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 inlagda

Bild (thumbnail)
Post
Challenges and Opportunities within Design for Additive Manufacturing
(2024) Dasappa Ashoka, Vijaya Kumar; Sidnekoppa, Luqmaan; Chalmers tekniska högskola / Institutionen för industri- och materialvetenskap; Chalmers University of Technology / Department of Industrial and Materials Science; Hryha, Eduard; Soundarapandiyan, Gowtham; Hryha, Eduard; Knuts, Sören; Andersson, Petter
Additive manufacturing (AM) is transforming the design and manufacturing industries by allowing production of complex geometrical components that enhances efficiency and performance. However, despite its significant potential, AM poses several challenges, particularly in achieving cost-effective mass production. This study focused on interviewing engineers within an aerospace product development organization to identify the challenges and opportunities related to AM powder bed development. The responses were analyzed, and the findings were visually represented using an (Design for additive manufacturing) DfAM AIM diagram. Additionally a benchmark was performed with the additional industrial partner that has experienced in this area. The results from the interviews highlighted several key challenges in the AM design process and within the organization design practice. Design for additive manufacturing is a specialist area closely related to specific AM process. The study also identifies knowledge gaps and lack of communication between people and parts of organization. Process simulation software is not a standardized platform within the company, and difficult to use which means that front loading in design is still hard to perform. A conclusion is that, the organization needs to undertake several strategic improvements such as bridging the knowledge gap and improve the communication. Implementing standardized design guidelines specific to AM will streamline the design process and reduce reliance on trial-and-error methods. Furthermore, by integrating advanced simulation tools early in the design phase support structures can be optimized, and material waste can be reduced, and manufacturing outcome can be more accurately predicted. This integration will not only improve efficiency but also reduce costs. Lastly, investing in research and development to refine post-processing techniques and explore alternative materials could further enhance the economic viability of AM.
Bild (thumbnail)
Post
Pharmaceutical assay search with AI
(2024) Alladin, Ali; Chalmers tekniska högskola / Institutionen för matematiska vetenskaper; Jonasson, Johan; Jonasson, Johan
Retrieving historical assay data in pharmaceutical research is often restricted by reliance on specific metadata, overlooking the contextual information in associated protocol documents. This thesis investigates the potential of utilizing these plain English protocol documents alongside Natural Language Processing (NLP) techniques to implement semantic search for assays. A baseline TF-IDF model and the Transformer models BERT, SBERT, and Longformer were used to get embeddings of protocol documents from a corpus of historical protocols. Their performance in retrieving relevant historical protocols was evaluated based on key technical criteria, where the TF-IDF models and BERT using the chunking technique showed the best results. However, limitations in the evaluation scope introduce some uncertainty to the findings, highlighting the need for more rigorous validation. Nevertheless, the conclusions suggest that integrating NLP-driven semantic search systems could reduce the time and manual effort required for assay retrieval, even though the current approach may need further refinement for practical application. These insights are a promising foundation for developing AI-powered search systems used for pharmaceutical texts.
Bild (thumbnail)
Post
Analysis and Study of Self-driving bikes
(2024) Kuduva Prakash, Brijesh; Jayachandran, Harish; Chalmers tekniska högskola / Institutionen för elektroteknik; Sjöberg, Jonas; Björnsson, Carina
Abstract Advancements in vehicle technologies and their active safety systems have necessitated various testing methods to ensure safety and reliability. Traditional methods using stationary bikes or simple mobile platforms lack realistic behaviour as compared to a bicyclist. Testing for bikes is crucial as cycling is a major mode of transportation for the general population. The increase in bicycles [1] and e-scooter usage [2], prompts Original equipment manufacturers, to decrease and mitigate the safety risks for these users termed as vulnerable road users. The Self-driving bike project aims to bridge this gap by developing bikes and e-scooter that mimic human riding behavior. These create realistic test scenarios to evaluate vehicle safety systems which is supported by a collaboration with Volvo Cars. This thesis contributes to the development of self-driving bikes and e-scooters by focusing on areas, including the remodelling of the steering motor mount for the e-scooter for improved durability, addressing cable management issues, and designing a 3D-printed roller for indoor testing of the e-scooter. Fine-tuning the steering motor through ESCON Studio and calibrating the motor contributed to optimize the steering angle range. The configuration and setting up of self-driving bikes and e-scooter for field tests at Astazero, collecting and analyzing data to refine performance along with an analysis of indoor and outdoor tests, focusing on balance, steering rates, and other performance metrics under various conditions. These contributions enhance the testing mechanisms for self-driving bikes and e-scooter, bridging the gap between stationary methods and real-world scenarios.
