Christopher Taylor’s Post

🚀 Unveiling 𝗡𝗲𝘂𝗿𝗮𝗹𝗗𝗘𝗠: The Future of Industrial Flow Simulations! Imagine modeling complex granular flows and fluidized beds with accuracy and speed that traditional methods just can't match. From mining to pharma, agriculture to manufacturing, 𝗡𝗲𝘂𝗿𝗮𝗹𝗗𝗘𝗠 is set to revolutionize how industries model particulate systems with: • 📈 𝗥𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗦𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻: Achieving 2800x faster modeling compared to traditional methods, enabling instant feedback and decision-making. • 🔄 𝗠𝘂𝗹𝘁𝗶𝗽𝗹𝗲 𝗨𝘀𝗲 𝗖𝗮𝘀𝗲𝘀: Ideal for modeling bulk material storage in hoppers, fluidized bed reactors for material processing, and much more. • 🌐 𝗦𝗰𝗮𝗹𝗮𝗯𝗹𝗲 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴: Our multi-branch neural operators let us simulate massive particle counts and fluid interactions, enhancing accuracy even for transient phenomena. Thank you to our brilliant researchers, Benedikt Alkin, Tobias Kronlachner , Samuele Papa, Thomas Lichtenegger, Johannes Brandstetter and the entire NXAI team. Your dedication has opened new doors in industrial engineering! Explore more 1. on our project page https://lnkd.in/dPx9gsxP and 2. in our paper https://lnkd.in/dpt9mXqH Join us as we push the boundaries of simulations!

Nanoparticle aggregates often have a fractal structure, formed during synthesis. In our recent work, we show the influence of Van der Waals forces on nanoparticle cluster formation. To this end, we built an off-grid 3D model to simulate aggregate growth under various conditions. We found that increasing the attractive Van der Waals forces resulted in clusters with more open structures with lower fractal dimensions. Moreover, we demonstrate that 2D models do not capture this effect, while 3D models do. I want to thank all the co-authors for their respective contributions to this work! The article can be found here: https://lnkd.in/eQEfPxEJ Maximilian Theisen Ankur Deep Bordoloi Samir Salameh Gabrie Meesters Ruud van Ommen ARC CBBC Pro2Tech | TU Delft Process & Product Technology Institute

📣 Join the talk and get updated on recent innovations in the field of particle technology.

What's new in brake technology? The µ-Symposium on 25 October 2024 in Chemnitz offers the opportunity for a final update this year. The conference will start with the topic of the effectiveness of metal sulphides in brake linings (Tribotecc). The middle section is dedicated to different approaches in methodology: map-based prediction of commercial vehicle brake emissions (Link Engineering) and digitally supported product development of fine dust-optimised brake systems using the example of ICE and BEV (HPL Technologies). The use of AI in visual condition monitoring allows us to gain us an insight into the field of rail vehicle technology (DB Cargo). Research topics form the last block of presentations: Particle-reinforced aluminium aluminium matrix materials as alternative friction linings (TU Chemnitz) and data science and brake science and brake modelling (SiDyS). For further information and to register, please visit https://lnkd.in/eQ5g3AEj

MIT researchers, using NSF ACCESS-funded resources at SDSC, have developed a cutting-edge #machinelearning method to analyze chemical patterns in metallic alloys. This breakthrough in computational models offers unprecedented insights. https://ow.ly/8hwy50SxwzU #deeplearning #artificialintelligence

The Minerals, Metals & Materials Society has been a great partner in developing the vision for Metamorphic Manufacturing. They gathered an internationally respected expert group in 2019 to develop a foundational study Metamorphic Manufacturing: Shaping the Future of On-Demand Components.  Another expert group was assembled to consider the core issues of how data, artificial intelligence, machine control and component qualifications will shape this emerging industry. The new report was released this week: A Revolution in Digital Manufacturing: Integrating Machines, Robotics, Artificial Intelligence, and Forming Technologies.  This supports core elements of the HAMMER-ERC mission. https://lnkd.in/gnW7Kb-T Glenn Daehn, Wei Chen, Tony Schmitz, Robert Gao Jian Cao, John Lewandowski, Jag Sankar

🚀 Proud to announce the Special Issue of ACSE, curated by Simona Perotto, Nicola Ferro and myself: “Frontiers in Methods and Models in Engineering: State-of-the-Art and Emerging Solutions” This issue showcases cutting-edge research on mathematical models, numerical methods, and data-driven approaches to solve challenges in materials science, energy technology, and sustainability. Key contributions include: 🌱 Smart agriculture with satellite data and AI 🔋 Physics-Informed Neural Networks for lithium-ion battery modeling ☀️ Heat transfer optimization in solar power receivers with POD 🌿 Ecological modeling with PDEs 🛠 Advances in peridynamics simulations Discover how computational science is driving engineering innovation: https://lnkd.in/dAw--2C8 #Sustainability #MathematicalModeling #NumericalMethods #DataDriven

