Study on the Intelligent Control of Springback in Stretch Bending Process Based on Neural Networks

2006 ◽  
pp. 604-607
Author(s):  
Yong Jun Wang ◽  
Jun Biao Wang ◽  
Sheng Min Wei ◽  
Jiang Jun Jiang
Keyword(s):  
Neural Networks ◽  
Intelligent Control ◽  
Stretch Bending ◽  
Bending Process

Study on the Intelligent Control of Springback in Stretch Bending Process Based on Neural Networks

2006 ◽  
Vol 532-533 ◽  
pp. 604-607
Author(s):  
Yong Jun Wang ◽  
Jun Biao Wang ◽  
Sheng Min Wei ◽  
Jiang Jun Jiang
Keyword(s):  
Mechanical Properties ◽  
Neural Networks ◽  
Friction Coefficient ◽  
Intelligent Control ◽  
Process Parameters ◽  
Control Method ◽  
Friction Condition ◽  
Whole Process ◽  
Stretch Bending ◽  
Bending Process

In extrusion stretch bending process, there are many factors which affect springback of the workpiece such as mechanical properties of the material, friction condition and process parameters. The springback of same batch of extrusion is different at same forming parameters because of the variation of the mechanical properties of the material and the friction condition. A method of intelligent control of springback in stretch bending process is proposed by using ANN(artificial neural networks). The online identification model of the mechanical properties of the material and friction coefficient and the online prediction control model of springback of workpiece in stretch bending process are established by using ANN ,which are trained by the data of analysis calculation. It realizes the intelligent control on springback of stretch bending to online identify the material properties and friction coefficient and predict springback and adjust process parameters dynamically through the whole process of stretch bending. The results from the experiment state that the intelligent control method can suit the variation of mechanical properties of material and friction condition and improve the geometry precision.

Research on springback characteristic in flexible multi-points 3D stretch-bending process

2021 ◽  
pp. 095440542110284
Author(s):  
Song Gao ◽  
Tonggui He ◽  
Qihan Li ◽  
Yingli Sun ◽  
Jicai Liang
Keyword(s):  
Simulation Model ◽  
High Efficiency ◽  
Vertical Direction ◽  
Factors Affecting ◽  
Stretch Bending ◽  
Element Simulation ◽  
Bending Process ◽  
Bending Radius ◽  
Aluminum Profiles ◽  
Bending Sequence

The problem of springback is one of the most significant factors affecting the forming accuracy for aluminum 3D stretch-bending parts. In order to achieve high-efficiency and high-quality forming of such kind of structural components, the springback behaviors of the AA6082 aluminum profiles are investigated based on the flexible multi-points 3D stretch-bending process (3D FSB). Firstly, a finite element simulation model for the 3D FSB process was developed to analyze the forming procedure and the springback procedure. The forming experiments were carried out for the rectangle-section profile to verify the effectiveness of the simulation model. Secondly, the influence of tension on springback was studied, which include the pre-stretching and the post-stretching. Furthermore, the influences of the bending radius and bending sequence are revealed. The results show that: (1) The numerical model can be used to evaluate the effects of bending radius and process parameters on springback in the 3D FSB process effectively. (2) The pre-stretching has little effect on the horizontal springback reduction, but it plays a prominent role in reducing the springback in the vertical direction. (3) The increase of bending deformation in any direction will lead to an increase of springback in its direction and reduce the springback in the other direction. Besides, it reduces the relative error in both directions simultaneously. This research established a foundation to achieve the precise forming of the 3D stretch-bending parts with closed symmetrical cross-section.

