Substation apparatus failure plays a major role in reliability of power delivery systems. Traditionally, most utilities perform regular maintenance in order to prevent equipment breakdown. Condition-based maintenance strategy monitors the condition of the equipment by measuring and analyzing key parameters and recommends optimum maintenance actions. Equipment such as transformers and standby batteries which are valuable and critical assets in substations has attracted increased attentions in recently years. An automated monitoring and diagnosis tool for power transformers based on dissolved gas analysis, ANNEPS v4.0, was developed. The new tool extended the existing expert system and artificial neural network diagnostic engine with automated data acquisition, display, archiving, and alarm notification functions. This thesis also studied substation batteries types and failure mode and surveyed the market of current on-line battery monitors. A practical battery monitoring system architecture was proposed. Analysis rules of measured parameters were developed. The above study and results can provide basics for further designing of a simple battery monitoring system in industry applications. / Master of Science

This dissertation presents a comprehensive exploration and implementation of attention mechanisms and transformers on several healthcare-related and assistive applications. The overarching goal is to demonstrate successful implementation of the state-of-the-art approaches and provide validated models with their superior performance to inform future research and development. In Chapter 1, attention mechanisms are harnessed for the fine-grained classification of white blood cells (WBCs), showcasing their efficacy in medical diagnostics. The proposed multi-attention framework ensures accurate WBC subtype classification by capturing discriminative features from various layers, leading to superior performance compared to other existing approaches used in previous work. More importantly, the attention-based method showed consistently better results than without attention in all three backbone architectures tested (ResNet, XceptionNet and Efficient- Net). Chapter 2 introduces a self-supervised framework leveraging vision transformers for object detection, semantic and custom algorithms for collision prediction in application to assistive technology for visually impaired. In addition, Multimodal sensory feedback system was designed and fabricated to convey environmental information and potential collisions to the user for real-time navigation and grasping assistance. Chapter 3 presents implementation of transformer-based method for operation-relevant human activity recognition (HAR) and demonstrated its performance over other deep learning model, long-short term memory (LSTM). In addition, feature engineering was used (principal component analysis) to extract most discriminatory and representative motion features from the instrumented sensors, indicating that the joint angle features are more important than body segment orientations. Further, identification of a minimal number and placement of wearable sensors for use in real-world data collections and activity recognitions, addressing the critical gap found in the respective field to enhance the practicality and utility of wearable sensors for HAR. The premise and efficacy of attention-based mechanisms and transformers was confirmed through its demonstrated performance in classification accuracy as compared to LSTM. These research outcomes from three distinct applications of attention-based mechanisms and trans- formers and demonstrated performance over existing models and methods support their utility and applicability across various biomedical and human activity research fields. By sharing the custom designed model architectures, implementation methods, and resulting classification performance has direct impact in the related field by allowing direct adoption and implementation of the developed methods.

Autonomous driving requires lightweight and robust perception systems that can rapidly and accurately interpret the complex driving environment. This dissertation investigates the transformative capacity of discrete wavelet transform (DWT), inverse DWT, CNNs, and transformers as foundational elements to develop lightweight perception architectures for autonomous vehicles. The inherent properties of DWT, including its invertibility, sparsity, time-frequency localization, and ability to capture multi-scale information, present an inductive bias. Similarly, transformers capture long-range dependency between features. By harnessing these attributes, novel wavelet-enhanced deep learning architectures are introduced. The first contribution is introducing a lightweight backbone network that can be employed for real-time processing. This network balances processing speed and accuracy, outperforming established models like ResNet-50 and VGG16 in terms of accuracy while remaining computationally efficient. Moreover, a multiresolution attention mechanism is introduced for CNNs to enhance feature extraction. This mechanism directs the network's focus toward crucial features while suppressing less significant ones. Likewise, a transformer model is proposed by leveraging the properties of DWT with vision transformers. The proposed wavelet-based transformer utilizes the convolution theorem in the frequency domain to mitigate the computational burden on vision transformers caused by multi-head self-attention. Furthermore, a proposed wavelet-multiresolution-analysis-based 3D object detection model exploits DWT's invertibility, ensuring comprehensive environmental information capture. Lastly, a multimodal fusion model is presented to use information from multiple sensors. Sensors have limitations, and there is no one-fits-all sensor for specific applications. Therefore, multimodal fusion is proposed to use the best out of different sensors. Using a transformer to capture long-range feature dependencies, this model effectively fuses the depth cues from LiDAR with the rich texture derived from cameras. The multimodal fusion model is a promising approach that integrates backbone networks and transformers to achieve lightweight and competitive results for 3D object detection. Moreover, the proposed model utilizes various network optimization methods, including pruning, quantization, and quantization-aware training, to minimize the computational load while maintaining optimal performance. The experimental results across various datasets for classification networks, attention mechanisms, 3D object detection, and multimodal fusion indicate a promising direction in developing a lightweight and robust perception system for robotics, particularly in autonomous driving.

