Additive manufacturing (AM) is a relatively new domain of manufacturing processes that began with its first patent in 1986. Since then, AM processes quickly grew in popularity due to their flexibility, superior efficiency in high mix low volume manufacturing settings, and lower material costs compared to more subtractive processes. Despite its increasing popularity, AM processes remain behind subtractive processes in terms of quality and the speed at which new technologies are integrated. Introducing Industry 4.0 technologies is an excellent opportunity to address the need for quality assurance tools for AM processes. First, the question of how the quality of additively manufactured parts can be increased to match parts created through subtractive processes must be asked. In this dissertation, two machine learning (ML) models are developed and utilized in a federated environment to mimic what one would see in a production setting. The proposed models increase AM part quality by (1) predicting the resulting geometry of an AM process and (2) compensating for geometric errors by altering the initial stereolithography (STL) file before slicing. In addition to performing geometric error prediction and compensation, the models were enhanced to be resilient to changes in geometry by training on segments of a 3D object rather than the whole object. Next, process parameters from fused-filament fabrication (FFF) processes were added to the ML models to add resilience process parameter variance. Lastly, the ML models were deployed in a federated environment created from three FFF 3D printers that collaboratively created a dataset for the ML models. Collectively, these works expand the research area created by AM, federated learning, and error compensation. This proposal addresses research gaps in the current literature by first setting the prediction and compensation resolution of voxel-based ML methods to a static 100 µm, thereby reducing the error associated with each voxel. Secondly, process parameters are introduced to the model, further increasing prediction and compensation accuracy compared to predicting on the geometry alone. Lastly, the models are deployed in a federated AM environment with multiple 3D printers acting as clients to reduce each client's time spent generating data while maintaining model performance. / Doctor of Philosophy / Additive manufacturing (AM) is a relatively new field where parts are created by extruding material to build a product in the desired shape. A key advantage of such a process is that it is more flexible than those subtractive processes, which remove material from a part. On the other hand, parts produced by AM processes generally have lower quality due to the very specific environments necessary to obtain high-quality parts. Because there is an increased desire to make customized parts (high mix) in small amounts (low volume), AM processes are seeing a rise in popularity, but there is still a need to improve the quality of these produced parts. Furthermore, these environments where AM is utilized generally have multiple 3D printers that manufacturers can leverage to create comprehensive datasets for model development. This dissertation uses machine learning (ML) to collect data from AM processes and reduce AM process errors. By comparing the process's input with the process's output, an ML model can estimate the result of the AM process, including potential defects. This dissertation addresses research gaps in current literature by reducing the error associated with converting the input and output 3D objects to voxels, using parameters to the AM process in the ML models, and using the ML models with 3D printers in a networked environment while forbidding sharing private data.

Allt fler misshandelsbrott sker i Sverige enligt Brå. För att reducera detta kan det som fångats på övervakningskameror användas i brottsutredningar, för att senare användas som bevismaterial till att döma den eller de skyldiga till brottet. Genom att optimera övervakningen kan företag använda sig av automatiserad igenkänning. Automatisering för igenkänningen av normala kontra onormala beteenden går att lösa med djupinlärning. Syftet med denna undersökning är att finna en lämplig modell som kan identifiera det onormala beteendet (till exempel ett slagsmål). Modell arkitekturen som användes under projektet var 3D ResNet, eftersom den klara av en djupare arkitektur. Ett djupare nätverk, innebär bättre prediktion av problemet. 3DResNet-34 var den modell arkitekturen som gav högst noggrannhet med 93,33%. Implementering av projektet utfördes i ramverket PyTorch. Undersökningen har visat att med           hjälp av överförd inlärning går det att återanvända kunskap från förtränade modeller och applicera dessa kunskaper på det aktuella problemet. Detta bidrar till en mer pålitligare modell med noggrann prediktion på nytt övervaknings           material. / According to Brå, more assault crimes are taking place in Sweden. To reduce this, information that was captured on surveillance cameras can be used in criminal investigations, to convict the perpetrator or perpetrators of the crime. To optimize monitoring, companies can use automation. Automation of the recognition of normal versus abnormal activities can be solved with deep learning. The purpose of this study is to find a suitable model that can identify               the abnormal activity (for example, a fight). The model architecture used during the project was 3D ResNet, because it was capable of handling deeper architectures. Having a deeper network means better prediction of the problem.           3D ResNet-34 was the model architecture that gave the highest accuracy with 93,33%. Implementation of the project was carried out in the framework of PyTorch. The study has shown that with the help of transfer learning it is possible to transfer knowledge from pre-trained models and apply this          knowledge to the current problem. This contributes to a more reliable model with accurate prediction for new surveillance footage.

