Bending control method for metal plate processing
By integrating equipment and closed-loop control logic, and utilizing AI prediction and wireless communication modules, the problem of low bending and springback accuracy in metal sheet processing has been solved, achieving high-precision, low-scrap, and high-energy-consumption metal sheet processing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ANHUI LUHONG INTELLIGENT EQUIPMENT CO LTD
- Filing Date
- 2026-03-02
- Publication Date
- 2026-05-05
AI Technical Summary
The lack of coordination in existing bending equipment in sheet metal processing results in low bending springback accuracy. Traditional control methods have limited precision and cannot adapt to multi-variety, small-batch production. Equipment monitoring relies on local systems, which is inefficient.
The system employs integrated equipment, combining AI prediction modules and wireless communication modules to form a closed-loop control logic. By predicting the springback amount through AI and compensating and adjusting it in real time, it achieves remote monitoring and collaborative control, including system initialization, AI-assisted feeding, intelligent bending and springback compensation, adaptive adjustment of tool retraction, and closed-loop optimization.
It improves bending angle accuracy to within ±0.2°, reduces scrap rate by 30%, increases equipment utilization by 15%, reduces energy consumption by 20%, and is suitable for processing sheet materials of different sizes and materials.
Smart Images
Figure CN121972541A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of metal sheet processing technology, and in particular to a bending control method for metal sheet processing. Background Technology
[0002] Sheet metal bending is a key process in sheet metal forming, involving multiple steps such as feeding, clamping, bending, and tool retraction. In existing technologies, bending equipment typically employs discrete control, with each component operating independently and lacking coordination.
[0003] Bending springback is a core issue affecting forming accuracy, causing the actual bending angle to deviate from the target angle (often exceeding ±1°). Traditional control methods, such as PID regulation or empirical compensation, have limited accuracy and cannot adapt to multi-variety, small-batch production. Furthermore, equipment monitoring relies on local systems, and fault diagnosis and maintenance require on-site operation, resulting in low efficiency. Summary of the Invention
[0004] The purpose of this invention is to provide a bending control method that uses an AI prediction module to predict the bending springback in real time and adjust bending parameters in advance, combined with wireless communication to achieve remote monitoring and collaborative control. This method solves the problems of low springback control accuracy and isolated equipment operation in existing technologies.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a bending control method for metal sheet processing, the method being implemented based on integrated equipment, the integrated equipment including a tool retraction assembly, a feeding rotary pressure arm device, and a crankshaft connecting rod bending structure; the method integrates an AI prediction module and a wireless communication module through a control platform to form closed-loop control logic, including the following steps: Step S1: System initialization and data acquisition. Connect to the cloud database via the wireless communication module to download historical bending data, and use sensors to collect the initial parameters of the sheet material. Step S2: AI-assisted feeding and clamping control, fine-tuning the feeding position and clamping force based on the springback compensation amount output by the AI prediction module; Step S3: Intelligent bending and springback compensation control, adjusting the bending target angle based on AI prediction results, and driving the crankshaft connecting rod bending mechanism to apply force; Step S4: Tool retraction and adaptive adjustment, adjust the tool pressing length according to the prediction results, and monitor the equipment status through wireless communication; Step S5: Closed-loop optimization and remote maintenance, using post-bending data to update the AI model, supporting remote diagnosis and parameter adjustment.
[0006] Preferably, the AI prediction module uses a deep learning model. The model input includes the material, thickness, bending angle, and historical springback data of the sheet metal, and the model output is the springback compensation amount. The deep learning model is an LSTM or CNN architecture and supports online learning.
[0007] Preferably, the wireless communication module transmits data through an IoT platform and encrypts the data using the MQTT protocol; the wireless communication module is integrated into the calibration table of the feeding rotary pressure arm device and the sensors of the crankshaft connecting rod bending mechanism, and uploads angle, displacement and pressure data in real time.
[0008] Preferably, the AI-assisted feeding and clamping control in step S2 specifically includes: The feeding device adjusts its position based on the feeding offset predicted by AI, and achieves misalignment through the deflection mechanism of the main pressure arm; The clamping mechanism drives the pressure head through a three-cylinder system, and the AI module dynamically adjusts the clamping force based on real-time pressure data. During the feeding process, the wireless communication module synchronizes the calibration data to the cloud platform.
[0009] Preferably, the intelligent bending and springback compensation control in step S3 specifically includes: The bending mechanism receives AI instructions and adjusts the bending target angle to the sum of the original angle and the springback compensation amount; The crankshaft connecting rod drive mechanism operates in coordination with drive motors one through four, causing the bending fixed seat to float up and down. After bending, the sensor measures the actual angle, and the springback error data is used for model optimization.
