An adaptive control method and system for the power spinning forming of metal components

Abstract

The present invention relates to the field of intelligent manufacturing, and discloses an adaptive control method for the power spinning forming of metal components, which includes the following steps: establishing a virtual machining environment based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model; in the virtual machining environment, obtaining corresponding reward values by simulating the power spinning process and the springback process of the metal component; taking the feed rate, machining state, and reward value as inputs, establishing a mapping relationship through a deep learning algorithm, obtaining a trained control model, and outputting an optimized feed rate; deploying the trained control model to the motion controller of the spinning machine to achieve real-time control of the spinning machine. Through the present invention, it is possible to change the acting force of the spinning wheel on the metal forming, improve the stress-strain state in the plastic deformation zone, and improve the workpiece quality.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, and specifically to an adaptive control method and system for the power spinning forming of metal components. Background Art

[0002] Metal power spinning is a non-chip machining process. This process is easy to manufacture seamless ring-shaped parts and is particularly suitable for the forming of components with a large diameter-thickness ratio. The metal power spinning process has the advantages of saving material consumption, low cost, simple equipment, and high product quality, and has been widely used in key national industrial fields such as aerospace and machinery. With the rapid development of advanced manufacturing technologies, the industrial field has increasing requirements for the forming quality of spun parts, and the quality control of metal spinning forming has become extremely urgent. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an adaptive control method for the power spinning forming of metal components, including the following steps:

[0004] Step 1, based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model, establish a virtual machining environment;

[0005] Step 2, in the virtual machining environment, by simulating the shear spinning process and the springback process of the metal component power spinning, reproduce the spinning forming process and forming defects of the metal component; by deploying a sensor simulation module, monitor the machining state, machining quality, and stress state of the metal component in real time, and obtain the corresponding reward value through a preset evaluation function in combination with the machining quality index;

[0006] Step 3, taking the feed rate, machining state, and reward value as inputs, establish a mapping relationship through a deep learning algorithm, and optimize through a reinforcement learning algorithm, update the deep learning strategy, and after multiple trainings under different working conditions, obtain a trained control model and output an optimized feed rate;

[0007] Step 4, deploy the trained control model to the motion controller of the spinning machine, obtain the real-time machining state data of the spinning machine through real-time data acquisition, and realize the real-time control of the spinning machine by online updating and optimizing the feed rate according to the real-time machining state data.

[0008] Further, the establishment of the virtual machining environment based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model includes:

[0009] According to the stress-strain relationship of metal materials during the spinning process, a constitutive model of metal materials is constructed; according to the vibration and stiffness of the machine tool, a dynamic characteristic model of the machine tool is established; a springback model is established to predict the springback behavior of the workpiece after spinning forming; through the constitutive model of metal materials, the dynamic characteristic model of the machine tool and the springback model, a complete virtual machining environment is formed.

[0010] Further, in the virtual machining environment, by simulating the power spinning and shear spinning processes and the springback process of metal components, the spinning forming process and forming defects of metal components are reproduced; by deploying a sensor simulation module, the machining state, machining quality and stress state of metal components are monitored in real time, and through a preset evaluation function, combined with machining quality indicators, the corresponding reward values are obtained, including:

[0011] Through the sensors in the virtual environment, various state parameters of the metal component during the spinning process are monitored in real time to obtain the machining simulation results; according to the simulation results, the machining quality of the metal component is evaluated to obtain the machining quality evaluation results; according to the stress state of the metal component during the machining process, the monitoring results of the low-stress state are obtained; according to the machining state, the machining quality evaluation results and the monitoring results of the low-stress state, the corresponding reward values are calculated.

[0012] Further, taking the feed rate, machining state, and reward value as inputs, a mapping relationship is established through a deep learning algorithm, and through the reinforcement learning algorithm for optimization, the deep learning strategy is updated. After multiple trainings under different working conditions, a trained control model is obtained, and the optimized feed rate is output, including:

[0013] Select a deep learning algorithm, take the feed rate, machining state, and reward value as inputs, construct a deep learning model, and learn the mapping relationship between the input and output through training data; use the reinforcement learning algorithm to optimize the deep learning model, and conduct multiple trainings under different working conditions to complete the training of the control model; the deep learning algorithm is one of the convolutional neural network and the recurrent neural network.

