Chalmers Open Digital Repository
Välkommen till Chalmers öppna digitala arkiv!
Här hittar du:
- Studentarbeten utgivna på lärosätet, såväl kandidatarbeten som examensarbeten på grund- och masternivå
- Digitala specialsamlingar, som t ex Chalmers modellkammare
- Utvalda projektrapporter
Forskningspublikationer, rapporter och avhandlingar hittar du i research.chalmers.se
Enheter i Chalmers ODR
Välj en enhet för att se alla samlingar.
Senast publicerade
- Comparative Analysis of the Dielectric Performance of Insulating Liquids Aged Under Different Conditions(2026) Svalander, Moses; Dahlman Zakrisson, JoelThis thesis investigates the dielectric performance of transformer oils used as high voltage insulation media in membrane-sealed and fully enclosed transformers and compares the impact of the sealing methods. Oil samples were collected from real transformer units, which were operating under different conditions over several years and the insulating oils in them were exposed to combined electrical-thermal stresses of different duration and intensity. It was anticipated that the intake of air through the sealings of the transformer tanks might initiate chemical reactions under such stresses thus causing accelerated ageing of the insulation. To examine this hypothesis, dielectric measurements and breakdown tests were conducted according to the respective standards. The former included measurements of complex capacitances, loss factor and oil resistivity whereas the latter provided the maximum electric field strength the oil could withstand. The obtained results indicate a rapid initial degradation in some of the measured parameters, while the evidence for storagetime- dependent degradation remains inconclusive. Under the applied test conditions, the oil samples taken from fully enclosed transformer tank showed a slightly better performance than those from membrane-sealed tank. To explicate this finding, an attempt to correlate the measured results with the histories of the respective transformer units is undertaken. Ageing factors affecting performance of the insulating oils, including chemical degradation under electro-thermal stresses and possible contaminations, are discussed based on the performed literature study.
- KULTURVERKSTAN I GAMLESTADEN(2026) Friström, Carl
- Understanding Work-Efficiency of label correcting Single Source Shortest Path Algorithms - How does relaxed data structures affect the work-efficiency of parallel algorithms?(2026) Berglöf, Alfred; Berg, JohanRelaxed priority schedulers have been successfully used to improve the scalability of parallel Single-Source Shortest Path (SSSP) algorithms by reducing synchronization overhead. However, relaxation introduces deviations from strict priority ordering, which may generate redundant work and reduce work-efficiency. This thesis investigates the relationship between scheduling relaxation and work-efficiency in label-correcting parallel SSSP algorithms. To study this relationship, we extend an existing benchmarking framework with extra measurements. In addition, we introduce a configurable relaxed scheduler, DrPQ, which enables controlled experimentation with different degrees of relaxation. Experiments are conducted on road networks, Random Hyperbolic Graphs (RHGs), and grid graphs using several state-of-the-art relaxed priority scheduler implementations. Our results show that relaxation error is correlated with redundant work, but is not sufficient on its own to explain work-efficiency. The relationship between relaxation and work-efficiency is influenced by graph structure, queue size, and algorithm design. High-degree RHGs generally tolerate larger relaxation errors while maintaining high work-efficiency, whereas road and grid graphs are more sensitive to the same level of relaxation. Furthermore, the skew between median rank error and delay provides additional information about scheduler behavior: some schedulers achieve high work-efficiency despite large skew, although this skew is not a universal predictor of redundant work. Overall, the thesis demonstrates that work-efficiency in label-correcting parallel SSSP algorithms emerges from the interaction between relaxation errors, graph characteristics, and scheduler design. No single relaxation metric consistently predicts redundant work across all graph classes, reinforcing that work-efficiency depends on the interplay of multiple factors rather than any isolated measure of relaxation.
- Scalable Parallel Branch-and-Bound with Relaxed Concurrent Queues - A Study of Relaxed Concurrent Queues in Branch-and-Bound using a Generic Solver Framework(2026) Sandh, LudvigRelaxed concurrent queues improve scalability by weakening strict ordering guarantees in exchange for reduced synchronisation overhead and contention. While previous work has demonstrated strong performance for such data structures in isolation, their applicability within complete algorithms and practical workloads remains less explored. This thesis investigates the use of relaxed concurrent queues, particularly the MultiQueue and related relaxed concurrent data structures, in branch-and bound (BnB) algorithms. A literature study identifies BnB as a suitable application domain due to its tolerance for non-strict exploration order, after which a generic modular framework for parallel BnB solving is developed. The framework separates problem-specific logic from parallel execution and queue management, greatly simplifying the development of high-performing branch-and-bound solvers. The framework is evaluated primarily through a case study on the Maximum Clique Problem using DIMACS benchmark instances on a 512-thread multicore system, alongside demonstrations on knapsack variants. The experimental results show that relaxed concurrent designs generally scale better than strict concurrent alternatives under high contention, with relaxed queues significantly outperforming strict concurrent counterparts while maintaining competitive search quality. Solvers using our framework with relaxed concurrent data structures sometimes outperform competitors, showing the value and further potential of our framework. The results demonstrate that data-structure-level parallelism based on relaxed concurrent queues can compete with algorithm-level parallelisation strategies in several cases, particularly at very high thread counts.