Publication

Adaptive Laser Welding Control: A Reinforcement Learning Approach

Résumé

Despite extensive research efforts in the field of laser welding, the imperfect repeatability of the weld quality still represents an open topic. Indeed, the inherent complexity of the underlying physical phenomena prevents the implementation of an effective controller using conventional regulators. To close this gap, we propose the application of Reinforcement Learning for closed-loop adaptive control of welding processes. The presented system is able to autonomously learn a control law that achieves a predefined weld quality independently from the starting conditions and without prior knowledge of the process dynamics. Specifically, our control unit influences the welding process by modulating the laser power and uses optical and acoustic emission signals as sensory input. The algorithm consists of three elements: a smart agent interacting with the process, a feedback network for quality monitoring, and an encoder that retains only the quality critic events from the sensory input. Based on the data representation provided by the encoder, the smart agent decides the output laser power accordingly. The corresponding input signals are then analyzed by the feedback network to determine the resulting process quality. Depending on the distance to the targeted quality, a reward is given to the agent. The latter is designed to learn from its experience by taking the actions that maximize not just its immediate reward, but the sum of all the rewards that it will receive from that moment on. Two learning schemes were tested for the agent, namely S-Learning and Policy Gradient. The required training time to reach the targeted quality was 20 min for the former technique and 33 min for the latter.

À propos de ce résultat
Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.
Concepts associés (34)
Soudage
Le soudage est un procédé d'assemblage permanent qui assure la continuité de la matière à assembler ; la soudure est le résultat obtenu (mais le terme est souvent utilisé pour le procédé). Cette continuité est ce qui distingue le soudage d'autres techniques d'assemblage mécaniques (rivetage, boulonnage, ...) ou par adhésion (collage), ainsi que les techniques de brasage L'opération peut être appliquée aux métaux ainsi qu'aux matières thermoplastiques (voir Soudage de plastiques) et au bois (voir Soudage du bois).
Laser
thumb|250px|Lasers rouges (660 & ), verts (532 & ) et bleus (445 & ). thumb|250px|Rayon laser à travers un dispositif optique. thumb|250px|Démonstration de laser hélium-néon au laboratoire Kastler-Brossel à l'Université Pierre-et-Marie-Curie. Un laser (acronyme issu de l'anglais light amplification by stimulated emission of radiation qui signifie « amplification de la lumière par émission stimulée de radiation ») est un système photonique.
Diode laser
Une diode laser est un composant opto-électronique à base de matériaux semi-conducteurs. Elle émet de la lumière monochromatique cohérente (une puissance optique) destinée, entre autres, à transporter un signal contenant des informations sur de longues distances (dans le cas d'un système de télécommunications) ou à apporter de l'énergie lumineuse pour le pompage de certains lasers (lasers à fibre, laser DPSS) et amplificateurs optiques (OFA, Optical Fiber Amplifier).
Afficher plus
Publications associées (79)

Control of intermetallics in the interface in directed energy deposition of titanium on stainless steel 316L

Di Cui

Application of a single metal or alloy is often restricted by its properties from optimal combination of performance and cost. Therefore, there is a vast need of joining dissimilar metals for various applications in biomedical, aerospace, automobile and ma ...
EPFL2023

Single-frequency violet and blue laser emission from AlGaInN photonic integrated circuit chips

Tobias Kippenberg, Anat Siddharth

Chip-based, single-frequency and low phase-noise integrated photonic laser diodes emitting in the violet (412 nm) and blue (461 nm) regime are demonstrated. The GaN-based edge-emitting laser diodes were coupled to high-quality on-chip micro-resonators for ...
Optica Publishing Group2023

Third-harmonic generation monitoring of femtosecond-laser-induced in-volume functional modifications

Christian Enz, Yves Bellouard, Assim Boukhayma, Andrea Kraxner, Olivier Bernard

During the last two decades, ultrafast in-volume laser-based processing of transparent materials has emerged as a key 3D-printing method for manufacturing a variety of complex integrated photonic devices and micro-parts. Yet, identifying suitable laser pro ...
2023
Afficher plus
MOOCs associés (13)
Neuronal Dynamics - Computational Neuroscience of Single Neurons
The activity of neurons in the brain and the code used by these neurons is described by mathematical neuron models at different levels of detail.
Neuronal Dynamics - Computational Neuroscience of Single Neurons
The activity of neurons in the brain and the code used by these neurons is described by mathematical neuron models at different levels of detail.
Neuronal Dynamics 2- Computational Neuroscience: Neuronal Dynamics of Cognition
This course explains the mathematical and computational models that are used in the field of theoretical neuroscience to analyze the collective dynamics of thousands of interacting neurons.
Afficher plus