Sökning: "transformatorer"
Visar resultat 21 - 25 av 129 uppsatser innehållade ordet transformatorer.
21. Computer Vision for Document Image Analysis and Text Extraction
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Automatic document processing has been a subject of interest in the industry for the past few years, especially with the recent technological advances in Machine Learning and Computer Vision. This project investigates in-depth a major component used in Document Image Processing known as Optical Character Recognition (OCR). LÄS MER
22. Transformer-based Source Code Description Generation : An ensemble learning-based approach
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Code comprehension can be significantly benefited from high-level source code summaries. For the majority of the developers, understanding another developer’s code or code that was written in the past by them, is a timeconsuming and frustrating task. LÄS MER
23. Evaluation of mini-grid versus stand-alone projects: a tool for decision-making on household connections in last-mile rural electrification projects
Master-uppsats, KTH/Skolan för industriell teknik och management (ITM)Sammanfattning : Electrification of isolated rural areas in developing countries is a key action for the total eradication of energy poverty. A growing number of authors argue that this electrification should be done with distributed energy systems that combine different renewable energy sources, rather than trying to reach these communities with grid extensions. LÄS MER
24. Task-agnostic knowledge distillation of mBERT to Swedish
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Large transformer models have shown great performance in multiple natural language processing tasks. However, slow inference, strong dependency on powerful hardware, and large energy consumption limit their availability. LÄS MER
25. Predicting Top-Oil-Temperature in Power Transformers with Light Gradient Boosting Machine
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Online monitoring of mineral-oil-immersed transformer variables provide capabilities such as failure avoidance, optimization of load capacity and maintenance as well as extending the remaining life expectancy. A central role is played by the Top-Oil-Temperature which in its own is an indicator of harmful events or transformer states. LÄS MER