Spelling suggestions: "subject:"surrogate model."" "subject:"surrogates model.""
51 |
Image Segmentation, Parametric Study, and Supervised Surrogate Modeling of Image-based Computational Fluid DynamicsIslam, Md Mahfuzul 05 1900 (has links)
Indiana University-Purdue University Indianapolis (IUPUI) / With the recent advancement of computation and imaging technology, Image-based computational fluid dynamics (ICFD) has emerged as a great non-invasive capability to study biomedical flows. These modern technologies increase the potential of computation-aided diagnostics and therapeutics in a patient-specific environment. I studied three components of this image-based computational fluid dynamics process in this work.
To ensure accurate medical assessment, realistic computational analysis is needed, for which patient-specific image segmentation of the diseased vessel is of paramount importance. In this work, image segmentation of several human arteries, veins, capillaries, and organs was conducted to use them for further hemodynamic simulations. To accomplish these, several open-source and commercial software packages were implemented.
This study incorporates a new computational platform, called InVascular, to quantify the 4D velocity field in image-based pulsatile flows using the Volumetric Lattice Boltzmann Method (VLBM). We also conducted several parametric studies on an idealized case of a 3-D pipe with the dimensions of a human renal artery. We investigated the relationship between stenosis severity and Resistive index (RI). We also explored how pulsatile parameters like heart rate or pulsatile pressure gradient affect RI.
As the process of ICFD analysis is based on imaging and other hemodynamic data, it is often time-consuming due to the extensive data processing time. For clinicians to make fast medical decisions regarding their patients, we need rapid and accurate ICFD results. To achieve that, we also developed surrogate models to show the potential of supervised machine learning methods in constructing efficient and precise surrogate models for Hagen-Poiseuille and Womersley flows.
|
52 |
Fast Simulations of Radio Neutrino Detectors : Using Generative Adversarial Networks and Artificial Neural NetworksHolmberg, Anton January 2022 (has links)
Neutrino astronomy is expanding into the ultra-high energy (>1017eV) frontier with the use of in-ice detection of Askaryan radio emission from neutrino-induced particle showers. There are already pilot arrays for validating the technology and the next few years will see the planning and construction of IceCube-Gen2, an upgrade to the current neutrino telescope IceCube. This thesis aims to facilitate that planning by providing faster simulations using deep learning surrogate models. Faster simulations could enable proper optimisation of the antenna stations providing better sensitivity and reconstruction of neutrino properties. The surrogates are made for two parts of the end-to-end simulations: the signal generation and the signal propagation. These two steps are the most time-consuming parts of the simulations. The signal propagation is modelled with a standard fully connected neural network whereas for the signal generation a conditional Wasserstein generative adversarial network is used. There are multiple reasons for using these types of models. For both problems the neural networks provide the speed necessary as well as being differentiable -both important factors for optimisation. Generative adversarial networks are used in the signal generation because of the inherent stochasticity in the particle shower development that leads to the Askaryan radio signal. A more standard neural network is used for the signal propagation as it is a regression task. Promising results are obtained for both tasks. The signal propagation surrogate model can predict the parameters of interest at the desired accuracy, except for the travel time which needs further optimisation to reduce the uncertainty from 0.5 ns to 0.1 ns. The signal generation surrogate model predicts the Askaryan emission well for the limited parameter space of hadronic showers and within 5° of the Cherenkov cone. The two models provide a first step and a proof of concept. It is believed that the models can reach the required accuracies with more work.
|
53 |
Development of a Design Tool in CAD for Fused Deposition Modelled Coolant Nozzles in Grinding : Design automation of coolant nozzlesNeguembor, Joachim January 2022 (has links)
This thesis covers the process of automating the design of coolant nozzles used for cylindrical grinding. Coolant nozzles are used to supply coolant, an oil and water mixture used to cool the metal workpiece and lubricate the grinding wheel. In the automotive industry, grinding is used to reduce the surface roughness of the workpiece. However, a large amount of heat is generated, risking the heat treatment of the steel to be compromised, for this, coolant is supplied to minimize the heat caused by friction. A nozzle is used, aiming a jet to the zone that generates heat. Commonly used nozzles are adjustable, leading to variation in cooling performance if misaligned. The design of fixed nozzles is developed in this thesis to reduce variation and automatise the design for multiple applications. The automatically designed nozzles are fused deposition modeled and tested. The design automation tool is tested repeatedly and improved successively in the span of the thesis. This lead to a great extent of implementation of design automation. Which lead to a facilitation in reaching of the work zone and avoid obstacles. Also, the tool managed to create nozzle tubes for a multitude of machines. The tool is able to generate, aim, orient, and individually dimension multi-nozzle tubes. Design of Experiment methodology is implemented to find nozzle designs with improved velocity and flow rate and minimize the air mixture with the coolant. Several nozzle designs are tested and fitted into a surrogate model that is, in turn, optimized. The results of the tests led to a greater understanding of how the nozzle geometry restricts the flow rate when attempts of reaching higher velocities of the coolant jet are made. The surrogate models created, also made it possible to find the range of designs which best suits different applications, whereby a Pareto front was able to be populated with a range of different designs alternating in flow rate, velocity and coherency ratio.
