Pipe high-precision free hot bending forming method and device, medium and equipment

Abstract

The invention discloses a high-precision free hot bending forming method, device, medium and equipment for pipes, and belongs to the technical field of plastic forming. According to the high-precision free hot bending forming method for the pipes, a pre-trained forming compensation model is adopted; according to the method, the track compensation amount of the machining track and the thermal power correction value of the working temperature are obtained, target pipe forming is controlled through the track compensation amount and the thermal power correction value, the machining track and the heating power of the pipe can be compensated and corrected in the forming process, the machining precision is improved, and the product quality is improved. The problems that mold use is reduced and the forming precision is not ideal in current pipe hot bending are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of pipe high-precision free thermal bending forming method, device, medium and equipment, belong to plastic forming technical field. BACKGROUND

[0002] In the industrial fields such as aviation, automobile, power plant, building, due to the requirements of lightweight design and customization manufacturing, the hot bending forming method of ultra-high strength steel pipe stands out.In the process of bending component, the hot bending forming method controls the trajectory of mechanical arm and the movement of bending roll die to control the pipe to realize complex three-dimensional forming, suitable for completing the component forming of complex variable curvature;But due to the reduction of dependence on die, both the manufacturing flexibility and response ability are improved, and the finished product precision of ultra-high strength steel pipe free thermal bending forming is required. SUMMARY

[0003] The purpose of the present application is to overcome the deficiencies in the prior art, provide a kind of pipe high-precision free thermal bending forming method, device, medium and equipment, can compensate and correct the processing trajectory of pipe and heating power, improve the processing precision, solve the problem of ideal forming precision that current pipe hot bending reduces die use.

[0004] To solve the above technical problems, the present application is realized by using the following technical scheme: The present application provides a kind of pipe high-precision free thermal bending forming method, comprising: determine the corresponding processing trajectory and working temperature according to the geometric parameters of target pipe; input the corresponding processing trajectory and working temperature into the pre-trained forming compensation model, determine the trajectory compensation amount of processing trajectory and the heat power correction value of working temperature; control target pipe forming according to trajectory compensation amount and heat power correction value.

[0005] Further, the corresponding processing trajectory and working temperature are determined according to the geometric parameters of target pipe, comprising: According to the geometric parameters of target pipe, calculate the corresponding execution end pose sequence and dynamic path; According to execution end pose sequence and dynamic path, determine the working temperature and processing trajectory of heating position.

[0006] Further, the training method of the pre-trained forming compensation model comprises: According to the selected processing trajectory and working temperature of pipe, obtain pipe real-time deformation error, temperature deviation and corresponding historical compensation data; Based on deep Q neural network, map real-time deformation error, temperature deviation and corresponding historical compensation data into observable virtual space. The agent acts in the virtual space and obtains the corresponding reward until the agent can perform all actions and get the highest total reward.

[0007] Further, the real-time deformation error is represented as: ; The real-time temperature deviation is represented as: ; The corresponding historical compensation data is represented as: ; represents the real-time deformation error, N represents the number of scanned pipe profile point coordinates, represents the scanned pipe actual profile point coordinate, i represents the pipe theoretical profile point coordinate, i represents the L2 norm; represents the real-time temperature deviation, represents the pipe measured temperature, represents the pipe material optimal temperature; represents the historical compensation data corresponding to the real-time deformation error and the real-time temperature deviation, represents the trajectory compensation amount at the time t, represents the trajectory compensation amount at the time t, represents the heating power at the time t, represents the compensation accuracy at the time t. Further, the deep Q neural network based on the real-time deformation error, the temperature deviation and the corresponding historical compensation data is mapped to the observable virtual space, comprising: The virtual space is represented as: ;

[0008] represents the state space, including the state of the agent at different times ; represents the state space, including the state of the agent at different times ; ; represents the curvature deviation, represents the axial temperature difference, represents the radial temperature difference, represents the residual stress distribution; ​​​​represents an action space, including actions made by the agent at different times ; ; represents a trajectory compensation amount represents a thermal power correction value of a working temperature represents a reward function, including reward values obtained by the agent at different times .

[0009] Further, the reward values obtained by the agent at different times are represented by the following formula : ; represents a time efficiency represents a rebound prediction amount ; represents a modulus of elasticity, represents a strain rate.

