraint force change data and geometric deformation data of the component in the welding process in real time; dynamically adjusting the welding process according to the multi-source feedback data to realize active regulation of welding stress and deformation; giving each component a unique identification, and binding machining parameters, process monitoring data and final detection results of each link to the identification; and constructing and continuously optimizing a process prediction model by using a machine learning algorithm, so as to recommend or fine-tune process parameters for new components.

Description

Technical Field

[0001] This invention belongs to the field of steel component processing, and specifically relates to a method for processing box-shaped steel components. Background Technology

[0002] The processing of box-type steel components relies on discrete semi-automatic equipment and human experience. Precision assurance in the material cutting, assembly, and welding stages is independent, leading to significant error accumulation and reliance on post-weld correction. The transient stresses and deformations induced by the welding thermal process are complex; traditional methods, relying on fixed sequences and post-weld correction, have limited effectiveness and damage the base material. Furthermore, it is difficult to quickly respond to design changes or batch variations in materials. Existing improvements mostly focus on single-point technologies, such as using laser-cut beveling or optimizing welding sequences; they lack intelligent processing methods with a closed-loop linkage of perception, decision-making, and execution. Summary of the Invention

[0003] To achieve the above objectives, the technical solution of the present invention is as follows: a method for processing box-type steel components, the method comprising the following steps, S1. Generate processing data based on the component BIM model, and drive the laser cutting equipment to complete the one-time precision cutting of the plate and bevel; S2. Obtain the actual point cloud of the cut board through 3D scanning, compare it with the design model, and guide the actuator to apply active corrective force to the board during the assembly process based on the comparison results to complete the pre-assembly. S3. Welding is performed under the planning of the welding digital twin model, while infrared thermal field distribution data, restraint force change data and geometric deformation data of the component are collected in real time during the welding process; the welding process is dynamically adjusted based on the multi-source feedback data to achieve active control of welding stress and deformation. S4. Assign a unique identifier to each component and bind the processing parameters, process monitoring data and final test results of each stage to the identifier; S5. Utilize machine learning algorithms to build and continuously optimize process prediction models to recommend or fine-tune process parameters for new components.

[0004] Preferably, step S2, which guides the actuator to apply active corrective force to the sheet metal during assembly based on the comparison results, includes the following steps: S21. Input the comparison results of the actual point cloud and the design point cloud into the prediction model. The prediction model outputs the grasping strategy and motion trajectory instructions for correcting the deformation of the plate. S22. According to the instructions, the actuator applies a controllable force or torque through the end effector during the process of grasping and moving the plate to the assembly position, so that the shape of the plate approaches the design model.

[0005] Preferably, the prediction model is a deep learning model trained based on historical plate deformation data and successful correction strategy data.

[0006] Preferably, the dynamic adjustment welding process includes one or more of the following actions: Dynamically adjust the welding sequence, welding speed, or welding energy input of the welding robot; dynamically adjust the restraint state or position of the clamps used to fix components; instruct the welding robot or auxiliary heat source to compensate for heating or cooling of specific areas.

[0007] Preferably, step S3 further includes a stress buffer construction step: after or during the main weld welding, a layer of low-strength stress buffer material is clad in the stress concentration area of ​​the component by additive manufacturing.

[0008] Preferably, in step S5, the input features of the process prediction model include plate specifications, bevel type, assembly gap, initial welding parameters, and real-time thermodynamic data; the output targets include predicted deformation, predicted residual stress distribution, or weld quality grade.

[0009] Preferably, an intelligent processing system for box-type steel components to implement the method includes: a central control and data processing platform for running a digital twin model and intelligent decision-making algorithms; a laser precision cutting unit connected to the platform; a three-dimensional vision-guided assembly unit including a scanner for acquiring three-dimensional point cloud data of the plate, an execution mechanism for performing gripping and assembly, and an assembly control module for processing point cloud data and generating correction instructions; a multi-physics field collaborative welding unit including a welding robot, an infrared thermal imager for monitoring the temperature field of the component, a fixture integrating a force sensor, a measuring device for monitoring the deformation of the component, and a welding collaborative controller for processing multi-source feedback and outputting adjustment instructions; and a data acquisition and traceability unit for acquiring data from each stage and binding it to component identification.