|
54 |
[pt] MODELO SUBSTITUTO PARA FLUXO NÃO SATURADO VIA REGRESSÃO POLINOMIAL EVOLUCIONÁRIA: CALIBRAÇÃO COM O ENSAIO DE INFILTRAÇÃO MONITORADA / [en] SURROGATE MODEL FOR UNSATURATED FLOW THROUGH EVOLUTIONARY POLYNOMIAL REGRESSION: CALIBRATION WITH THE MONITORED INFILTRATION TESTRUAN GONCALVES DE SOUZA GOMES 26 February 2021 (has links)
[pt] A análise de fluxo de água sob condição transiente não saturada requer o conhecimento das propriedades hidráulicas do solo. Essas relações constitutivas, denominadas curva característica e função de condutividade hidráulica, são descritas através de modelos empíricos que geralmente possuem vários parâmetros que devem ser calibrados com relação a dados coletados. Muitos dos parâmetros nos modelos constitutivos não podem ser medidos diretamente em campo ou laboratório, mas somente podem ser inferidos de forma significativa a partir de dados coletados e da modelagem inversa. Para obter os parâmetros do solo com a análise inversa, um algoritmo de otimização de busca local ou global pode ser aplicado. As otimizações globais são mais capazes de encontrar parâmetros ótimos, no entanto,
a solução direta, por meio da modelagem numérica é computacionalmente custosa. Portanto, soluções analíticas (modelo substituto) podem superar essa falha acelerando o processo de otimização. Nesta dissertação, apresentamos a Regressão Polinomial Evolucionária (EPR) como uma ferramenta
para desenvolver modelos substitutos do fluxo não saturado. Um rico conjunto de dados de parâmetros hidráulicos do solo é usado para calibrar o nosso modelo, e dados do mundo real são utilizados para validar nossa metodologia. Nossos resultados demonstram que o modelo da EPR prevê com precisão os dados de carga de pressão. As simulações do modelo se mostram concordantes com as simulações do programa Hydrus. / [en] Water flow analyses under transient soil hydraulic conditions require knowledge of the soil hydraulic properties. These constitutive relationships, named soil-water characteristic curve (SWCC) and hydraulic conductivity function (HCF) are described through empirical models which generally have several parameters that must be calibrated against collected data. Many of the parameters in SWCC and HCF models cannot be directly measured in field or laboratory but can only be meaningfully inferred
from collected data and inverse modeling. In order to obtain the soil parameters with the inverse process, a local or global optimization algorithm may be applied. Global optimizations are more capable of fiding optimum parameters, however the direct solution through numerical modeling are time consuming. Therefore, analytical solutions (surrogate models) may overcome this shortcomming by accelerating the optimization process. In this work we introduce Evolutionary Polynomial Regression (EPR) as a
tool to develop surrogate models of the physically-based unsaturated flow. A rich dataset of soil hydraulic parameters is used to calibrate our surrogate model, and real-world data are then utilized to validate our methodology. Our results demonstrate that the EPR model predicts accurately the observed pressure head data. The model simulations are shown to be in good agreement with the Hydrus software package.
|
55 |
Optimization based Analysis of Highly Automated Driving SimulationSatyamohan, Sharmila 08 July 2024 (has links)
In recent years, there have been remarkable advancements in automated driving systems. Consumer protection organizations, such as Euro NCAP, play a pivotal role in enhancing the overall safety of these modern vehicles. While previous emphasis has been on passive safety, the significance of active safety systems has surged in recent years. Evaluating the performance of these systems now relies on standardized test scenarios designed to simulate real-world accidents. Addressing this challenge, the future necessitates the incorporation of virtual methods to supplement traditional track tests. Given the complex nature of high-dimensional test cases, an exhaustive grid search is exceedingly time-consuming. In light of this challenge, we present a novel testing method utilizing search-based testing with Bayesian Optimization to efficiently navigate and explore the expansive search space of Euro NCAP CCR scenarios to identify the performance-critical scenarios.