[0010] Further, the agent is caused to act in the virtual space and obtain corresponding rewards until the agent can perform all actions and obtain the highest total reward value, including: when the agent state , triggering an action , the agent state is updated to , and a reward value is obtained.

[0011] The second aspect of the present application provides a high-precision free thermal bending forming device for pipe materials, comprising: an acquisition module configured to determine corresponding machining trajectories and working temperatures according to geometric parameters of a target pipe material; a compensation module configured to input the corresponding machining trajectories and working temperatures into a pre-trained forming compensation model to determine a trajectory compensation amount for the machining trajectories and a thermal power correction value for the working temperatures; a processing module configured to control forming of the target pipe material according to the trajectory compensation amount and the thermal power correction value.

[0012] The third aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the high-precision free thermal bending forming method for pipe materials as described above.

[0013] The present application also provides a computer device, comprising: a memory configured to store instructions; A processor is configured to execute the instructions to cause the device to perform a pipe high-precision free thermal bending forming method as described above.

[0014] Compared with the prior art, the present application has the following advantages: 1、The present application adopts a pre-trained forming compensation model, obtains a track compensation amount for a processing track and a thermal power correction value for a working temperature according to the processing track and the working temperature of the target pipe, and controls the target pipe forming through the track compensation amount and the thermal power correction value, so that the processing track and the heating power of the pipe can be compensated and corrected during the forming process, the processing precision is improved, and the problems of reducing the use of molds and the unsatisfactory forming precision in the current pipe thermal bending are solved.

[0015] 2、The present application is based on real-time deformation error, temperature deviation and corresponding historical compensation data of the pipe, adopts a reinforcement learning to train a forming compensation model, associates the curvature deviation with the track compensation amount and the thermal power correction value, and avoids the performance degradation problem caused by thermal softening or low-temperature brittleness of the material. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the pipe high-precision free thermal bending forming method provided by the embodiment of the present application; Figure 2 is a structural schematic diagram of a robot execution module provided by the embodiment of the present application; (in the figure, 1, a mechanical arm fixing ring; 2, an auxiliary heat source; 3, a split inductive coil; 4, a multi-source sensing module; 5, a mechanical arm) Figure 3 is a training method flowchart of the pre-trained forming compensation model provided by the embodiment of the present application; Figure 4 is a pipe high-precision free thermal bending forming device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0017] The present application will be further described below in conjunction with the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and cannot be used to limit the protection scope of the present application.

[0018] Embodiment 1 As shown in the figure, the pipe high-precision free thermal bending forming method comprises: Figure 1 According to the geometric parameters of the target pipe, the corresponding processing track and working temperature are determined, specifically: According to the geometric parameters of the target pipe CAD model, 500 groups of key point cloud data are extracted, and the curvature change interval is According to the geometric parameters of the target pipe CAD model, 500 groups of key point cloud data are extracted, and the curvature change interval is In the inflection point P1, P2 of the curvature mutation region, three control points are sequentially inserted, which are represented as: control point A, inflection point P1, control point B, inflection point P2, control point C; The corresponding robot mechanical arm six-axis 500 group coordinate pose sequence and the bending roller dynamic path function are calculated: ; The transpose matrix is represented as: The horizontal transverse displacement of the pipe bending is represented as: The vertical height change is represented as: The axial feeding depth is represented as:

[0019] According to the execution end pose sequence and the dynamic path, the working temperature of the heating position and the machining track are determined; As shown in Figure 2 , taking the robot execution module as an example, the execution end, the multi-source sensing module 4, the mechanical arm 5 and the controller (not shown) are shown; Among them, the execution end includes a mechanical arm fixing ring 1, multiple groups of auxiliary heat sources 2 and a three-level split induction coil 3; The mechanical arm fixing ring 1 is connected with the mechanical arm; The controller is connected with the mechanical arm 5, the multiple groups of auxiliary heat sources 2, the three-level split induction coil 3 and the multi-source sensing module 4 respectively; It should be noted that the multi-source sensing module 4 includes a laser scanner, a strain sensor and an infrared thermal imager.

[0020] The heating of the target pipe is in the form of gradient heating. In this embodiment, the axial gradient coil is configured as three groups of split induction coils with a spacing of 36 mm, the optimal temperature of the inlet coil is 900℃, the optimal temperature of the middle coil is 950℃, and the optimal temperature of the outlet coil is 1000℃; The radial gradient adopts four groups of auxiliary infrared heaters, and the optimal temperature of the outside of the target pipe is 860℃ and the optimal temperature of the inside is 740℃.

