Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Adaptive decay: Investigating ruins through architectural filming
No full textIn the mid-19th century, the sawmill industry in Ådalen experienced a significant boom, with numerous sawmills operating in the area. One of these was the Köja sawmill, notable for being the first in the region to feature a steam-powered saw. However, as time passed, economic conditions worsened, leading to the sawmill’s closure in 1940. Over the decades, the building has deteriorated, with its facade developing holes, and its windows losing all their glass. The only remaining building is the engine house, along with part of the foundation of the sawmill where the timber was cut.
The documentation highlights and visualizes the aesthetic elements of a ruin and its essential characteristics through Tim Edensor’s Industrial Ruins and Christopher Bretecher’s Abandoned Potential conceptual framework. The design approach incorporates Jonathan Hill’s concept which suggests that the design process should integrate new fragments with the old, creating a palimpsest of layers. These layers should not only coexist but actively interact, evolving as a living entity. The design strategy focuses on preserving the ruin’s aesthetics while simultaneously introducing new materials and functions.
Proposal for Köja ruin is a space where the ruin is both preserved for the future and allowed to continue its natural decay. This dual approach is achieved by designating a portion of the ruin where no interventions are made, allowing the decay to unfold naturally. This process serves as a way to interpret and contextualize the site’s history and future. Preservation, on the other hand, is realized by repurposing the ruin into a ruin experience centre. The proposal is presented through conventional architectural drawings as well as a film. The film helps the viewer understand the different spatial, conceptual layers and nuances that Köja ruins has with a structural sequence of, showing the existing ruin, its construction, the proposed transformation and a speculative future scenario. This project contributes to a deeper understanding of how we can repurpose the ruins but still keep its characteristics.
“How can a ruin be investigated through filming?”
“How can a ruin’s aesthetics and process of decay be preserved when re-adapted?
Vehicle Design Optimization Using AI/ML Methods
No full textEnergy efficiency and performance are important attributes when developing and designing vehicles. With the transition of the automotive industry towards energy efficiency and sustainability, it is ever more important to save computational resources. Traditional vehicle design optimization heavily relies on computationally expensive simulations. The simulations carried out in this project particularly focus on the powertrain of vehicles. This work focuses on developing a surrogate-assisted, multiobjective optimization framework that efficiently finds the optimal values for the given variables. Surrogate modeling is an engineering method used when the outcome of an experiment cannot be easily computed, so a mathematical approximation is applied. In this case, we use machine learning models to predict the outcome of expensive simulations. These trained model(s) is then used for optimization instead of running optimization on the simulation model directly. First, we generate 4096 Sobol-sampled configurations spanning different parameters like gear ratios and electric motors. We train and compare different surrogate models like Random Forest, XGBoost, and LightGBM on these data, achieving test R2 scores up to 0.96 with Random Forest. Next, we employ the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to explore trade-offs among various conflicting objectives, extracting a Pareto front of optimal designs. A weighted-sum post-processing step or a constrained method later selects a single best-trade-off configuration, which full simulation validates. This framework slashes computational cost and empowers rapid, data-driven vehicle powertrain design
Using Transformers for Chemical Toxicity Prediction
No full textPollution from toxic chemicals threatens both biodiversity and human health, resulting
in significant costs for society. To mitigate these impacts, chemical emission
regulations, such as EC50, are employed. These regulations typically establish environmentally
safe concentrations of chemicals based on data from in vivo experiments,
which are usually time-consuming, expensive, and sometimes ethically problematic
to conduct. As alternative means to predict chemical toxicity, previous studies have
proposed computational methods (e.g., QSAR) and machine learning approaches,
including transformer-based models. In one of these previous studies, a pre-trained
transformer-based model was employed to predict the EC50 values of chemicals for
fish. The chemical structures were represented using SMILES notation and served
as the input to the model, which consisted of a RoBERTa component followed by
a fully connected feed-forward neural network. The present master’s thesis builds
upon this study by using the same dataset and model framework. It aims to compare
the toxicity prediction performance of fine-tuned-only models with different
model hyperparameters related to the model architecture and analyze the influence
of these hyperparameters. In addition, the thesis aims to evaluate the impact of
pre-training by comparing these models with different model hyperparameters to a
pre-trained and fine-tuned ChemBERTa model. The effect of model architecture
was examined only for the RoBERTa component by varying the following model
hyperparameters: embedding size, number of encoder layers, and number of attention
heads. The results indicated that increasing the embedding size and the
number of encoder layers improved prediction performance. In contrast, no clear
pattern regarding the impact of the number of attention heads on prediction performance
was observed. Additionally, pre-training appeared to be necessary since the
ChemBERTa-based model outperformed all non-pretrained models. These findings
contribute to the development of transformer-based machine learning models for
chemical toxicity prediction by indicating the optimal directions regarding model
architecture and pre-training approaches. Thus, future research may include evaluating
whether these findings hold for larger model hyperparameter values, as well
as for other chemical representations, toxicity endpoints, and species beyond EC50
and fish
Vägen till levande kontorsmiljöer i handelspräglade områden; en studie av vakanser i Nordstans kontorslokaler och möjliga strategier för ökad attraktivitet
No full textKöpcentrumet Nordstan i Göteborg är en av Sveriges mest centralt belägna
handelsplatser, men platsen i sig har upplevts otrygg och kontorsverksamheten i
byggnaden har uppfattats som osynlig. Studien syftar till att förklara vilka de främsta
orsakerna till kontorsvakanser i Nordstan samt vad som kan åtgärdas för att minska
dessa vakanser. Arbetet bygger på en kvalitativ studie med semistrukturerade
intervjuer, respondenterna består av fastighetsägare med koppling till köpcentrum och
kontorshyresgäster i Nordstan.