[0010] Preferably, the actuator in the three-dimensional vision-guided assembly unit is a multi-joint robot with a flexible gripper at its end that can apply multi-dimensional force control.

[0011] Preferably, the fixture in the multiphysics collaborative welding unit is a flexible fixture system with active posture adjustment capability.

[0012] Preferably, it also includes an auxiliary stress control unit, which includes a robot equipped with an additive manufacturing nozzle for cladding stress-absorbing material onto the component.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: The box-type steel component processing method provided by the present invention transforms the traditional open-loop process into a closed loop, suppressing error transmission from the source and greatly improving the accuracy of the assembly foundation; and realizing real-time active control of welding stress deformation. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the processing method described in this invention. Detailed Implementation

[0015] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0016] Example: Figure 1 As shown, to achieve the above objectives, the technical solution of the present invention is as follows: a method for processing box-type steel components, the method comprising the following steps, S1. Generate processing data based on the component BIM model, and drive the laser cutting equipment to complete the one-time precision cutting of the plate and bevel; S2. Obtain the actual point cloud of the cut board through 3D scanning, compare it with the design model, and guide the actuator to apply active corrective force to the board during the assembly process based on the comparison results to complete the pre-assembly. S3. Welding is performed under the planning of the welding digital twin model, while infrared thermal field distribution data, restraint force change data and geometric deformation data of the component are collected in real time during the welding process; the welding process is dynamically adjusted based on the multi-source feedback data to achieve active control of welding stress and deformation. S4. Assign a unique identifier to each component and bind the processing parameters, process monitoring data and final test results of each stage to the identifier; S5. Utilize machine learning algorithms to build and continuously optimize process prediction models to recommend or fine-tune process parameters for new components.

[0017] Furthermore, step S2, which guides the actuator to apply active corrective force to the sheet metal during assembly based on the comparison results, includes the following steps: S21. Input the comparison results of the actual point cloud and the design point cloud into the prediction model. The prediction model outputs the grasping strategy and motion trajectory instructions for correcting the deformation of the plate. S22. According to the instructions, the actuator applies a controllable force or torque through the end effector during the process of grasping and moving the plate to the assembly position, so that the shape of the plate approaches the design model.

[0018] Furthermore, the prediction model is a deep learning model trained based on historical plate deformation data and successful correction strategy data.

[0019] Furthermore, the dynamically adjusted welding process includes one or more of the following actions: Dynamically adjust the welding sequence, welding speed, or welding energy input of the welding robot; dynamically adjust the restraint state or position of the clamps used to fix components; instruct the welding robot or auxiliary heat source to compensate for heating or cooling of specific areas.

[0020] Furthermore, step S3 also includes a stress buffer construction step: after or during the main weld welding, a layer of low-strength stress buffer material is clad in the stress concentration area of ​​the component by additive manufacturing.

[0021] Furthermore, in step S5, the input features of the process prediction model include plate specifications, bevel type, assembly gap, initial welding parameters, and real-time thermodynamic data; the output targets include predicted deformation, predicted residual stress distribution, or weld quality grade.

[0022] Furthermore, an intelligent processing system for box-type steel components to implement the method includes: a central control and data processing platform for running a digital twin model and intelligent decision-making algorithms; a laser precision cutting unit connected to the platform; a three-dimensional vision-guided assembly unit including a scanner for acquiring three-dimensional point cloud data of the plate, an execution mechanism for performing gripping and assembly, and an assembly control module for processing point cloud data and generating correction instructions; a multi-physics field collaborative welding unit including a welding robot, an infrared thermal imager for monitoring the temperature field of the component, a fixture integrating a force sensor, a measuring device for monitoring the deformation of the component, and a welding collaborative controller for processing multi-source feedback and outputting adjustment instructions; and a data acquisition and traceability unit for acquiring data from each stage and binding it to component identification.