[0021] As shown in Figure 3 , the corresponding machining track and working temperature are input into the pre-trained forming compensation model to determine the track compensation amount of the machining track and the thermal power correction value of the working temperature; wherein the training method of the pre-trained forming compensation model comprises: According to the selected machining track and working temperature of the pipe, the real-time deformation error, temperature deviation and corresponding historical compensation data of the pipe are obtained, specifically: The real-time deformation error is represented as: ; The real-time temperature deviation is represented as: ; The corresponding historical compensation data is represented as: ; represents real-time deformation error, N represents the number of scanned pipe profile point coordinates, represents the scanned pipe profile point coordinates, i represents the pipe theoretical profile point coordinates, i represents the L2 norm; represents real-time temperature deviation, represents the measured temperature of the pipe, represents the optimal temperature of the pipe material; represents the historical compensation data corresponding to the real-time deformation error and the real-time temperature deviation, represents the trajectory compensation amount at represents the trajectory compensation amount at represents the heating power at represents the compensation accuracy at

[0022] Based on deep Q neural network, the real-time deformation error, temperature deviation and corresponding historical compensation data are mapped into an observable virtual space, specifically: The virtual space is represented as: ; represents the state space, including the state of the agent at different times ; ; represents the curvature deviation, measured by a laser scanner; represents the axial temperature difference, represents the radial temperature difference, measured by an infrared thermal imager; represents the residual stress distribution, measured by a strain sensor; represents the action space, including the actions made by the agent at different times ; ; represents the trajectory compensation amount; represents the thermal power correction value of the working temperature; ​​​​​​ represents a reward function, including the reward values obtained by the agent at different times ; ; represents time efficiency; represents a rebound prediction quantity; ; represents the elastic modulus, represents the strain rate.

[0023] The agent is caused to act in a virtual space and obtain a corresponding reward until the agent can perform all actions and obtain the highest total reward; wherein when the agent state , the action is triggered, the agent state is updated to , and the reward value is obtained; According to the trajectory compensation quantity and the thermal power correction value, the target pipe is formed, taking the inflection point P1 of the curvature mutation zone as an example, at the control point A, the and are calculated , and the specific trajectory compensation quantity is calculated ; ; represents a dynamic rebound coefficient, and the value is 0.85; At the control point A near the inflection point P1 of the curvature mutation zone, the reverse displacement is superimposed to correct the pose of the robot arm.

[0024] Embodiment 2 As shown in Figure 4 , the pipe high-precision free thermal bending forming device comprises: An acquisition module is configured to determine a corresponding processing trajectory and working temperature according to geometric parameters of a target pipe; A compensation module is configured to input the corresponding processing trajectory and working temperature into a pre-trained forming compensation model to determine a trajectory compensation quantity for the processing trajectory and a thermal power correction value for the working temperature; A processing module is configured to control the forming of the target pipe according to the trajectory compensation quantity and the thermal power correction value.

[0025] Embodiment 3 A computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the following method: determine a corresponding processing trajectory and a working temperature according to a geometric parameter of the target pipe; input the corresponding processing trajectory and the working temperature into a pre-trained forming compensation model to determine a trajectory compensation amount for the processing trajectory and a heat power correction value for the working temperature; control the forming of the target pipe according to the trajectory compensation amount and the heat power correction value.

[0026] Embodiment 4 A computer device comprises: a memory for storing instructions; a processor for executing the instructions to cause the device to perform a method comprising: determining a corresponding processing trajectory and a working temperature according to a geometric parameter of the target pipe; inputting the corresponding processing trajectory and the working temperature into a pre-trained forming compensation model to determine a trajectory compensation amount for the processing trajectory and a heat power correction value for the working temperature; controlling the forming of the target pipe according to the trajectory compensation amount and the heat power correction value.

[0027] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0028] The present application is described with reference to flowcharts according to the embodiments of the present application. It should be understood that each flow in the flowcharts can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flow Figure 1 an apparatus for implementing functions specified in one or more flows or one or more blocks. Figure 1

[0029] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus that implements the functions specified in one or more flows or one or more blocks. Figure 1 an apparatus for implementing functions specified in one or more flows or one or more blocks.