[0014] Further, the trained control model is deployed to the motion controller of the spinning machine. Through real-time data acquisition, the real-time machining state data of the spinning machine is obtained. According to the real-time machining state data, the feed rate is updated and optimized online to realize the real-time control of the spinning machine, including:

[0015] The trained control model is deployed to the motion controller of the spinning machine. During the actual machining process, the motion state of the spinning machine and the machining state of the metal component are collected in real time; the collected real-time data is input into the control model, and the feed rate is updated and optimized online according to the current machining state; the optimized feed rate is output to the motion controller of the spinning machine to realize the real-time motion control of the spinning machine.

[0016] An adaptive control system for the power spinning forming of metal components, applying the described adaptive control method for the power spinning forming of metal components, is characterized in that it includes a spinning machine motion controller, a power spinning simulation module for metal components, a data acquisition module, a data processing module, and a communication module;

[0017] The spinning machine motion controller, the power spinning simulation module for metal components, the data acquisition module, and the communication module are respectively connected to the data processing module.

[0018] The beneficial effect of the present invention is that by simulating the machining of metal components, an optimized feed rate is obtained, the main process parameter feed rate of the spinning forming is automatically adjusted, the force exerted by the spinning wheel on the metal forming is changed, and the stress-strain state in the plastic deformation zone is improved, thereby improving the workpiece quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic flow diagram of an adaptive control method for the power spinning forming of metal components;

[0020] Figure 2 A schematic principle diagram of an adaptive control system for the power spinning forming of metal components;

[0021] Figure 3 It is an implementation schematic diagram of an adaptive control method for the power spinning forming of metal components. DETAILED DESCRIPTION OF THE INVENTION

[0022] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following description.

[0023] The features and performance of the present invention will be further described in detail below with reference to the embodiments.

[0024] As Figure 1 shown, an adaptive control method for the power spinning forming of metal components includes the following steps:

[0025] Step 1: Based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model, a virtual machining environment is established;

[0026] Step 2: In the virtual machining environment, by simulating the power spinning shear spinning process and the springback process of the metal component, the spinning forming process and forming defects of the metal component are reproduced; by deploying a sensor simulation module, the machining state, machining quality, and stress state of the metal component are monitored in real time, and through a preset evaluation function, combined with the machining quality index, the corresponding reward value is obtained;

[0027] Step 3: Taking the feed rate, processing state, and reward value as inputs, establish a mapping relationship through a deep learning algorithm, optimize it through a reinforcement learning algorithm, update the deep learning strategy, and after multiple trainings under different working conditions, obtain a trained control model and verify the output of the optimized feed rate;

[0028] Step 4: Deploy the trained control model to the motion controller of the spinning machine. Through real-time data acquisition, obtain the real-time processing state data of the spinning machine. According to the real-time processing state data, realize the real-time control of the spinning machine by online updating and optimizing the feed rate.

[0029] Establish a virtual machining environment based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model, including:

[0030] Construct the constitutive model of the metal material according to the stress-strain relationship of the metal material during the spinning process; establish the dynamic characteristic model of the machine tool according to the vibration and stiffness of the machine tool; establish a springback model to predict the springback behavior of the workpiece after spinning forming; form a complete virtual machining environment through the constitutive model of the metal material, the dynamic characteristic model of the machine tool, and the springback model.

[0031] In the virtual machining environment, by simulating the power spinning and shear spinning processes and the springback process of the metal component, reproduce the spinning forming process and forming defects of the metal component; by deploying a sensor simulation module, monitor the processing state, processing quality, and stress state of the metal component in real time, and obtain the corresponding reward value through a preset evaluation function in combination with the processing quality index, including:

[0032] Through the sensors in the virtual environment, monitor various state parameters of the metal component during the spinning process in real time to obtain the processing simulation results; evaluate the processing quality of the metal component according to the simulation results to obtain the processing quality evaluation results; obtain the monitoring results of the low-stress state according to the stress state of the metal component during the processing; calculate the corresponding reward value according to the processing state, processing quality evaluation results, and monitoring results of the low-stress state.