MODELING OF NEURAL PREDICTIVE CONTROL OF IRRIGATION MACHINES

2021 ◽  
pp. 14-22
Author(s):  
G. N. KAMYSHOVA ◽  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Real Time ◽  
Predictive Control ◽  
Intelligent Control ◽  
Machine Learning Algorithms ◽  
Intelligent Control Systems ◽  
Artificial Neural ◽  
The One ◽  
Neural Predictive Control

The purpose of the study is to develop new scientific approaches to improve the efficiency of irrigation machines. Modern digital technologies allow the collection of data, their analysis and operational management of equipment and technological processes, often in real time. All this allows, on the one hand, applying new approaches to modeling technical systems and processes (the so-called “data-driven models”), on the other hand, it requires the development of fundamentally new models, which will be based on the methods of artificial intelligence (artificial neural networks, fuzzy logic, machine learning algorithms and etc.).The analysis of the tracks and the actual speeds of the irrigation machines in real time showed their significant deviations in the range from the specified speed, which leads to a deterioration in the irrigation parameters. We have developed an irrigation machine’s control model based on predictive control approaches and the theory of artificial neural networks. Application of the model makes it possible to implement control algorithms with predicting the response of the irrigation machine to the control signal. A diagram of an algorithm for constructing predictive control, a structure of a neuroregulator and tools for its synthesis using modern software are proposed. The versatility of the model makes it possible to use it both to improve the efficiency of management of existing irrigation machines and to develop new ones with integrated intelligent control systems.

Intelligent Control Experiment Study for V-Shape Free Bending of Wide Sheet Metal

2008 ◽  
Vol 575-578 ◽  
pp. 186-191
Author(s):  
Jun Zhao ◽  
Chun Jian Su ◽  
Ying Ping Guan
Keyword(s):  
Sheet Metal ◽  
Intelligent Control ◽  
Control Experiment ◽  
Bending Process ◽  
Prediction And Control ◽  
The Neural Networks ◽  
Die Material ◽  
Object Of Study ◽  
And Control ◽  
Free Bending

The main problem in bending process of sheet metal is that it is difficult to control bending springback accurately. Springback produced from the unloading of bending makes the shape and size incongruent between bending workpiece and working portion of die. Because the final shape of bending workpiece is related with the whole deformation process, the geometric parameter of die, material performance parameter will have great effect on springback. Therefore, the springback problem is very complicated and the prediction and control of springback is the key to improve the accuracy of bending workpiece. Taking the V free bending of wide sheet as an object of study, the neural networks technology and data acquisition system based on LabVIEW are used to establish intelligent control experiment system for V free bending of wide sheet metal. The control accuracy of system is high and it provides the basis for the realization of intelligent control for V-shape free bending of wide sheet metal in practice in future.

Forming Limits Under Stretch-Bending Through Distortionless and Distortional Anisotropic Hardening

2018 ◽  
Vol 140 (12) ◽  
Author(s):  
Ji He ◽  
Bin Gu ◽  
Yongfeng Li ◽  
Shuhui Li
Keyword(s):  
Detailed Comparison ◽  
Forming Limit ◽  
Anisotropic Hardening ◽  
Hardening Model ◽  
Stretch Bending ◽  
Bending Process ◽  
Hardening Models ◽  
Out Of Plane ◽  
Forming Limit Curves ◽  
Insight Into

The necking behavior of sheet metals under stretch-bending process is a challenge for the forming limit prediction. State-of-the-art forming limit curves (FLCs) allow the prediction under the in-plane stretching but fall short in the case under out-of-plane loading condition. To account for the bending and straightening deformation when sheet metal enters a die cavity or slide along a radius, anisotropic hardening model is essential to reflect the nonproportional loading effect on stress evolution. This paper aims to revisit the M-K analysis under the stretch-bending condition and extend it to accommodate both distortionless and distortional anisotropic hardening behavior. Furthermore, hardening models are calibrated based on the same material response. Then the detailed comparison is proposed for providing better insight into the numerical prediction and necking behavior. Finally, the evolution of the yield surface and stress transition states is examined. It is found that the forming limit prediction under stretch-bending condition through the M-K analysis strongly depends on the employed anisotropic hardening model.

Sign in / Sign up

Export Citation Format

Share Document