Örebro University and Akademiska Hus have started an initiative towards smart buildings. Avery important role to this is Digital Twin for buildings. Digital twin for buildings is a virtualcopy of a physical building. And by adding a Data Driven Simulation System, an even moresmart building could be achieved. Given a humidity-, temperature-, illuminance- and motionsensor in a specific corridor at the Örebro University, this thesis will ascertain what can bedone by creating a Data Driven Simulation System and using these sensors to achieve thedesired smart building. In this thesis, a simulation was created with simulated sensors andpedestrians. The simulation is a clone of the real world, by using real life sensors andapplying the data to the simulated sensors, this was partially achieved.

Accelerating physics-based simulations has been an evergreen topic across different scientific communities. This dissertation is devoted to this subject addressing bottlenecks in state-of-the-art approaches to the simulation of fluids of large-scale scenes, viscous threads, magnetic fluids, and the simulation of fibers and thin structures. The contributions within the thesis are rooted in mathematical modeling and numerical simulation as well as in machine learning. The first part deals with the simulation of incompressible flow in a multigrid fashion. For the variational viscous equation, geometric multigrid is inefficient. An Unsmoothed Aggregation Algebraic Multigrid method is devised with a multi-color Gauss-Seidel smoother, which consistently solves this equation in a few iterations for various material parameters. This framework is 2.0 to 14.6 times faster compared to the state-of-the-art adaptive octree solver in commercial software for the large-scale simulation of both non-viscous and viscous flow. In the second part, a new physical model is devised to accelerate the macroscopic simulation of magnetic fluids. Previous work is based on the classical Smoothed-Particle Hydrodynamics (SPH) method and a Kelvin force model. Unfortunately, this model results in a force pointing outwards causing significant levitation problems limiting the application of more advanced SPH frameworks such as Divergence-Free SPH (DFSPH) or Implicit Incompressible SPH (IISPH). This shortcoming has been addressed with this new current loop magnetic force model resulting in more stable and fast simulations of magnetic fluids using DFSPH and IISPH. Following a different trajectory, the third part of this thesis aims for the acceleration of iterative solvers widely used to accurately simulate physical systems. We speedup the simulation for rod dynamics with Graph Networks by predicting the initial guesses to reduce the number of iterations for the constraint projection part of a Position-based Dynamics solver. Compared to existing methods, this approach guarantees long-term stability and therefore leads to more accurate solutions.

Traffic congestion is a one hundred billion dollar problem in the US. The cost of congestion has been trending upward over the last few decades, but has experienced slight decreases in recent years partly due to the impact of congestion reduction strategies. The impact of these strategies is however largely experienced on freeways and not arterials. This discrepancy in impact is partially linked to the lack of real-time, arterial traffic information. Toward this end, this research effort seeks to address the lack of arterial traffic information. To address this dearth of information, this effort developed a methodology to provide accurate estimates of arterial performance measures to transportation facility managers and travelers in real-time. This methodology employs transmitted point sensor data to drive an online, microscopic traffic simulation model. The feasibility of this methodology was examined through a series of experiments that were built upon the successes of the previous, while addressing the necessary limitations. The results from each experiment were encouraging. They successfully demonstrated the method's likely feasibility, and the accuracy with which field estimates of performance measures may be obtained. In addition, the method's results support the viability of a "real-world" implementation of the method. An advanced calibration process was also developed as a means of improving the method's accuracy. This process will in turn serve to inform future calibration efforts as the need for more robust and accurate traffic simulation models are needed. The success of this method provides a template for real-time traffic simulation modeling which is capable of adequately addressing the lack of available arterial traffic information. In providing such information, it is hoped that transportation facility managers and travelers will make more informed decisions regarding more efficient management and usage of the nation's transportation network.