Many organisations face problems with data digitisation and continuous data gathering. They often gather and store this data in outdated systems that are difficult to search through. In our thesis, we utilise the engineering method to investigate the feasibility of incorporating artificial intelligence to search a large corpus of data and find accurate answers. To achieve the thesis goal, we conducted a literature review, studying existing solutions that enhance flexibility and facilitate artificial intelligence operations to search in databases. This resulted in the choice of utilising OpenSearch. Within OpenSearch, we conducted an experiment investigating which sentence transformer for embedding the contextual meaning of sentences could be best utilised for semantic search in the database. We then evaluated the sentence transformers´s performance with the MS MARCO dataset measuring both speed and accuracy. Through the experiment we found two sentence transformers that outperformed the rest by a slight margin and that all the sentence transformers performed similarly overall. A notable result is that the sentence transformers specifically dedicated to semantic search and sentence transformers with larger dimensions did not perform better. Further, these results showed the easy combination of existing search engines that incorporate artificial intelligence to semantically search in the documentation and showed that this could be used within organisations to handle a large corpus of data.

The diploma thesis deals with conventional transformers and current and voltage sensors. The first half of thesis describes mainly the basic concepts and accuracy classes of these converters used in electrical substations. The emphasis is given primarily on the differences. In the second half of thesis is presented an exact type of converters, which are installed in the substation Medlánky. There is an analysis of measured data from long – term monitoring from this substation and comparing the results from conventional transformers compared to the results from more modern sensors.

