Today
Bookmark and Share    
Full Text
Peer Reviewed
See detailLaser welding of copper to aluminum with spiral trajectory and identification of excessive aluminum melting
Mathivanan, Karthik UL; Plapper, Peter UL; Mathivanan, Karthik UL

in Journal of Laser Applications (2022)

Laser welding of copper and aluminum is challenging due to the formation of complex intermetallic phases. Only a defined amount of Al and Cu can be melted because of the limited solubility of Al–Cu ... [more ▼]

Laser welding of copper and aluminum is challenging due to the formation of complex intermetallic phases. Only a defined amount of Al and Cu can be melted because of the limited solubility of Al–Cu systems. Finding the optimum melting is critical for a strong joint. Optical emission during the welding process contains the metal vapor of Al metal that is being welded. This is a good indicator for monitoring the welding process. This research paper focuses on the optical emission of Al from the bottom sheet during welding of Cu (top) and Al (bottom) in overlapped configuration for a spiral trajectory. The emitted signal in the range of 395 nm (±3 nm) from the bottom sheet of aluminum is used to identify excessive Cu–Al welding. The tensile shear strength, microstructure, and welding signal in the time domain for optimum and excessive weld conditions are investigated. In this study, a technique using a photodiode is shown to identify the excessive melting of Al during the welding process in real time. [less ▲]

Detailed reference viewed: 51 (2 UL)
Full Text
Peer Reviewed
See detailA RNN-Based Hyper-Heuristic for Combinatorial Problems
Kieffer, Emmanuel UL; Duflo, Gabriel UL; Danoy, Grégoire UL et al

in A RNN-Based Hyper-Heuristic for Combinatorial Problems (2022)

Designing efficient heuristics is a laborious and tedious task that generally requires a full understanding and knowledge of a given optimization problem. Hyper-heuristics have been mainly introduced to ... [more ▼]

Designing efficient heuristics is a laborious and tedious task that generally requires a full understanding and knowledge of a given optimization problem. Hyper-heuristics have been mainly introduced to tackle this issue and are mostly relying on Genetic Programming and its variants. Many attempts in the literature have shown that an automatic training mechanism for heuristic learning is possible and can challenge human-based heuristics in terms of gap to optimality. In this work, we introduce a novel approach based on a recent work on Deep Symbolic Regression. We demonstrate that scoring functions can be trained using Recurrent Neural Networks to tackle a well-know combinatorial problem, i.e., the Multi-dimensional Knapsack. Experiments have been conducted on instances from the OR-Library and results show that the proposed modus operandi is an alternative and promising approach to human- based heuristics and classical heuristic generation approaches. [less ▲]

Detailed reference viewed: 51 (8 UL)