[0033] Taking the feed rate, processing state, and reward value as inputs, establish a mapping relationship through a deep learning algorithm, optimize it through a reinforcement learning algorithm, update the deep learning strategy, and after multiple trainings under different working conditions, obtain a trained control model and verify the output of the optimized feed rate, including:

[0034] Select a deep learning algorithm, use the feed rate, machining status, and reward value as inputs to build a deep learning model, and learn the mapping relationship between the inputs and outputs through training data; use a reinforcement learning algorithm to optimize the deep learning model, and perform multiple trainings under different working conditions to complete the training of the control model; the deep learning algorithm is one of the convolutional neural network and the recurrent neural network.

[0035] Deploy the trained control model to the motion controller of the spinning machine. Through real-time data acquisition, obtain the real-time machining status data of the spinning machine. According to the real-time machining status data, realize the real-time control of the spinning machine by online updating and optimizing the feed rate, including:

[0036] Deploy the trained control model to the motion controller of the spinning machine. During the actual machining process, collect the motion status of the spinning machine and the machining status of the metal component in real time; input the collected real-time data into the control model, and online update and optimize the feed rate according to the current machining status; output the optimized feed rate to the motion controller of the spinning machine to realize the real-time motion control of the spinning machine.

[0037] As Figure 2 shown, an adaptive control system for the power spinning forming of metal components, applying the described adaptive control method for the power spinning forming of metal components, is characterized by including a motion controller of the spinning machine, a power spinning simulation module of the metal component, a data acquisition module, a data processing module, and a communication module;

[0038] The motion controller of the spinning machine, the power spinning simulation module of the metal component, the data acquisition module, and the communication module are respectively connected to the data processing module.

[0039] Specifically, the adaptive control method for the power spinning forming of metal components provided by the present invention is aimed at the following two situations:

[0040] 1) During the power spinning forming process, the random vibration of the machine tool causes the stress and strain state of the metal in the plastic deformation zone to change, resulting in the phenomenon of plastic flow instability of the metal during the forming process, and easily causing forming defects such as non-conforming die sticking, local bulging, blank swinging, and flange wrinkling during the power spinning forming of metal components.

[0041] 2) The springback of the spun workpiece after spinning results in a change in the formed size of the workpiece. Its control principle: Based on the precursor symptoms of instability defects that occur during the power spinning metal forming process, automatically adjust the main process parameter, the feed rate, of the power spinning forming to change the acting force of the spinning wheel on the metal forming, improve the stress and strain state in the plastic deformation zone, and thus improve the workpiece quality.

[0042] The optimization control of the present invention is mainly composed of an offline control system and an online control system, and its process is asFigure 3 As shown in the figure. The main process is as follows:

[0043] (1) Design of the offline control system. First, in the offline learning stage, a virtual machining environment is established. In the virtual environment, the power spinning and shear spinning processes and the entire springback process of the metal component are simulated. Integrating the actual machining conditions, the actual machining process of the metal component spinning forming is reproduced, and the forming defects such as non-conforming die sticking, local bulging, blank wobbling, and flange wrinkling that are prone to occur during the spinning forming of the metal component are presented. The machining state of the metal component is corresponded to the machining quality, machining accuracy, and low stress to obtain a reward value. A mapping relationship is established through deep learning with the feed rate, machining state, and reward value, and the deep learning strategy is updated through optimization by the reinforcement learning algorithm. After multiple trainings under different working conditions, a well-learned control model is obtained, and an optimized feed rate is obtained.

[0044] (2) Online control system. The trained control model is deployed to the motion controller of the spinning machine. Along with the real-time machining state of the spinning, it is updated online and the feed rate is optimized in real time, so as to achieve real-time control.