The use of Machine Learning (ML) in business has increased significantly over the past years. Creating high quality and robust models requires a lot of data, which is at times infeasible to obtain. As more people are becoming concerned about their data being misused, data privacy is increasingly strengthened. In 2018, the General Data Protection Regulation (GDPR), was announced within the EU. Models that use either sensitive or personal data to train need to obtain that data in accordance with the regulatory rules, such as GDPR. One other data related issue is that enterprises who wish to collaborate on model building face problems when it requires them to share their private corporate data [36, 38]. In this thesis we will investigate how one might overcome the issue of directly accessing private data when training ML models by employing Federated Learning (FL) [38]. The concept of FL is to allow several silos, i.e. separate parties, to train models with the same objective, using their local data and then with the learned model parameters create a central model. The objective of the central model is to obtain the information learned by the separate models, without ever accessing the raw data itself. This is achieved by averaging the separate models’ weights into the central model. FL thus facilitates opportunities to train a model on large amounts of data from several sources, without the need of having access to the data itself. If one can create a model with this methodology, that is not significantly worse than a model trained on the raw data, then positive effects such as strengthened data privacy, cross-enterprise collaboration and more could be attainable. In this work we have used a financial data set consisting of 25242 equity research reports, provided by Skandinaviska Enskilda Banken (SEB). Each report has a recommendation label, either Buy, Sell or Hold, making this a multi-class classification problem. To evaluate the feasibility of FL we fine-tune the pre-trained Transformer model AlbertForSequenceClassification [37] on the classification task. We create one baseline model using the entire data set and an FL model with different experimental settings, for which the data is distributed both uniformly and non-uniformly. The baseline model is used to benchmark the FL model. Our results indicate that the best FL setting only suffers a small reduction in performance. The baseline model achieves an accuracy of 83.5% compared to 82.8% for the best FL model setting. Further, we find that with an increased number of clients, the performance is worsened. We also found that our FL model was not sensitive to non-uniform data distributions. All in all, we show that FL results in slightly worse generalisation compared to the baseline model, while strongly improving on data privacy, as the central model never accesses the clients’ data. / Företags nyttjande av maskininlärning har de senaste åren ökat signifikant och för att kunna skapa högkvalitativa modeller krävs stora mängder data, vilket kan vara svårt att insamla. Parallellt med detta så ökar också den allmänna förståelsen för hur användandet av data missbrukas, vilket har lätt till ett ökat behov av starkare datasäkerhet. 2018 så trädde General Data Protection Regulation (GDPR) i kraft inom EU, vilken bland annat ställer krav på hur företag skall hantera persondata. Företag med maskininlärningsmodeller som på något sätt använder känslig eller personlig data behöver således ha fått tillgång till denna data i enlighet med de rådande lagar och regler som omfattar datahanteringen. Ytterligare ett datarelaterat problem är då företag önskar att skapa gemensamma maskininlärningsmodeller som skulle kräva att de delar deras bolagsdata [36, 38]. Denna uppsats kommer att undersöka hur Federerad Inlärning [38] kan användas för att skapa maskinlärningsmodeller som överkommer dessa datasäkerhetsrelaterade problem. Federerad Inlärning är en metod för att på ett decentraliserat vis träna maskininlärningsmodeller. Detta omfattar att låta flera aktörer träna en modell var. Varje enskild aktör tränar respektive modell på deras isolerade data och delar sedan endast modellens parametrar till en central modell. På detta vis kan varje enskild modell bidra till den gemensamma modellen utan att den gemensamma modellen någonsin haft tillgång till den faktiska datan. Givet att en modell, skapad med Federerad Inlärning kan uppnå liknande resultat som en modell tränad på rådata, så finns många positiva fördelar så som ökad datasäkerhet och ökade samarbeten mellan företag. Under arbetet har ett dataset, bestående av 25242 finansiella rapporter tillgängliggjort av Skandinaviska Ensilda Banken (SEB) använts. Varje enskild rapport innefattar en rekommendation, antingen Köp, Sälj eller Håll, vilket innebär att vi utför muliklass-klassificering. Med datan tränas den förtränade Transformermodellen AlbertForSequence- Classification [37] på att klassificera rapporterna. En Baseline-modell, vilken har tränats på all rådata och flera Federerade modellkonfigurationer skapades, där bland annat varierande fördelningen av data mellan aktörer från att vara jämnt fördelat till vara ojämnt fördelad. Resultaten visar att den bästa Federerade modellkonfigurationen endast presterar något sämre än Baseline-modellen. Baselinemodellen uppnådde en klassificeringssäkerhet på 83.5% medan den bästa Federerade modellen uppnådde 82.8%. Resultaten visar också att den Federerade modellen inte var känslig mot att variera fördelningen av datamängd mellan aktorerna, samt att med ett ökat antal aktörer så minskar klassificeringssäkerheten. Sammanfattningsvis så visar vi att Federerad Inlärning uppnår nästan lika goda resultat som Baseline-modellen, samtidigt så bidrar metoden till avsevärt bättre datasäkerhet då den centrala modellen aldrig har tillgång till rådata.

