Student Projects

Nanyuan Zhang, PFA E93, nanyuan.zhang@inspire.ch Mert Ilten, PFA E93, mert.ilten@inspire.ch

The Machine Concepts research group at inspire AG is offering the following position: Project Engineer in AI-Driven Process Monitoring for Manufacturing We are seeking a motivated engineer to join our team for a funded innovation project in collaboration with a leading Swiss industrial partner. The focus is on developing AI-supported, sensor-based solutions for real-time monitoring and optimization of manufacturing processes. These solutions aim to assess process conditions during production to enhance stability, reduce variability, and enable data-driven optimization and maintenance strategies.

**Your Tasks** Develop and implement AI methods for in-process monitoring of surface quality and tool wear Analyze industrial sensor data (e.g., vibration, power, acoustic signals) to detect anomalies and trends Combine physical process models with data-driven methods to improve prediction accuracy Integrate and validate sensor systems on full-scale industrial machinery Collaborate closely with a Swiss machine manufacturer to bring research into practical application **Required Experience** CH/EU/EFTA citizenship or a valid Swiss work permit Master’s degree (ETH, university) in engineering, mechatronics, computer science, or a related field Experience in machine learning, signal processing, or condition monitoring Programming proficiency in Python, MATLAB, or comparable environments Strong interest in applying AI to real-world manufacturing problems **Required People Skills** Self-motivated, structured, and eager to explore innovative technologies Strong analytical mindset with hands-on problem-solving abilities Comfortable working in interdisciplinary, industry-oriented teams Fluency in German and English is required This role offers a unique opportunity to work at the forefront of applied research and innovation. You will contribute directly to the development of smart manufacturing technologies and work closely with Swiss industry to create a measurable impact. The position combines academic insight with industrial relevance, providing an ideal foundation for professional growth in applied research and development.

Please send your full application (cover letter, transcripts, CV, references) with reference ”AI-Driven Process Monitoring” to our HR department (hr@inspire.ch). For technical questions, please contact Dr. Markus Maier, Head of Machine Concepts (markus.maier@inspire.ch). Please also visit our websites www.inspire.ch and https://mohr.ethz.ch/. By submitting application documents, you accept the provisions of the privacy policy of inspire.

White metal bimetal bearings are vital for high-performance applications due to their excellent tribological properties. These bearings consist of a steel backing for mechanical strength and a thin layer of white metal (typically tin-based alloys) on the surface that interfaces with the shaft. White metal is used because it offers excellent conformability and low friction, making it ideal for use in engines, turbines, and heavy-duty machinery. However, conventional casting methods used to produce these bearings are wasteful, labor-intensive, and hard to automate. Laser Deposition Brazing (LDB) offers a more efficient and automatable alternative, yet struggles with thermal instability, particularly when processing low-melting-point white metals. This research project, which is a collaborative work with an industry partner active in the aerospace, automotive and energy sector, aims to enable stable LDB through the development of a closed-loop controlled cryogenic spray cooling (CSC) system using frozen CO₂ as a cooling medium. As part of this broader research effort, this Master’s thesis will focus on the CFD-based design optimization of the cryogenic cooling nozzle—a key component for enabling effective and adaptable cooling in real-time. The thesis will develop a simulation-based design framework to engineer and optimize a multi-jet CO₂ cooling nozzle, which can ensure stable and controlled cryogenic cooling during the deposition. The work involves both simulation and design tasks, with potential for experimental validation via 3D-printed prototypes. **Task 1: Literature Review** - Review state-of-the-art cooling nozzle designs in cryogenic spray cooling and relevant multiphase flow simulation methods. - Compare CFD approaches for gas jet modeling and cooling performance assessment. **Task 2: Computational Fluid Dynamics (CFD) Simulation Setup** - Define simulation settings (transient vs. steady-state, pressure-based vs. density-based solvers, turbulence models such as k-ε or k-ω, thermal boundary conditions). - Create nozzle geometry in Ansys DesignModeler or similar. - Generate a high-quality computational mesh using Ansys Meshing or compatible tools. - Set up simulations using either in-house OpenFOAM-based simulation software or commercial tools. - Implement boundary conditions for cryogenic CO₂ flow (e.g., pressure inlet, velocity outlet). - Select and apply appropriate numerical schemes for compressible and multiphase flow simulation. - Optional: Manufacture a test nozzle using additive manufacturing (for model validation). **Task 2: Design Parameter Variation and Optimization** - Identify key design variables: nozzle inlet/outlet dimensions, jet angle, nozzle length, number of jets. - Set a range of operating conditions: gas flow rate, nominal working distance. - Define performance metrics: effective cooling area, jet uniformity, working distance stability, cooling power density. - Perform systematic variation of geometry and flow parameters using CFD simulations. - Conduct sensitivity analysis to assess design robustness under variable operating conditions. - Optional: Apply design optimization techniques such as genetic algorithms or gradient-based methods to maximize performance.

By the end of the thesis, the following deliverables should be achieved: **D1:** Summary report of the literature review outlining: - State-of-the-art in cryogenic cooling nozzles and multiphase jet modeling - Overview of CFD strategies for gas flow and spray cooling simulations **D2:** Fully developed CFD simulation setup including geometry, solver settings, boundary conditions, and (if possible) validation results. **D3:** Optimized cooling nozzle design models. At least three promising multi-jet nozzle designs should be proposed and evaluated based on defined performance metrics.

Luis Neff, luis.neff@inspire.ch

This project is set to limited visibility by its publisher. To see the project description you need to log in at SiROP. Please follow these instructions: - Click link "Open this project..." below. - Log in to SiROP using your university login or create an account to see the details. If your affiliation is not created automatically, please follow these instructions: http://bit.ly/sirop-affiliate

The project will extend existing FEM models for SPS to include 3D geometries and hybrid material interactions, utilizing Si₃N₄-based ceramics composites. The investigation will follow a two-stage approach: Material Characterization & Parameter Determination: • Identification and determination of the necessary mechanical, thermal, and electrical properties specific to ceramic composition. • Utilization of experimental data to refine the input parameters for the FEM simulations. Numerical & Experimental Validation: • Design and development of a graphite tool with optimized geometry to promote uniform densification during the SPS process. • Simulation of the thermo-electro-mechanical behavior during SPS processing. • Comparison of FEM predictions with experimentally measured density, dimensional accuracy, and microstructural characteristics of the sintered components This approach will improve the accuracy of FEM models for SPS, facilitating better control over material densification and the fabrication of high-performance, complex ceramic components.

In this project, the aim is to investigate processing of near-net-shaped ceramic composite material components by use of Spark Plasma Sintering. • Enhance FEM capabilities for simulating near-net shaped ceramic sintering in SPS. • Determine material-dependent input parameters for FEM models. • Design, fabricate, and test graphite tooling for optimized densification. • Validate the final model through experimental sintering trials. • Compare simulation results with experimental data for density and geometry accuracy.

Dr. Natalia Kovalska nkovalska@ethz.ch