Claims

1. A self-adaptive control method for high-force spinning of metal components, characterized in that: The following steps are involved: Step 1: Establish a virtual machining environment based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool, and the springback model; Step 2: In a virtual processing environment, the spinning process and forming defects of metal components are reproduced by simulating the strong spinning, shearing and springback processes of metal components. The processing state, processing quality and stress state of metal components are monitored in real time by deploying sensor simulation modules, and the corresponding reward value is obtained through the preset evaluation function combined with the processing quality index. Step 3: Taking feed rate, processing status, and reward value as input, a mapping relationship is established through a deep learning algorithm, and then the deep learning strategy is updated through a reinforcement learning algorithm. After multiple trainings under different working conditions, a trained control model is obtained, and the optimized feed rate is verified and output; Step 4: Deploy the trained control model to the motion controller of the spinning machine. Through real-time data acquisition, the real-time processing status data of the spinning machine is obtained. Based on the real-time processing status data, the real-time control of the spinning machine is achieved through online updating and optimization of the feed rate.

2. The method for adaptively controlling the high-force spinning of metal components according to claim 1, characterized in that: The virtual machining environment is established based on the constitutive model of the metal material of the metal component, the dynamic characteristic model of the machine tool and the springback model, including: According to the stress-strain relationship of metal materials during the spinning process, the constitutive model of metal materials is constructed; according to the vibration and stiffness of the machine tool, the dynamic characteristic model of the machine tool is established; a rebound model is established to predict the springback behavior of the workpiece after spinning; through the constitutive model of metal materials, the dynamic characteristic model of the machine tool and the springback model, a complete virtual processing environment is formed.

3. The method for adaptively controlling the high-force spinning of metal components according to claim 2, characterized in that: In the virtual processing environment, the spinning process and forming defects of metal components are reproduced by simulating the strong spinning, shearing and springback process of metal components; the processing state, processing quality, processing accuracy and low stress state of metal components are monitored in real time by deploying sensor simulation modules, and the corresponding reward value is obtained by combining the preset evaluation function with the processing quality index, including: Through sensors in the virtual environment, various state parameters of metal components during the spinning process are monitored in real time to obtain processing simulation results; based on the simulation results, the processing quality of the metal components is evaluated to obtain processing quality evaluation results; based on the stress state of the metal components during the processing, the monitoring results of the low stress state are obtained; based on the processing state, the processing quality evaluation results and the monitoring results of the low stress state, the corresponding reward value is calculated.

4. The self-adaptive control method for high-force spinning of metal components according to claim 3, characterized in that: The feed rate, processing state, and reward value are input, a mapping relationship is established through a deep learning algorithm, and the optimization is performed through a reinforcement learning algorithm, the deep learning strategy is updated, and after multiple trainings under different working conditions, a trained control model is obtained, and the optimized feed rate is output, including: A deep learning algorithm is selected, and a deep learning model is constructed with feed rate, processing status, and reward value as input. The mapping relationship between input and output is learned through training data. The deep learning model is optimized using a reinforcement learning algorithm, and multiple trainings are performed under different working conditions to complete the training of the control model. The deep learning algorithm is one of a convolutional neural network and a recurrent neural network.

5. The self-adaptive control method for high-force spinning of metal components according to claim 4, characterized in that: The trained control model is deployed to the motion controller of the spinning machine, and the real-time processing status data of the spinning machine is obtained through real-time data acquisition. According to the real-time processing status data, the real-time control of the spinning machine is realized by online updating and optimizing the feed rate, including: The trained control model is deployed to the spinning machine motion controller. During the actual processing, the motion state of the spinning machine and the processing state of the metal components are collected in real time. The collected real-time data is input into the control model, and the feed rate is updated and optimized online according to the current processing state. The optimized feed rate is output to the spinning machine motion controller to realize real-time spinning machine motion control.

6. A self-adaptive control system for high-force spinning of metal components, using a self-adaptive control method for high-force spinning of metal components as claimed in any one of claims 1 to 5, characterized in that: It includes a spinning machine motion controller, a metal component high-force spinning simulation module, a data acquisition module, a data processing module and a communication module; The spinning machine motion controller, the metal component strong spinning simulation module, the data acquisition module and the communication module are respectively connected to the data processing module.

Images