Patent monitoring is about tracking the upcoming inventions in a particular field, predicting future trends, and specific intellectual property rights of interest. It is the process of finding relevant patents on a particular topic based on a specific query. With patent monitoring, one can keep them updated on the new technology in the market. Also, they can find potential licensing opportunities for their inventions. The outputs of patent monitoring are essential for companies, academics, and inventors looking forward to using the latest patents that can enhance further innovation. Nevertheless, there is no widely accepted best approach to patent monitoring. Usually, most patent monitoring systems are based on complex search and find, often leading to insignificant hit rates and highly human intervention. As the number of patents published each year increases massively and with patents being critical to accelerating innovation, the current approach to patent monitoring has two main drawbacks. Firstly, human-driven patent monitoring is time consuming and expensive process. In addition, there is a risk of overlooking interesting documents due to inadequate searching tools and processes, which could cost companies fortunes while at the same time hindering further innovation and creativity. This thesis presents a smart patent monitoring assistant tool that applies natural language processing. The use of several natural language processing methods is investigated to find, classify and rank relevant documents. The tool was trained on a dataset that contains the title, abstract, and claims of patent documents. Given a dataset of patent documents, the aim of this thesis is to create a tool that can classify patents into two classes relevant and not relevant. Furthermore, the tool can rank documents based on relevancy. The evaluation result of the tool gave satisfying results when it came to receiving the expected patents. In addition, there is a significant improvement in terms of performance for memory usage and the time it took to train the model and get results. / Patentövervakning handlar om att övervaka kommande uppfinningar, förutsäga framtida trender, eller specifika immateriella rättigheter och används för att hitta relevanta patent inom ett visst område. Med patentövervakning är det möjligt att hålla patent uppdaterade enligt den senaste tekniken på marknaden samt att hitta potentiella möjligheter att licensiera innehavda patent till tredje part. Målgruppen för patentövervakning är företag, akademiker, och uppfinnare som vill hitta de senaste patenten för att uppnå maximal innovation. Dock finns det ingen generell metod för att bedriva patentövervakning. Vanligtvis används komplexa sökmetoder som resulterar i undermåliga resultat och kräver manuellt ingripande. I och med att andelen patent ökar varje år har nuvarande metod två huvudsakliga nackdelar. Till att börja med är mänsklig patentövervakning en tidskrävande och dyr process. Vidare är det en betydande risk att missa viktiga eller på andra sätt intressanta dokument till följd av en bristande sökprocess. Detta kan möjligtvis resultera i att företag missar stora möjligheter samt utebliven innovation och kreativitet. Detta arbete presenterar ett smart verktyg för patentövervakning baserat på natural language processing. Vi analyserar användningen av ett flertal processer för att hitta, klassificera, och rangordna relevant dokument. Verktyget tränades på ett dataset som innehåller patentets titel, abstrakt, och vad patentet gör anspråk på. Givet ett godtyckligt dataset är målet med detta arbete att utveckla ett verktyg med förmågan att klassificera relevanta och icke-relevanta patent samt rangordna dessa utifrån relevans. Resultatet visar att verktyget gav tillfredsställande gällande att hitta önskvärda patent. Vidare uppnåddes en signifikant förbättring när det gäller prestanda för minnesanvändning och tiden som krävs för att träna modeller och erhålla resultat.

Recent advancements in Transformers have unlocked a new relational analysis technique for Reinforcement Learning (RL). This thesis researches the models for DownLink Link Adaptation (DLLA). Radio resource management methods such as DLLA form a critical facet for radio-access networks, where intricate optimization problems are continuously resolved under strict latency constraints in the order of milliseconds. Although previous work has showcased improved downlink throughput in an online RL approach, time dependence of DLLA obstructs its wider adoption. Consequently, this thesis ventures into uncharted territory by extending the DLLA framework with sequence modelling to fit the Transformer architecture. The objective of this thesis is to assess the efficacy of an autoregressive sequence modelling based offline RL Transformer model for DLLA using a Decision Transformer. Experimentally, the thesis demonstrates that the attention mechanism models environment dynamics effectively. However, the Decision Transformer framework lacks in performance compared to the baseline, calling for a different Transformer model. / De senaste framstegen inom Transformers har möjliggjort ny teknik för Reinforcement Learning (RL). I denna uppsats undersöks modeller för länkanpassning, närmare bestämt DownLink Link Adaptation (DLLA). Metoder för hantering av radioresurser som DLLA utgör en kritisk aspekt för radioåtkomstnätverk, där invecklade optimeringsproblem löses kontinuerligt under strikta villkor kring latens och annat, i storleksordningen millisekunder. Även om tidigare arbeten har påvisat förbättrad länkgenomströmning med en online-RL-metod, så gäller att tidsberoenden i DLLA hindrar dess bredare användning. Följaktligen utökas här DLLA-ramverket med sekvensmodellering för att passa Transformer-arkitekturer. Syftet är att bedöma effekten av en autoregressiv sekvensmodelleringsbaserad offline-RL-modell för DLLA med en Transformer för beslutsstöd. Experimentellt visas att uppmärksamhetsmekanismen modellerar miljöns dynamik effektivt. Men ramverket saknar prestanda jämfört med tidigare forsknings- och utvecklingprojekt, vilket antyder att en annan Transformer-modell krävs.