Studien påvisar att Nordstans kontorslokaler har låg synlighet, vilket gör att
allmänheten i många fall inte är medvetna om att det finns kontorversamhet i området.
Denna brist på synlighet har bidragit till att Nordstans kontorslokaler blir bortvalda
för andra kontorsområden. Genom studien har det även framkommit att Nordstan har
haft ett dåligt rykte gällande trygghet. Trots mycket arbete med tryggheten så ligger
samma image av Nordstan kvar hos allmänheten, vilket gör att företag kan undvika att
bosätta sig i området. .
För att öka intresset hos kontorsområdet så betonar studien att tydligare synliggöra
kontorsmiljöerna. Genom att öka medvetenheten om att det finns kontor i Nordstan så
kan fler företag överväga Nordstan som en framtida plats för etablering. Vidare visar
studien att trygghetsarbetet måste visualiseras tydligare för allmänheten för att gradvis
förändra den negativa bild av Nordstan som många har. På så sätt kan
kontorsvakanserna successivt minska och Nordstans upplevelse som plats stärkas
Design of tomorrow’s marine engine covers
No full textThis thesis explores the redesign of the D6 engine cover with the objectives of improving
noise reduction, aesthetics, and durability. The primary motivation for the
project was to achieve a measurable reduction in engine noise through modifications
to the cover alone, without altering the engine itself. Additionally, the existing
cover has remained unchanged for several years, highlighting an opportunity for design
improvement. The project adopts a traditional design methodology based on
sequential engineering, guiding the process from problem identification to concept
generation and prototyping.
A series of design concepts were developed and evaluated according to acoustic performance,
visual appeal, and structural resilience. These concepts were iteratively
refined, leading to the selection of one final design considered the most promising
based on a combination of theoretical performance and practical feasibility. This
final concept was further developed using CAD modeling tools and subsequently
prototyped through 3D printing, enabling preliminary physical evaluation.
Although the project resulted in a tangible prototype, the findings underline the
need for continued research, particularly in advanced acoustic testing, material optimization,
and long-term durability studies, to validate the design under realistic
operating conditions. This work provides a foundation for future development and
highlights the value of design-focused interventions in addressing performance and
aesthetic challenges in marine engine component
Safe and Reliable Switching of Reactive Capacitor Banks in Sweden’s Industrial Sector
No full textAbstract
The main scope of this project aims to evaluate methods of limiting inrush current when connecting capacitor banks to a 12.1 kV grid in an industrial environment. Traditionally, Swedish industries have connected large capacitor banks in parallel to the load for reactive power compensation, using asynchronous switching methods. However, the need of reactive power compensation is changing, due to an increase of power electronics and drive systems, requiring lesser amount of reactive power compensation but greater need of filtering. Establishing guiding parameters from accepted international standard IEC 61000-2-4 to measure the performance and behavior of different methods, was a major part of the evaluation method.