The ecological and economical production of high-quality cold forged parts requires a high level of process reliability. The measurement of eccentric forces during the forming process plays an important role for the subsequent product quality and at the same time represents a major challenge. The Fraunhofer IST is therefore developing integrated and wear-resistant thin-film sensors to enable in-situ data acquisition. In combination with the use of artificial intelligence (AI), the measured data can be used to predict process conditions and product quality. 🎯🔧 A research team from Fraunhofer IST and the Institute for Metal Forming Technology (IFU) of the Universität Stuttgart, where Martin Rekowski conducted the investigations, has just published great work on thin-film sensors for force measurement in cold forging entitled ”Thin-film sensors for data-driven concentricity prediction in cup backward extrusion”. The paper introduces a novel piezoelectric thin-film sensor to detect geometric imperfections such as concentricity deviations in cold forming. These deviations are detected by measuring the eccentric load caused by the elastic deformation of the die. By processing the measured force data with a support vector regression algorithm, the concentricity deviations of the manufactured parts can be efficiently predicted. The publication also discusses various sensors and measurement methods used in metal forming to collect process data and indirectly measure concentricity deviations.📊🔍 The thin film sensor offers high flexibility in design and can be integrated directly into the active process zone. 💪🌟 You want to dive into the details? You can find the full open access paper here:  👉https://lnkd.in/eyvdu-CT Congratulations to the brilliant research team behind this achievement:  Martin Rekowski, Karl C. Grötzinger, Anna Schott and Mathias Liewald   #thinfilmsensors #publication #coldforming #sensortechnology #machinelearning

Publication Alert 📢 Excited to share our latest research article, "Feature-Driven Density Prediction of Maraging Steel Additively Manufactured Samples Using Supervised Machine Learning," now published on IEEE Xplore. This study utilizes default machine input parameters (laser power and speed), an integrated pyrometer sensor, and physics-based features in Laser Powder Bed Fusion (LPBF) to accurately predict the density of maraging steel samples produced with the Aconity3D GmbH printer. The pyrometer sensor captures real-time thermal emissivity data, which enables precise monitoring of melt pool conditions, supporting Machine Learning models in predicting part quality and consistency for enhanced AM control and optimization. This feature-driven approach with pyrometer data creates a strong framework for quality monitoring and lays the groundwork for future advances in layer-wise quality predictions in AM. Special thanks to my supervisor, Raghavan Nagarajan, for his invaluable support and guidance, and my mentor Shaista Hussain. I would like to thank John kgee Ong and Qing Yang Tan for their continuous support on data collection from pyrometer sensor setup. #LaserPowderBedFusion #AconityMINI3D #PyrometerSensor #MaragingSteel #MachineLearning #QualityMonitoring Read more here: 

𝗧𝗿𝗮𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 𝗰𝗵𝗲𝗺𝗶𝗰𝗮𝗹 𝗴𝗮𝘀 𝘀𝗲𝗻𝘀𝗼𝗿𝘀 𝗳𝗮𝗰𝗲 𝗮 𝗯𝗶𝗴 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲: they react to multiple gases at once, leading to inaccurate readings. But what if a sensor could detect just one specific gas with pinpoint precision? That’s the vision behind the 𝗔𝗰𝗰𝗦𝗲𝗻𝘀 𝗣𝗿𝗼𝗷𝗲𝗰𝘁. Together with our project partners Materials Center Leoben Forschung GmbH (MCL) and University of Leoben, we have 𝗰𝗼𝗺𝗯𝗶𝗻𝗲𝗱 𝗺𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗩𝗦𝗣𝗮𝗿𝘁𝗶𝗰𝗹𝗲’𝘀 𝗻𝗮𝗻𝗼𝗽𝗿𝗶𝗻𝘁𝗶𝗻𝗴 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆 𝘁𝗼 𝘁𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺 𝗺𝗮𝘁𝗲𝗿𝗶𝗮𝗹 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁 𝗳𝗿𝗼𝗺 𝗮 𝘁𝗿𝗶𝗮𝗹-𝗮𝗻𝗱-𝗲𝗿𝗿𝗼𝗿 𝗽𝗿𝗼𝗰𝗲𝘀𝘀 𝗶𝗻𝘁𝗼 𝗮 𝗱𝗮𝘁𝗮-𝗱𝗿𝗶𝘃𝗲𝗻, 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲𝗱 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵. With the ability to rapidly test and optimize material combinations, we are driving a new era for gas sensor technology — making sensors faster, more accurate, and more efficient than ever before. 🎥 In this video, Dr. Anton Köck, Group Leader Sensor Solutions at MCL explains the project’s vision and impact. #gassensing #nanoporouslayers #nanomaterials #gassensors Leandro Sacco