Subsurface porous media, like aquifers, petroleum reservoirs, and geothermal systems, are vital for natural resources and environmental management. Extensive research has been conducted to understand flow and transport in these media, addressing challenges in hydrocarbon extraction, carbon storage and waste management. Classifying the type of porous media (e.g., sandstone, carbonate) is often the first step in the rock characterization process, and it provides critical information regarding the physical properties of the porous media. Therefore, we utilize multivariate statistical methods with discriminant analysis to categorize porous media samples which proved to be efficient by achieving excellent classification accuracy on testing datasets and served as a surrogate tool to study key porous media characteristics. While recent advances in three-dimensional (3D) imaging of core samples have enabled digital subsurface characterization, the exorbitant computational cost associated with direct numerical simulation in 3D remains a persistent challenge. In contrast, machine learning (ML) models are much more efficient, though their use in subsurface characterization is still in its infancy. Therefore, we introduce a novel 3D convolution neural network (CNN) for end-to-end prediction of permeability. By increasing dataset size, diversity, and optimizing the network architecture, our model surpasses the accuracy of existing 3D CNN models for permeability prediction. It demonstrates excellent generalizability, accurately predicting permeability in previously unseen samples. However, despite the efficiency of the developed 3D CNN model for accurate and fast permeability prediction, its utility remains limited to small subdomains of the digital rock samples. Therefore, we introduce an upscaling technique using a new analytical solution to calculate effective permeability in a 3D digital rock composed of 2 × 2 × 2 anisotropic cells. By incorporating this solution into physics-informed neural network (PINN) models, we achieve highly accurate results. Even when upscaling previously unseen samples at multiple levels, the PINN with the physics-informed module maintains excellent accuracy. This advancement enhances the capability of ML models, like 3D CNN, for efficient and accurate digital rock analysis at the core scale. After successfully applying ML models in permeability prediction, we now extend their application to another important parameter in subsurface engineering projects: effective thermal conductivity, which is a key parameter in engineering projects like radioactive waste repositories, geothermal energy production, and underground energy storage. To address the need for large training data and processing power in ML models, we propose a novel framework based on transfer learning. This approach allows prior knowledge from previous applications to be transferred, resulting in faster and more efficient implementation of new relevant applications. We introduce CNN models trained on various porous media samples that leverage transfer learning to predict porous media sample thermal conductivity accurately. Our approach reduces training time, processing power, and data requirements, enabling effective prediction and analysis of porous media properties such as permeability and thermal conductivity. It also facilitates the application of ML to other properties, improving efficiency and accuracy. / Thesis / Doctor of Philosophy (PhD)

The goal of this thesis has been global geolocalization using only visual input and a 3D database for reference. In recent years Convolutional Neural Networks (CNNs) have seen huge success in the task of classifying images. The flattened tensors at the final layers of a CNN can be viewed as vectors describing different input image features. Two networks were trained so that satellite and aerial images taken from different views of the same location had feature vectors that were similar. The networks were also trained so that images taken from different locations had different feature vectors. After training, the position of a given aerial image can then be estimated by finding the satellite image with a feature vector that is the most similar to that of the aerial image.  A previous method called Where-CNN was used as a baseline model. Batch-Hard triplet loss, the Adam optimizer, and a different CNN backbone were tested as possible augmentations to this method. The models were trained on 2640 different locations in Linköping and Norrköping. The models were then tested on a sequence of 4411 query images along a path in Jönköping. The search region had 1449 different locations constituting a total area of 24km2.  In Top-1% accuracy, there was a significant improvement over the baseline, increasing from 61.62% accuracy to 88.62%. The environment was modeled as a Hidden Markov Model to filter the sequence of guesses. The Viterbi algorithm was then used to find the most probable path. This filtering procedure reduced the average error along the path from 2328.0 m to just 264.4 m for the best model. Here the baseline had an average error of 563.0 m after filtering.  A few different 3D methods were also tested. One drawback was that no pretrained weights existed for these models, as opposed to the 2D models, which were pretrained on the ImageNet dataset. The best 3D model achieved a Top-1% accuracy of 70.41%. It should be noted that the best 2D model without using any pretraining achieved a lower Top-1% accuracy of 49.38%. In addition, a 3D method for efficiently doing convolution on sparse 3D data was presented. Compared to the straight-forward method, it was almost 2.5 times faster while still having comparable accuracy at individual query prediction.  While there was a significant improvement over the baseline, it was not significant enough to provide reliable and accurate localization for individual images. For global navigation, using the entire Earth as search space, the information in a 2D image might not be enough to be uniquely identifiable. However, the 3D CNN techniques tested did not improve the results of the pretrained 2D models. The use of more data and experimentation with different 3D CNN architectures is a direction in which further research would be exciting.