[0030] ​These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the functions specified in the flowchart Figure 1 The steps of the functions specified in the flowchart or the flowcharts.

[0031] The embodiments of the present application are described above with reference to the drawings, but the present application is not limited to the specific embodiments described above, and the specific embodiments described above are merely illustrative, but not restrictive, and those skilled in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, which are all within the protection of the present application.

Claims

1. A high-precision free hot bending forming method for pipes, characterized in that: include: Determine the corresponding processing trajectory and working temperature based on the geometric parameters of the target pipe; The corresponding machining trajectory and working temperature are input into a pre-trained forming compensation model to determine the trajectory compensation amount for the machining trajectory and the thermal power correction value for the working temperature; The target tube forming is controlled according to the trajectory compensation amount and the thermal power correction value.

2. The high-precision free hot bending forming method for pipes according to claim 1, characterized in that: Determining the corresponding processing trajectory and working temperature according to the geometric parameters of the target pipe includes: According to the geometric parameters of the target pipe, the corresponding execution end pose sequence and dynamic path are calculated; According to the execution end posture sequence and dynamic path, the working temperature and processing trajectory of the heating position are determined.

3. The high-precision free hot bending forming method for pipes according to claim 1, characterized in that: The training method of the pre-trained forming compensation model includes: According to the processing trajectory and working temperature of the selected pipe, the real-time deformation error, temperature deviation and corresponding historical compensation data of the pipe are obtained; Based on the deep Q neural network, the real-time deformation error, temperature deviation and corresponding historical compensation data are mapped into the observable virtual space; Make the agent move in the virtual space and obtain corresponding rewards until the agent can perform all actions and obtain the highest total reward.

4. The high-precision free hot bending forming method for pipes according to claim 3, characterized in that: The real-time deformation error is expressed as: ； The real-time temperature deviation is expressed as: ； The corresponding historical compensation data is expressed as: ； represents the real-time deformation error, N Indicates the number of coordinates of the scanned pipe contour points, Indicates the scanned pipe i The actual contour point coordinates, Indicates the pipe i Theoretical contour point coordinates, represents the L2 norm; Indicates real-time temperature deviation, Indicates the actual measured temperature of the pipe. Indicates the optimal temperature of the pipe material; Indicates historical compensation data corresponding to real-time deformation error and real-time temperature deviation. Indicates The trajectory compensation amount at the moment, Indicates The trajectory compensation amount at the moment, Indicates Heating power at the moment, Indicates Compensation accuracy at all times.

5. The high-precision free hot bending forming method for pipes according to claim 3, characterized in that: The deep Q neural network is used to map real-time deformation error, temperature deviation and corresponding historical compensation data into an observable virtual space, including: The virtual space is represented as: ； Represents the state space, including the state of the agent at different times ; ； represents the curvature deviation, represents the axial temperature difference, Radial temperature difference, represents the residual stress distribution; Represents the action space, including the actions taken by the agent at different times ; ； Indicates the trajectory compensation amount; Indicates the thermal power correction value of the operating temperature; Represents the reward function, including the reward values ​​obtained by the agent at different times .

6. The high-precision free hot bending forming method for pipes according to claim 5, characterized in that: The reward value obtained by the agent at different times is expressed as follows : ； Indicates time efficiency; Represents the rebound prediction amount; ； represents the elastic modulus, represents the strain rate.

7. The high-precision free hot bending forming method for pipes according to claim 5, characterized in that: The agent is made to act in the virtual space and obtain corresponding rewards until the agent can perform all actions and obtain the highest total reward, including: when the agent is in the state When the action is triggered , agent state Updated to , get reward value .

8. A high-precision free hot bending forming device for pipes, characterized in that: include: The acquisition module is used to determine the corresponding processing trajectory and working temperature according to the geometric parameters of the target pipe; A compensation module is used to input the corresponding processing trajectory and working temperature into a pre-trained forming compensation model to determine the trajectory compensation amount for the processing trajectory and the thermal power correction value for the working temperature; The processing module is used to control the target tube forming according to the trajectory compensation amount and the thermal power correction value.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the high-precision free hot bending forming method for pipes according to any one of claims 1 to 7 is implemented.

10. A computer device, characterized in that: include: a memory for storing instructions; A processor is used to execute the instructions so that the device performs the high-precision free hot bending forming method for pipes as described in any one of claims 1 to 7.

Images