Examensarbeten för kandidatexamen
Länka till denna samling:
Browse
Senast publicerade
- PostDesign av en hybridmotor till en sondraket(2023) Hansson, Alexander; Kassem, Mohammed; Lindblad, Emil; Noord, Jesper; Nyberg, Atle; Rovelstad, Aaron; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Xisto, Carols; Capitano Patrao, Alexandre
- PostSäker biomimetisk luftinfångning av drönare(2023) Sterning, Åke; Nygren, Simon; Lahti, Leo; Åberg, Fredrik; Åkerlund, Daniel; Bergom, Filip; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Grönstedt, Tomas; Pons, Arion
- PostSimulering av kompressibelt flöde i Python. Utveckling av numeriska metoder för hantering av stötar i kompressibelt flöde(2023) Angel, Arvid; Johansson, Melvin; larsson, Lukas; Wiström, Jesper; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Andersson, Niklas; Ghosh, DebarsheeComputational fluid dynamics (CFD) simulations are an essential part of product development. If the simulations are physically correct, it can reduce the costs of experiments that otherwise would have to be done to evaluate the interactions between the product and the fluid it flows in or the fluid that flows past a statio nary product. When the fluid velocity equals to or exceeds the speed of sound, it introduces a phenomenon called shocks. The project aimed to investigate whether replacing the existing shock sensor in a CFD solver, called “G3Dflow”, could increase the accuracy and precision of the simulations. “G3Dflow” is a Python library provided by the Institution of Mecha nics and Maritime Sciences at Chalmers. The process of choosing a new shock sensor was mainly a matter of studying literature. One part of the project focused on how to calculate the gradients needed to compute the new sensor and how this affected the simulations. A major part of the project was evaluating the results of the simulations and comparing the results to literature describing test cases and the expected outcome. The results of the project are then to be used as a case for the course TME085-COMPRESSIBLE FLOW.
- PostUtveckling av turbulensmodeller med hjälp av maskininlärning i Python(2023) Batlouni, Fadi; Elm Jonsson, Benjamin; Fjeldså, Ole; Persson, Niclas; Ånestrand, Leo; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Andersson, Niklas; Davidson, LarsTurbulence modelling is a central component in the field of computational fluid dynamics (CFD), which aims to accurately simulate turbulent flows. To do this, solving the Navier-Stokes equations numerically is necessary. However, due to the fact that turbulence is rather chaotic, direct numerical simulation (DNS) is com putationally expensive. The idea of using Reynold-Averaged Navier Stokes (RANS) and turbulence models such as the k −ω model is to reduce the computational costs by simplifying the equations, at the cost of losing accuracy. The main aim of the project is to explore if some specific parameters used in these simplified equations, usually assumed to be constant, can be optimized with the use of machine learning (ML), and how the improved models fare against previous mo dels. The ML methods used are support vector regression (SVR), k nearest neighbors regression (kNN), and neural networks. The project found that optimizing Cµ with the k − ω model is inessential, since the model badly predicts k even though the fraction of k and ω can be correct making the optimization difficult. Nonetheless, optimizing other parameters still proves to be rewarding. The use of machine lear ning to improve turbulence models is considered promising and should be explored further.
- PostDegradation analysis of fuel cells for heavy-duty vehicle applications(2023) Nyström, Emil; Olger, Emma; Sever, Stjepana; Stewall, Emelie; Chalmers tekniska högskola / Institutionen för mekanik och maritima vetenskaper; Chalmers University of Technology / Department of Mechanics and Maritime Sciences; Sedarsky, David; Enelund, MikaelClimate change is one of the most pressing issues of today, and the emissions from fossil fuel-powered vehicles are major contributors. A potential solution to reduce the use of fossil fuels in vehicles is the implementation of fuel cells. However, the main challenge of using fuel cells is degradation, which results in a loss of performance. Therefore, on behalf of Volvo trucks, this project aims to model, analyze and draw conclusions about the performance and durability of a fuel cell used in trucks, with a special focus on the degradation in the catalyst layer caused by Ostwald ripening. The degradation is studied by modeling in Matlab and GT-Suite. The Matlab model studies the Ostwald ripening phenomena for two different drive cycles and outputs electrochemical surface area dependent on cell runtime. The data are then used to analyze the performance of a fuel cell stack in GT-Suite. The GT-Suite simulation calculates power output as well as polarization curves for the cell, which was done for each data point provided by the Matlab analysis. In relation to the Ostwald ripening phenomenon, a steady voltage displays a 17% loss of electrical power at 10 000 hours, in contrast, fluctuating voltage shows a 17% loss already at 6000 hours. It reveals a clear dependence on whether the voltage is steady or fluctuating. The study concludes that Ostwald ripening is a significant contributor to fuel cell degradation and needs to be taken into account for further development of the technology. Temperature and humidity have also been shown to ha an effect on fuel cell performance and need to be studied further in order for fuel cells to become a potential long-term solution for heavy-duty vehicles.