Bild (thumbnail)
Post
Design av transportbehållare för sjöräddningsdrönare
Persson, Samuel; Ranbring, Ted; Chalmers tekniska högskola / Institutionen för elektroteknik; Chalmers University of Technology / Department of Electrical Engineering; Thomas, Bertil; Falkman, Fredrik
Sammanfattning I dagsläget tar en utryckning av svenska sjöräddningssällskapet från land till olycksplats i genomsnitt 40 minuter. För att förbättra hantering av situationer har Sjöräddningssällskapet utvecklat en fixed-wing drönare som på 10 minuter kan ta sig till olycksplatsen och skicka bilder som kan analyseras av räddningspersonal. Därför undersöks nu möjligheten att bestycka drönare med gods som kan underlätta vid räddningar. Examensarbetet vid Svenska Sjöräddningssällskapet går ut på att utveckla en behållare som kan vinschas ner från en fixed-wing drönare. Behållaren är utvecklad för att möjliggöra transport av föremål ut till havs, vid larm kan den lastas med uppdragsspecifikt gods och flyga till destinationen. Inledningsvis utfördes en marknadsundersökning för att få en bättre förståelse för vilka befintliga lösningar och idéer som redan finns på marknaden. Utifrån detta framställdes det ett flertal olika koncept som sedan eliminerades till ett koncept kvarstod, vilket sedan realiserades och vidareutvecklades. Resultatet av detta arbete var en fungerande prototyp för behållarstabilisering med drönare. Genom fortsatt arbete, omfattande tester och implementering med drönare finns potentialen att uppnå målet att effektivisera räddningsinsatser till havs. Denna framstegsmöjlighet öppnar upp för innovation och teknologi kan bidra avsevärt till att öka säkerheten och effektiviteten vid räddningsinsatser till sjöss.
Bild (thumbnail)
Post
Attention-based Time Series Forecasting with Limited Data
(2024) Vadström, Gustav; Chalmers tekniska högskola / Institutionen för elektroteknik; Monti, Paolo; Banar, Jafar
Electricity outages are common in electrical power systems, and often caused by natural phenomena, human intervention, or faults in electrical components, such as transformers. A small number of these faults can be predicted by analysing the stream of voltage and current. Forecasting faults in electrical power systems can prevent electricity outages that cause production downtime and capital losses. However, data collected in power systems are usually limited and unbalanced because of the very few historical predictable faults. This study focused on evaluating more recently popular attention-based machine learning models for time series prediction in electrical power systems, in a context where data is a significant limitation. The data was real and consisted of disturbances recorded from power systems over sev eral years, along with documented faults. Two different model architectures were evaluated and compared: the Long short-term memory (LSTM) and the transformer. Three different model instances were trained: using features manually extracted from each disturbance recording, using manually extracted features with pre-training on a similar dataset, and using a signal embedding pipeline attached to each model processing raw waveforms. The results from all six training instances showed that the transformer performed better than the LSTM in terms of evaluation metrics, although the LSTM outputs were more interpretable, because the transformer had higher confidence in its outputs even during false predictions. A bottleneck was found in the small sequence lengths, with improvement shown when utilizing pre training on a similar dataset containing longer sequences. The integrated waveform feature embedding also showed improvement over the manually extracted features.