Inline closed-loop control of bending angles with machine learning supported springback compensation - Research paper

Aug 3, 2026

Closed-loop control of product properties is increasingly important in metal forming because it helps counter unavoidable uncertainties in semi-finished parts and process environments. By recording the controlled product properties inline and adapting the process through targeted changes of manipulated variables, closed-loop forming can reduce tolerances and improve consistency, robustness, and reliability. This paper demonstrates an air bending process where camera-based control determines bending angle and springback is compensated within a stroke using force-signal recording and machine-learning prediction of the loaded bending angle.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Keywords

Introduction

Uncertainty in metal forming

Closed-loop control concept and requirements

Servo presses and flexibility

Machine learning in manufacturing

Related work on bending angle and springback control

Data-driven and neural-network models

Control strategies requiring load-unloading cycles

Analytical and model-based approaches

International Journal of Material Forming (2024) 17:8 [https://doi.org/10.1007/s12289-023-01802-y](https://doi.org/10.1007/s12289-023-01802-y) Inline closed-loop control of bending angles with machine learning supported springback compensation Dirk Alexander Molitor1 · Viktor Arne1 · Christian Kubik1 · Gabriel Noemark1 · Peter Groche1 Received: 17 July 2023 / Accepted: 23 November 2023 / Published online: 13 December 2023 © The Author(s) 2023 Abstract Closed-loop control of product properties is becoming increasingly important in forming technology research and enables users to counteract unavoidable uncertainties in semi-ﬁnished product properties and process environments. Therefore, closedloop controlled forming processes are considered to have the potential to reduce tolerances on desired product properties, resulting in consistent qualities. The achievement of associated increases in robustness and reliability is linked to enormous requirements, which in particular include the inline recording of the product properties to be controlled and the subsequent adaptation of the process control through the targeted derivation of manipulated variables. The present paper uses the example of an air bending process to show how the bending angle can be controlled camera-based and how springback can be compensated within a stroke by recording force signals and subsequently predicting the loaded bending angle using machine learning algorithms. The results show that the combined application of camera-based control and machine learning assisted springback compensation leads to highly accurate bending angles, whereby the results strongly depend on the machine learning algorithms and associated data transformation processes used. Keywords Closed-loop control · Deep learning · Sheet metal forming Introduction Uncertainty plays a decisive role in metal forming and signiﬁcantly inﬂuences product quality and its variance. [1] Manufacturers strive for consistent quality and thus repeatable production of selected product properties. For this reason, manufacturers try to keep their processes, semi-ﬁnished products, tools and associated environmental inﬂuences as stationary as possible over the production time, which leads to reliable products but not to robust processes. Unavoidable uncertainties resulting, for example, from tool wear [2], machine stiffness [3] or variations in semi-ﬁnished product properties [4] must either be monitored individually or, if not taken into account, are reﬂected in ﬂuctuating product properties [5] . A promising approach that has been increasingly explored in academic research in recent years is aimed at closed-loop control of product properties [6] . The basic idea B Dirk Alexander Molitor [dirk.molitor@ptu.tu-darmstadt.de](dirk.molitor@ptu.tu-darmstadt.de) 1 Institute for Production Engineering and Forming Machines, Technical University of Darmstadt, Otto-Berndt-Straße 2, Darmstadt D-64287, Germany of this approach is to equip forming machines and corresponding tools with sensors and actuators, whereby product properties can be recorded stroke-by-stroke or inline and subsequently manipulated. However, the realisation of such closed-loop controls is highly challenging. Product properties must be sensed or observed with sufﬁcient frequency and models between the drives ofthe machine and product properties must be available that allow a controlled manipulation. The quality of these closed-loop controls is determined by their accuracy and ﬂexibility. The increasing use of servo presses with freely programmable drive motion paths [7] or task space controlled ram movements [8] has signiﬁcantly increased the accuracy and ﬂexibility of forming processes in recent years . For example, Groche et al. investigate ina deep-drawing process on a 3D Servo Press how multiple degrees of freedom of the press can be used to manipulate wrinkles and how different control strategies inﬂuence the process [9] . T

FAQs

Why is closed-loop control important in metal forming?

Uncertainties strongly affect product quality and its variance. Closed-loop control can reduce tolerances on desired properties by recording inline and adapting manipulated variables based on measured outcomes.

How does the paper control bending angle and compensate springback?

Camera-based recording controls the bending angle, while springback compensation is performed within a stroke by recording force signals and using machine-learning algorithms to predict the loaded bending angle.

What does the paper say about factors affecting performance?

Highly accurate bending angles are achieved, but results depend strongly on the chosen machine learning algorithms and the associated data transformation processes.

Aug 3, 2026

Uploaded by River Wang