[pt] Apesar dos recentes avanços tecnológicos e do surgimento de computadores extremamente rápidos, a simulação numérica direta de escoamentos turbulentos ainda é proibitivamente cara para a maioria das aplicações de engenharia e até mesmo para algumas aplicações de pesquisa. As simulações utilizadas são, no geral, baseadas em grandezas médias e altamente dependentes de modelos de turbulência. Apesar de amplamente utilizados, tais modelos não conseguem prever adequadamente o escoamento médio em muitas aplicações, como o escoamento em um duto quadrado. Com o reflorescimento do Aprendizado de Máquina nos últimos anos, muita atenção está sendo dada ao uso de tais técnicas para substituir os modelos tradicionais de turbulência. Este trabalho estudou o uso de Redes Neurais como alternativa para aprimorar a simulação de escoamentos turbulentos. Para isso, a técnica PIV-Estereoscópico foi aplicada ao escoamento em um duto quadrado para obter dados experimentais de estatísticas do escoamento e campos médios de velocidade de 10 casos com diferentes números de Reynolds. Um total de 10 metodologias foram avaliadas para entender quais grandezas devem ser previstas por um algoritmo de aprendizado de máquina para obter simulações aprimoradas. A partir das metodologias selecionadas, excelentes resultados foram obtidos com uma Rede Neural treinada a partir dos dados experimentais para prever o termo perpendicular do Tensor de Reynolds e a viscosidade turbulenta. As simulações turbulentas auxiliadas pela Rede Neural retornaram campos de velocidade com menos de 4 por cento de erro, em comparação os dados medidos. / [en] Although the technological advances that led to the development of fast computers, the direct numerical simulation of turbulent flows is still prohibitively expensive to most engineering and even some research applications. The CFD simulations used worldwide are, therefore, based on averaged quantities and heavily dependent on mathematical turbulence models. Despite widely used, such models fail to proper predict the averaged flow in many practical situations, such as the simple flow in a square duct. With the re-blossoming of machine learning methods in the past years, much attention is being given to the use of such techniques as a replacement to the traditional turbulence models. The present work evaluated the use of Neural Networks as an alternative to enhance the simulation of turbulent flows. To this end, the Stereoscopic-PIV technique was used to obtain well-converged flow statistics and velocity fields for the flow in a square duct for 10 values of Reynolds number. A total of 10 methodologies were evaluated in a data-driven approach to understand what quantities should be predicted by a Machine Learning technique that would result in enhanced simulations. From the selected methodologies, accurate results could be obtained with a Neural Network trained from the experimental data to predict the nonlinear part of the Reynolds Stress Tensor and the turbulent eddy viscosity. The turbulent simulations assisted by the Neural Network returned velocity fields with less than 4 percent in error, in comparison with those previously measured.

<p dir="ltr">This research aims to address these critical challenges in uncertainty quantification. The objective is to employ data-driven approaches for UQ with physics models.</p>

The current thesis investigates data-driven simulation decision-making with field-quality consumer data. This is accomplished by outlining the benefits and uses of combining machine learning and simulation in the literature and by locating barriers to the use of machine learning (ML) in the simulation subsystems at a case study organization. Additionally, an implementation is carried out to demonstrate how Scania departments can use this technology to analyze their current data and produce results that support the exploration of the simulation space and the identification of potential design issues so that preventative measures can be taken during concept development. The thesis' findings provide an overview of the literature on the relationship between machine learning and simulation technologies, as well as limitations of using machine learning in simulation systems at large scale manufacturing organizations. Support vector machines, logistic regression, and Random Forest classifiers are used to demonstrate one possible use of machine learning in simulation.

[pt] Esta dissertação introduz um simulador híbrido inovador que combina um resolvedor de Equações Diferenciais Parciais (EDP) numérico de Elementos Finitos (FE) com uma Rede Neural de Passagem de Mensagens (MPNN) para realizar simulações de dinâmicas de sólidos deformáveis em malhas pouco refinadas. Nosso trabalho visa fornecer simulações precisas com um erro comparável ao obtido com malhas mais refinadas em discretizações FE,mantendo a eficiência computacional ao usar um componente MPNN que corrige os erros numéricos associados ao uso de uma malha menos refinada. Avaliamos nosso modelo focando na precisão, capacidade de generalização e velocidade computacional em comparação com um solucionador numérico de referência que usa malhas 64 vezes mais refinadas. Introduzimos um novo conjunto de dados para essa comparação, abrangendo três casos de referência numéricos: (i) deformação livre após um impulso inicial, (ii) alongamento e (iii)torção de sólidos deformáveis. Baseado nos resultados de simulação, o estudo discute as forças e fraquezas do nosso método. O estudo mostra que nosso método corrige em média 95,4 por cento do erro numérico associado à discretização, sendo até 88 vezes mais rápido que o solucionador de referência. Além disso, nosso modelo é totalmente diferenciável em relaçao a funções de custo e pode ser incorporado em uma camada de rede neural, permitindo que seja facilmente estendido por trabalhos futuros. Dados e código estão disponíveis em https://github.com/Kerber31/fast_coarse_FEM para investigações futuras. / [en] This thesis introduces a novel hybrid simulator that combines a numerical Finite Element (FE) Partial Differential Equation solver with a Message Passing Neural Network (MPNN) to perform simulations of deformable solid dynamics on coarse meshes. Our work aims to provide accurate simulations with an error comparable to that obtained with more refined meshes in FE discretizations while maintaining computational efficiency by using an MPNN component that corrects the numerical errors associated with using a coarse mesh. We evaluate our model focusing on accuracy, generalization capacity, and computational speed compared to a reference numerical solver that uses 64 times more refined meshes. We introduce a new dataset for this comparison, encompassing three numerical benchmark cases: (i) free deformation after an initial impulse, (ii) stretching, and (iii) torsion of deformable solids. Based on simulation results, the study thoroughly discusses our method s strengths and weaknesses. The study shows that our method corrects an average of 95.4 percent of the numerical error associated with discretization while being up to 88 times faster than the reference solver. On top of that, our model is fully differentiable in relation to loss functions and can be embedded into a neural network layer, allowing it to be easily extended by future work. Data and code are made available on https://github.com/Kerber31/fast_coarse_FEM for further investigations.