TY - GEN A1 - Sharma, Dikshant A1 - Radomsky, Lukas A1 - Mathiazhagan, Akilan A1 - Konda, Karunakar Reddy A1 - Hammami, Ghaieth A1 - Asli, Majid A1 - Höschler, Klaus A1 - Mallwitz, Regine T1 - Strut-based porous media heatsinks for high-performance power electronics thermal management in electrified aircrafts T2 - ASME Turbo Expo 2025 : Turbomachinery Technical Conference and Exposition : Volume 4: Controls, Diagnostics & Instrumentation; Cycle Innovations; Education; Electric Power : June 16–20, 2025, Memphis, Tennessee, USA N2 - Multi-level inverters are one promising solution for high-power applications, enabling higher efficiency and improved power quality over conventional inverters. The emergence of these converter topologies with a larger number of topological switches makes reliable, forced and even natural convection air cooling a feasible option for aircraft power electronics. The need for high heat dissipation rate, robust design and lightweight heatsinks has led to the development of strut-based porous media structures for forced air cooling. The current work focuses on investigating Kelvin, Body-Centered Cubic (BCC) and Simple Cubic (SC) periodic open cellular structured (POCS) lattice heatsink with a fixed porosity and a fixed unit cell size. 3D printed Kelvin and SC heatsinks using AlSi10Mg material are tested in an air duct experimental setup along with a conventional LAM aluminium heatsink. The Computational Fluid Dynamics (CFD) simulation model is validated with the experimental results and a 0D thermal model is developed using the CFD results. The CFD thermal results are in close accordance with the experimental results for the POCS heatsink within an error band of ±2%. The 0D results using the thermal data from CFD simulations also show a close comparison for the calculated semiconductor junction temperatures. The Kelvin heatsink performs the best thermally from the CFD analysis and has the least error when comparing the 0D and 3D-CFD results. Y1 - 2025 SN - 978-0-7918-8880-3 U6 - https://doi.org/10.1115/GT2025-152670 VL - 4 IS - V004T06A011 SP - 1 EP - 11 PB - The American Society of Mechanical Engineers CY - New York, NY ER - TY - GEN A1 - Kim, Dongsuk A1 - Gerstberger, Ulf A1 - Asli, Majid A1 - Höschler, Klaus T1 - U-Net driven semantic segmentation for detection and quantification of cracks on gas turbine blade tips T2 - Results in engineering N2 - Crack detection and quantification on gas turbine blades is crucial for component validation during the development phase and for operational efficiency in service, as unexpected cracks can compromise blade integrity and lead to early engine removals. Gas turbine blades operate under extreme thermal and mechanical stresses, making them particularly susceptible to crack formation. At the same time deterministic predictions of crack formation are subject to high uncertainty in material data and actual loading conditions. Accurate detection and quantification of cracks, therefore, is essential for the validation and calibration of life predictions in order to prevent in-service failures, to extend component lifespan, and to reduce maintenance costs. This study introduces a U-Net based semantic segmentation model designed to automate crack detection on turbine blade tips. The model was trained on a dataset of 210 surface images with and without evidence of cracks, each divided into 128  ×  128 pixel patches. Data augmentation techniques were applied to address the class imbalance between cracked and non-cracked pixels. The U-Net architecture, optimized with a Dice loss function, achieved a validation IoU of 0.7557, along with approximately 85% recall and precision in identifying cracked pixels. The pixel-based accuracy of the model primarily affects the quantification of cracks rather than their identification. A sliding window pipeline was implemented to extend the model’s applicability, enabling segmentation of entire blade tip images for comprehensive crack localization. While the model may occasionally miss low-contrast cracks, it holds potential as a supplementary tool for manual inspection as part of the life prediction validation. By providing automated crack localization and quantification, the model can assist in analyzing crack characteristics relative to engine operating conditions. KW - Crack detection KW - Semantic segmentation KW - Gas turbine blade KW - U-Net KW - Deep learning KW - Convolutional neural network Y1 - 2026 U6 - https://doi.org/10.1016/j.rineng.2025.108864 SN - 2590-1230 VL - 29 SP - 1 EP - 9 PB - Elsevier BV CY - Amsterdam ER -