[0010] Preferably, the tool retraction and adaptive adjustment in step S4 specifically include: The tool retraction assembly drives the tool pressing mechanism to move through cylinder one and cylinder two, and adjusts the tool pressing length based on the springback trend predicted by AI. The inflation control of the self-made miniature cylinders one and two is synchronized via a wireless module, enabling the rapid disassembly or expansion of the pressure knife body.
[0011] Preferably, the closed-loop optimization and remote maintenance described in step S5 specifically includes: After each batch of AI models is processed, they are retrained based on new data and the local model is updated via wireless network. The cloud platform provides visual reports, and operators can remotely adjust equipment parameters via a wireless interface.
[0012] Preferably, the method is applicable to the bending of metal sheets, with sheet size ranging from 10cm to 50cm, and bending angle deviation controlled within ±0.2°.
[0013] A control system for a bending control method in metal sheet processing, the control system comprising: The AI prediction module, deployed on the control platform, is used for rebound prediction. Wireless communication module for data transmission and remote monitoring; Sensor arrays, integrated into feeding, bending, and unloading equipment, are used for real-time data acquisition; The actuator assembly, including cylinders and drive motors, is used to drive the various components to move.
[0014] Compared with the prior art, the beneficial effects of the present invention are: Improved precision: Springback compensation keeps bending angle deviation within ±0.2° (compared to over ±1° with traditional methods), reducing scrap rate by 30%. Efficiency optimization: Wireless remote monitoring reduces manual intervention and increases equipment utilization by 15%; Energy saving and reliability: The crankshaft connecting rod energy-saving design reduces overall energy consumption by 20%, and AI prediction reduces excessive bending, extending equipment life; Highly adaptable: Supports sheets of different sizes (10cm to 50cm) and materials, adapting to complex bending processes. Attached Figure Description
[0015] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0016] Figure 1 This is a flowchart of a bending control method for metal sheet processing according to the present invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] Please see Figure 1 The present invention provides a technical solution: The core of this invention is: using a feeding device, bending mechanism, and tool retraction assembly as hardware, and integrating an AI prediction module and a wireless communication module into a control platform to form a closed-loop control logic of "sensing-prediction-compensation-monitoring". Specifically, it includes: AI prediction module: Based on deep learning models (such as LSTM or CNN), it analyzes the material, thickness, bending angle, and historical data of the sheet metal to output the springback compensation amount. The model training data comes from historical processing records and supports online learning.
[0019] Wireless communication integration: Connect device sensors and control units through an IoT platform (such as the MQTT protocol) to enable real-time data uploading, remote diagnostics, and parameter optimization.
[0020] Collaborative control process: Initialization → AI-assisted feeding → Intelligent bending and springback compensation → Adaptive tool retraction adjustment → Closed-loop optimization. Each step is deeply integrated with the existing equipment, for example: The feeding device fine-tunes its position based on AI predictions.
[0021] The bending mechanism receives AI instructions to adjust the bending angle.
[0022] The retraction assembly adjusts the pressing length based on the prediction results.
[0023] Example 1: System Initialization and Data Acquisition The control platform starts up, and the wireless communication module (using a 4G / 5G module) connects to the cloud database to download historical bending data (including springback records of similar sheet materials). At the same time, local sensors collect the initial parameters of the sheet material (2mm thickness, aluminum material).
[0024] AI prediction module integration: A pre-trained LSTM model is deployed in the control platform. Inputting sheet metal parameters and a 90° bending angle, the model outputs a predicted springback of 1.8°. Model weights are updated regularly via a cloud platform.
[0025] Wireless communication setup: The device is configured with an IoT node, and data is encrypted and transmitted to the cloud platform. Sensor data from the calibration station is uploaded in real time for remote monitoring.
[0026] Example 2: AI-assisted feeding and clamping control The feeding device is fine-tuned based on AI predictions: the cloud platform recommends a feeding offset of 0.5mm to avoid the impact of springback. During feeding, the main pressure arm is deflected by a cylinder, causing the clamping mechanism and the rotating mechanism to be misaligned, and the robot arm places the sheet material vertically.
[0027] Dynamic clamping mechanism: The main pressure arm presses down, and the cylinder-driven pressure head fixes the plate. The AI module adjusts the clamping force based on real-time pressure data to avoid excessive deformation. The wireless module uploads the pressure data, which the operator can view remotely via an app.
[0028] Example 3: Intelligent Bending and Springback Compensation Control The bending mechanism receives AI instructions: the control platform adjusts the target bending angle to 91.8° (90° + 1.8° compensation). The crankshaft connecting rod drive mechanism operates according to the compensation parameters, and the bending fixed seat drives the upper and lower bending blades to apply force.
[0029] Real-time feedback: After bending, the displacement sensor measures the actual angle as 89.9°, with a springback error of 0.1°. The data is uploaded to the cloud platform via a wireless module for online model learning.