Quality inspection is an essential part of almost any industrial production line. However, designing customized solutions for defect detection for every product can be costlyfor the production line. This is especially the case for short-series production, where theproduction time is limited. That is because collecting and manually annotating the training data takes time. Therefore, a possible method for defect detection using only synthetictraining data focused on geometrical defects is proposed in this thesis work. The methodis partially inspired by previous related work. The proposed method makes use of aninstance segmentation model and pose-estimator. However, this thesis work focuses onthe instance segmentation part while using a pre-trained pose-estimator for demonstrationpurposes. The synthetic data was automatically generated using different data augmentation techniques from a 3D model of a given object. Moreover, Mask R-CNN was primarilyused as the instance segmentation model and was compared with a rival model, HTC. Thetrials show promising results in developing a trainable general-purpose defect detectionpipeline using only synthetic data

Gaze estimation, a vital component in numerous applications such as humancomputer interaction, virtual reality, and driver monitoring systems, is the process of predicting the direction of an individual’s gaze. The predominant methods for gaze estimation can be broadly classified into intrusive and nonintrusive approaches. Intrusive methods necessitate the use of specialized hardware, such as eye trackers, while non-intrusive methods leverage images or recordings obtained from cameras to make gaze predictions. This thesis concentrates on appearance-based gaze estimation, specifically within the non-intrusive domain, employing various deep learning models. The primary focus of this study is to compare the efficacy of Vision Transformers (ViTs), a recently introduced architecture, with Convolutional Neural Networks (CNNs) for gaze estimation on RGB images. Performance evaluations of the models are conducted based on metrics such as the angular gaze error, stimulus distance error, and model size. Within the realm of ViTs, two variants are explored: pure ViTs and hybrid ViTs, which combine both CNN and ViT architectures. Throughout the project, both variants are examined in different sizes. Experimental results demonstrate that all pure ViTs underperform in comparison to the baseline ResNet-18 model. However, the hybrid ViT consistently emerges as the best-performing model across all evaluation datasets. Nonetheless, the discussion regarding whether to deploy the hybrid ViT or stick with the baseline model remains unresolved. This uncertainty arises because utilizing an exceedingly large and slow model, albeit highly accurate, may not be the optimal solution. Hence, the selection of an appropriate model may vary depending on the specific use case. / Ögonblicksbedömning, en avgörande komponent inom flera tillämpningar såsom människa-datorinteraktion, virtuell verklighet och övervakningssystem för förare, är processen att förutsäga riktningen för en individs blick. De dominerande metoderna för ögonblicksbedömning kan i stort sett indelas i påträngande och icke-påträngande tillvägagångssätt. Påträngande metoder kräver användning av specialiserad hårdvara, såsom ögonspårare, medan ickepåträngande metoder utnyttjar bilder eller inspelningar som erhållits från kameror för att göra bedömningar av blicken. Denna avhandling fokuserar på utseendebaserad ögonblicksbedömning, specifikt inom det icke-påträngande området, genom att använda olika djupinlärningsmodeller. Studiens huvudsakliga fokus är att jämföra effektiviteten hos Vision Transformers (ViTs), en nyligen introducerad arkitektur, med Convolutional Neural Networks (CNNs) för ögonblicksbedömning på RGB-bilder. Prestandautvärderingar av modellerna utförs baserat på metriker som den vinkelmässiga felbedömningen av blicken, felbedömning av stimulusavstånd och modellstorlek. Inom ViTs-området utforskas två varianter: rena ViTs och hybrid-ViT, som kombinerar både CNN- och ViT-arkitekturer. Under projektet undersöks båda varianterna i olika storlekar. Experimentella resultat visar att alla rena ViTs presterar sämre jämfört med basmodellen ResNet-18. Hybrid-ViT framstår dock konsekvent som den bäst presterande modellen över alla utvärderingsdatauppsättningar. Diskussionen om huruvida hybrid-ViT ska användas eller om man ska hålla sig till basmodellen förblir dock olöst. Denna osäkerhet uppstår eftersom användning av en extremt stor och långsam modell, även om den är mycket exakt, kanske inte är den optimala lösningen. Valet av en lämplig modell kan därför variera beroende på det specifika användningsområdet.