An innovative smart monitoring system for DED-LB/MW process stability based on machine learning techniques Jon Flores, Itziar Cabanes, Aitor Gutierrez, Kerman Lopez de Calle, Oscar Gonzalo, Eva Portillo International Journal of Advanced Manufacturing Technology, 2025 Within additive manufacturing (AM) processes, the wire laser metal deposition (DED-LB/MW) technique optimizes the benefits of this technology, including reduced material consumption, cost, and production time. However, it still lacks sufficient robustness to ensure a continuous and consistent high-quality process for the manufacture of three-dimensional components. Process stability is determined by the ability to maintain constant conditions in the transfer of the wire feed material into the melt pool, which can be affected by any variation in the process. This work presents a smart monitoring system for online diagnosis of the DED-LB/MW process, facilitating the automated and immediate identification of the stability level during manufacturing. The proposed methodology relies on image-based monitoring and the integration of a convolutional neural network (CNN) model trained to identify the categorized conditions in the material deposition process. This study introduces an innovative method to convert the discrete output of the model into a continuous stability indicator, offering a more detailed and accurate depiction of the transition between stable and unstable process states. The proposed smart monitoring system is rigorously validated through two representative case studies, demonstrating its ability to adapt to variations in standoff distance—a critical challenge in achieving consistent layer growth in 3D manufacturing. By leveraging the power of CNN models for online monitoring in DED-LB/MW, this work not only advances the state of the art in process monitoring but also establishes a solid foundation for the integration of closed-loop control systems in future additive manufacturing technologies.
Multiple data management for quality analysis in DED─LB⁄MW Jon Flores, David Alberdi, Itziar Cabanes, Oscar Gonzalo, Carlos Soriano International Journal of Advanced Manufacturing Technology, 2025 Directed energy deposition using laser beam and metal wire (DED─LB⁄MW) offers notable advantages for additive manufacturing (AM) of large metal components but remains limited by process variability and challenges in ensuring internal quality. This study presents an integrated multi-sensor monitoring approach combined with a spatiotemporal data fusion framework to enhance process understanding. Key thermal, geometric, and visual signals were acquired during deposition using a coaxial pyrometer, an optical coherence tomography (OCT) sensor, and a welding camera, and subsequently synchronized with robot trajectory data. Post-process computed tomography (CT) scans were used to incorporate internal porosity information. The fused dataset was mapped onto the manufactured geometry through a layer-wise 3D representation, enabling the spatial correlation of process features and defect formation. Results demonstrate that thermal accumulation, surface irregularities, and deficient bead overlap significantly affect deposition stability and internal quality. The proposed methodology supports advanced process analysis and lays the groundwork for data-driven control and quality assurance strategies in DED─LB⁄MW.
Preface Lecture Notes in Production Engineering, 2018
Case Study 1.2: Turning of Low Pressure Turbine Casing Oscar Gonzalo, Jose Mari Seara, Eneko Olabarrieta, Mikel Esparta, Iker Zamakona, Manu Gomez-Korraletxe, José Alberto de Dios Lecture Notes in Production Engineering, 2018
Machinability of Al-SiC metal matrix composites using WC, PCD and MCD inserts Oscar Gonzalo, Jokin Beristain, Alejandro Sandá Revista De Metalurgia, 2014 The aim of this work is the study of the machinability of aluminium-silicon carbide Metal Matrix Composites (MMC) in turning operations. The cutting tools used were hard metal (WC) with and without coating, different grades and geometries of Poly-Crystalline Diamond (PCD) and Mono-Crystalline Diamond (MCD). The work piece material was AMC225xe, composed of aluminium-copper alloy AA 2124 and 25% wt of SiC, being the size of the SiC particles around 3 μm. Experiments were conducted at various cutting speeds and cutting parameters in facing finishing operations, measuring the surface roughness, cutting forces and tool wear. The worn surface of the cutting tool was examined by Scanning Electron Microscope (SEM). It was observed that the Built Up Edge (BUE) and stuck material is higher in the MCD tools than in the PCD tools. The BUE acts as a protective layer against abrasive wear of the tool.
Concept design of a 5-axis portable milling machine for the in-situ processing of large pieces Proceedings of the 13th International Conference of the European Society for Precision Engineering and Nanotechnology Euspen 2013, 2013
Surface and mechanical properties of laser melted maraging steel Proceedings of the Euro International Powder Metallurgy Congress and Exhibition Euro Pm 2009, 2009
Computation and experimental validation of the oblique cutting process in AL2024-T4 and AISI 4340 Computational Plasticity Fundamentals and Applications Proceedings of the 8th International Conference on Computational Plasticity Complas Viii, 2005