[0030] Example 4: Tool Retraction and Adaptive Adjustment The retraction assembly adjusts the pressure blade length based on AI predictions: for the springback trend of the aluminum plate, the AI suggests expanding the pressure blade coverage area. The adjustment process is driven by cylinder one and cylinder two, and the inflation control of the self-made miniature cylinders one and two is synchronized via a wireless module.
[0031] Fault prediction: The cylinder status is monitored via wireless communication, and abnormal data triggers an alarm.
[0032] Example 5: Closed-loop optimization and remote maintenance After each batch of processing is completed, the AI model is retrained based on the new data and updated locally via wireless network. The cloud platform generates a report showing the bending accuracy trend.
[0033] Remote diagnostics: Operators can adjust the speed of the travel motor in document 2 via a wireless interface to optimize the feeding stroke.
[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A bending control method for processing metal sheets, characterized in that: The method is implemented based on integrated equipment, which includes a tool retraction assembly, a feeding rotary pressure arm device, and a crankshaft connecting rod bending structure. The method integrates an AI prediction module and a wireless communication module through a control platform to form closed-loop control logic, and includes the following steps: Step S1: System initialization and data acquisition. Connect to the cloud database via the wireless communication module to download historical bending data, and use sensors to collect the initial parameters of the sheet material. Step S2: AI-assisted feeding and clamping control, fine-tuning the feeding position and clamping force based on the springback compensation amount output by the AI prediction module; Step S3: Intelligent bending and springback compensation control, adjusting the bending target angle based on AI prediction results, and driving the crankshaft connecting rod bending mechanism to apply force; Step S4: Tool retraction and adaptive adjustment, adjust the tool pressing length according to the prediction results, and monitor the equipment status through wireless communication; Step S5: Closed-loop optimization and remote maintenance, using post-bending data to update the AI model, supporting remote diagnosis and parameter adjustment.
2. The bending control method for metal sheet processing according to claim 1, characterized in that: The AI prediction module uses a deep learning model. The model input includes the material, thickness, bending angle, and historical springback data of the sheet metal. The model output is the springback compensation amount. The deep learning model is an LSTM or CNN architecture and supports online learning.
3. The bending control method for metal sheet processing according to claim 1, characterized in that: The wireless communication module transmits data through an IoT platform and encrypts the data using the MQTT protocol. The wireless communication module is integrated into the calibration table of the feeding rotary pressure arm device and the sensors of the crankshaft connecting rod bending mechanism, and uploads angle, displacement and pressure data in real time.
4. The bending control method for metal sheet processing according to claim 1, characterized in that: The AI-assisted feeding and clamping control mentioned in step S2 specifically includes: The feeding device adjusts its position based on the feeding offset predicted by AI, and achieves misalignment through the deflection mechanism of the main pressure arm; The clamping mechanism uses a cylinder to drive the pressure head, and the AI module dynamically adjusts the clamping force based on real-time pressure data. During the feeding process, the wireless communication module synchronizes the calibration data to the cloud platform.
5. The bending control method for metal sheet processing according to claim 1, characterized in that: The intelligent bending and springback compensation control mentioned in step S3 specifically includes: The bending mechanism receives AI instructions and adjusts the bending target angle to the sum of the original angle and the springback compensation amount; The crankshaft connecting rod drive mechanism operates in coordination with drive motors one through four, causing the bending fixed seat to float up and down. After bending, the sensor measures the actual angle, and the springback error data is used for model optimization.
6. The bending control method for metal sheet processing according to claim 1, characterized in that: The tool retraction and adaptive adjustment mentioned in step S4 specifically include: The tool retraction assembly drives the tool pressing mechanism to move through cylinder one and cylinder two, and adjusts the tool pressing length based on the springback trend predicted by AI. The inflation control of the self-made miniature cylinders one and two is synchronized via a wireless module, enabling the rapid disassembly or expansion of the pressure knife body.
7. The bending control method for metal sheet processing according to claim 1, characterized in that: The closed-loop optimization and remote maintenance described in step S5 specifically include: After each batch of AI models is processed, they are retrained based on new data and the local model is updated via wireless network. The cloud platform provides visual reports, and operators can remotely adjust equipment parameters via a wireless interface.
8. The bending control method for metal sheet processing according to claim 1, characterized in that: The method is applicable to the bending of metal sheets, with sheet sizes ranging from 10cm to 50cm, and the bending angle deviation controlled within ±0.2°.
9. A control system for implementing the bending control method for metal sheet processing according to any one of claims 1-8, characterized in that: The control system includes: The AI prediction module, deployed on the control platform, is used for rebound prediction. Wireless communication module for data transmission and remote monitoring; Sensor arrays, integrated into feeding, bending, and unloading equipment, are used for real-time data acquisition; The actuator assembly, including cylinders and drive motors, is used to drive the various components to move.