[0023] Furthermore, the actuator in the three-dimensional vision-guided assembly unit is a multi-joint robot with a flexible gripper at its end that can apply multi-dimensional force control.

[0024] Furthermore, the fixture in the multiphysics collaborative welding unit is a flexible fixture system with active posture adjustment capability.

[0025] Furthermore, it also includes an auxiliary stress control unit, which includes a robot equipped with an additive manufacturing nozzle for cladding stress-reducing material onto the component.

[0026] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A method of processing a box-shaped steel member, characterized by, The method includes the following steps: S1. Generate processing data based on the component BIM model, and drive the laser cutting equipment to complete the one-time precision cutting of the plate and bevel; S2. Obtain the actual point cloud of the cut board through 3D scanning, compare it with the design model, and guide the actuator to apply active corrective force to the board during the assembly process based on the comparison results to complete the pre-assembly. S3. Welding is performed under the planning of the welding digital twin model, while infrared thermal field distribution data, restraint force change data and geometric deformation data of the component are collected in real time during the welding process; the welding process is dynamically adjusted based on the multi-source feedback data to achieve active control of welding stress and deformation. S4. Assign a unique identifier to each component and bind the processing parameters, process monitoring data and final test results of each stage to the identifier; S5. Utilize machine learning algorithms to build and continuously optimize process prediction models to recommend or fine-tune process parameters for new components.

2. The method of claim 1, wherein, Step S2, based on the comparison results, guides the actuator to apply active straightening force to the sheet metal during the assembly process, including the following steps: S21. Input the comparison results of the actual point cloud and the design point cloud into the prediction model. The prediction model outputs the grasping strategy and motion trajectory instructions for correcting the deformation of the plate. S22. According to the instructions, the actuator applies a controllable force or torque through the end effector during the process of grasping and moving the plate to the assembly position, so that the shape of the plate approaches the design model.

3. The method according to claim 1, characterized in that, The prediction model is a deep learning model trained based on historical plate deformation data and successful correction strategy data.

4. The method according to claim 1, characterized in that, The dynamically adjusted welding process includes one or more of the following actions: Dynamically adjust the welding sequence, welding speed, or welding energy input of the welding robot; dynamically adjust the restraint state or position of the clamps used to fix components; instruct the welding robot or auxiliary heat source to compensate for heating or cooling of specific areas.

5. The method according to claim 1, characterized in that, Step S3 further includes a stress buffer construction step: after the main weld is completed or during the welding process, a layer of low-strength stress buffer material is clad in the stress concentration area of ​​the component by additive manufacturing.

6. The method according to claim 1, characterized in that, In step S5, the input features of the process prediction model include plate specifications, bevel type, assembly gap, initial welding parameters, and real-time thermodynamic data; the output targets include predicted deformation, predicted residual stress distribution, or weld quality grade.

7. An intelligent processing system for box-type steel components for implementing the method according to any one of claims 1-6, characterized in that, include: A central control and data processing platform is used to run digital twin models and intelligent decision-making algorithms; a laser precision cutting unit is connected to the platform; a three-dimensional vision-guided assembly unit includes a scanner for acquiring three-dimensional point cloud data of the board, an actuator for performing gripping and assembly, and an assembly control module for processing point cloud data and generating correction instructions. The multi-physics collaborative welding unit includes a welding robot, an infrared thermal imager for monitoring the temperature field of the component, a fixture with integrated force sensors, a measuring device for monitoring the deformation of the component, and a welding collaborative controller for processing multi-source feedback and outputting adjustment commands. The data acquisition and traceability unit is used to collect data from each stage and bind it to the component identifier.

8. The system according to claim 7, characterized in that, The actuator in the three-dimensional vision-guided assembly unit is a multi-joint robot with a flexible gripper at its end that can apply multi-dimensional force control.

9. The system according to claim 7, characterized in that, The fixture in the multiphysics field collaborative welding unit is a flexible fixture system with active posture adjustment capability.

10. The system according to claim 7, characterized in that, It also includes an auxiliary stress control unit, which includes a robot equipped with an additive manufacturing nozzle for cladding stress-absorbing material onto the component.

Images