Depending on the amount of reactive power (Q) compensation and harmonic filtering needed, this report suggests different solutions for different use cases. For low frequent switching rates: A pre-insertion resistor is suitable for cases with major need of Q-compensation together with a small inrush reactor. A de-tuned filter is suitable when low amounts of Q-compensation is needed, but instead the need of filtering is large. Further, a TSC is applicable for any scenario of the above, with the additional feature of being able to switch asynchronously and more frequent than the regular synchronous/asynchronous breakers. Finally, a theoretical example of a strategy using variable resistance is also evaluated, with impressive results in limiting inrush currents during switching. Even though this is not yet typically deployed for 12.1 kV voltage levels, this is by far the most effective strategy evaluated
Interior Design of an Urban Air Mobility Vehicle for Use in Prehospital Medical Emergencies - Developing a Concept to Complement Today’s Ambulance Vehicles
No full textUrban Air Mobility (UAM) is a rapidly emerging field focused on creating aerial traffic systems in urban cities and utilising the air for transportation of passengers and cargo, to ease arising issues with traffic congestion and increased travel times. The research project Airmobility Emergency System (AMES) at Chalmers aims to develop a UAM
vehicle for implementation within emergency medical services in the near future, with focus on the Region Västra Götaland (VGR) in Sweden. UAM will serve as a complement when neither road ambulances nor helicopters are suitable options. This ensures that public emergency services can access accident sites regardless of ground conditions.
This project is part of AMES and focuses on developing a comfortable, functional, and user-friendly interior concept that meets the needs of ambulance personnel, enhancing their ability to provide care with effectiveness and efficiency. Based on literature, a survey and field studies conducted at the ambulance helicopter base and a road ambulance station in VGR, an interior design concept was developed and evaluated through 1:1 scale prototyping and CAD modelling. A final evaluation with paramedics in the full-scale
prototype, followed by refinements, resulted in a user-centred interior solution. The primary project deliverables consist of a visualisation of the concept through a comprehensive CAD model, supported by a detailed product speciation and an equipment list. Together, these deliverables propose a recommended interior design for an Urban Air Mobility (UAM) vehicle intended for future implementation in emergency medical services.
The proposed interior design concept features a seat–stretcher configuration with one stretcher positioned centrally. A seat is placed on each side of the stretcher, with an additional seat behind the patient’s head. Storage solutions are integrated at the back of the cabin to accommodate items not needed during flight. Additional storage is positioned along the ceiling. Loose equipment and bags can also be secured to the floor, enabled by the built-in L-tracks. In situations where the autonomous system would require human intervention, essential manoeuvring hardware such as stop buttons, displays and a dashboard, is incorporated into the design
Sensor-Based Virtual Fences for Industrial Robot Safety
No full textIn industrial environments, safety has traditionally been ensured using physical barriers such as fences or enclosures to separate humans from robotic systems. While effective, these static solutions limit flexibility and make it difficult to adapt the layout to changing production needs. As factories become more dynamic and collaborative, there is a growing need for smarter, more adaptable safety systems. In this project, a virtual safety system for an industrial robot was developed using LiDAR sensors and real-time data processing. The system was designed to replace traditional physical barriers by creating three safety zones around the robot a safe zone, a warning zone and a restricted zone, depending on the distance of approaching objects. The sensors and control electronics were built and tested in real life, while the behavior and reactions of the robot were evaluated through simulation. The system continuously monitored the area at a height of 15 to 20 cm above the floor and successfully detected objects and classified them into the correct zones. The tests showed that the system detected all intrusions correctly in both warning and stop zones. The average response time was around 10 ms, which is fast enough for real-time feedback. However, the system experienced false intrusions in some cases, especially when using larger zones and more active components—up to 601 false triggers during 20 minutes recorded in one test. The results demonstrate that the system was able to trigger appropriate responses based on risk level, and show that virtual safety zones could be a viable and flexible alternative to traditional physical fences in industrial robot applications
Improving energy grades in norwegian dwellings through local PV and battery systems a path to zero-emission
No full textThe EU has introduced the EPBD, which was implemented in 2024, and is a new directive
targeting energy performance in buildings, which aims to reach a zero-emission building
stock by 2050. Additionally, it states that 55% of the total reduction must come from
renovating 43% of the worst-performing buildings. This thesis aims to investigate how to
reach a zero-emission residential building stock in Norway, through investigating various
energy reduction measures in existing dwellings, and how they will contribute to the energy
grading based on the energy labeling system. The study’s results are based on literature
review, data collection, case studies, as well as utilizing mixed integer linear programming to
investigate the intricacies of using PV panels, batteries, and flexible loads. The analysis
shows in terms of cost, traditional measures such as insulation remain more economically
feasible. But it’s PV and hybrid systems that deliver the biggest energy reduction. A major
limitation is that battery systems currently receive no Enova subsidies, even though they have
a high impact on both grading and self-consumption. That’s something worth re-evaluating at
the policy level. Overall, PV and battery solutions show strong potential, both in older
buildings, like Rørvollveien 17, and in homes already close to grade A, like Jongsåsveien 31
A, and they can play a central role in reaching a zero-emission building stock