The methodology incorporates the Brake Threat Number as a robust criticality metric within the fitness function, providing a reliable indicator for assessing the inevitability of collisions. Furthermore, the research utilizes a surrogate model derived from the evaluation points used by the optimization algorithm to determine the performance-critical boundary that separates the critical and the non-critical
scenarios. Additionally, this approach leverages the surrogate model for conducting sensitivity analysis, explaining the impact of individual parameters on the system’s output.
|
56 |
[en] ENSEMBLE GREY AND BLACK-BOX SYSTEM IDENTIFICATION FOR FRICTION MODELS / [pt] IDENTIFICAÇÃO DE SISTEMA CONJUNTO CAIXA-CINZA E CAIXA- PRETA PARA MODELOS DE ATRITOWALISSON CHAVES FERREIRA PINTO 11 June 2021 (has links)
[pt] A abstração matemática de um processo físico é essencial em problemas de engenharia, pois muitas vezes pode ser impraticável ou impossível realizar experimentos no sistema real. Além disso, modelos matemáticos são mais flexíveis que protótipos físicos, permitindo um rápido refinamento dos projetos
do sistema para otimizar várias medidas de desempenho. As aplicações dos modelos podem ser divididas em quatro partes, a saber: projeto, estimativa, controle e monitoramento. Algumas aplicações específicas são i) simulações, ii) soft sensors, iii) avaliação de desempenho, iv) controle estatístico de qualidade
e v) detecção e diagnóstico de falhas. Este trabalho visa então: i) desenvolver diferentes classes de modelos capazes de simular com precisão a variável de saída de um sistema, ii) avaliar a eficiência dos algoritmos de otimização utilizados na tarefa de estimação de parâmetros, iii) avaliar qual modelo
de atrito é o mais adequado para descrever esse fenômeno em um sistema de posicionamento. Os resultados mostraram que o atrito no sistema de posicionamento apresenta comportamento não linear e assimétrico, já que alguns termos dos modelos de atrito relacionados às velocidades positiva e
negativa são significativamente diferentes um do outro. O resultado final do processo de otimização que usou um algoritmo de busca local foi altamente dependente das condições iniciais e do número de parâmetros estimados, o que elevou o erro de simulação. Entretanto, melhores estimativas da variável
de saída foram alcançadas quando essa abordagem foi combinada com outros modelos de diferentes classes. Através dessa última abordagem o erro relativo foi reduzido em mais de 20 porcento. As simulações realizadas com os parâmetros estimados pelos algoritmos evolucionários foram mais acuradas, eles foram capazes de reduzir o erro relativo em quase 30 porcento quando comparados com o
algoritmo de busca local. Considerando o segundo estudo de caso, o otimizador baseado em árvores de decisão se mostrou igualmente eficaz se comparado aos algoritmos evolucionários. O erro relativo das simulações usando os parâmetros estimados por esses algoritmos foi inferior a 8 porcento. Além disso, a forma do atrito reconstruído na segunda junta do manipulador robótico através dos parâmetros estimados pelos algoritmos está de acordo com o esperado. / [en] The mathematical abstraction of a physical process is essential in engineering problems, as it can often be impractical or impossible to perform experiments on the real system. Besides, mathematical models are more flexible than physical prototypes, allowing for quick refinement of system designs to optimize various performance measures. The applications of the models can be divided into four parts, namely: design, estimation, control and monitoring. Some specific applications are i) simulations, ii) soft sensors, iii) performance evaluation, iv) statistical quality control and, v) fault detection and diagnosis. This work aims to: i) develop different classes of models capable of accurately simulating the output variable of a system, ii) evaluate the efficiency of optimization algorithms used in the parameter estimation task, iii) assess which friction model is the most appropriate to describe this phenomenon in a positioning system. The results showed that the friction in the positioning system presents a nonlinear and asymmetric behavior since some terms of the friction models related to the positive and negative velocities are significantly different from each other. The final result of the optimization process that used a local search algorithm was highly dependent on the initial conditions and the number of estimated parameters, which increased the simulation error. However, better estimates of the output variable were achieved when this approach
was combined with other models of different classes. Through this last approach, the relative error was reduced by more than 20 percent. The simulations performed with the parameters estimated by the evolutionary algorithms were more accurate, they were able to reduce the relative error by almost 30 percent when compared with the local search algorithm. Considering the second case study,
the decision tree-based optimizer proved to be equally effective compared to evolutionary algorithms. The relative error of the simulations using the parameters estimated by these algorithms was less than 8 percent. Besides, the shape of the friction reconstructed in the second joint of the robotic manipulator through the parameters estimated by the algorithms is in accordance with the expected.
|
57 |
Prediction and minimization of excessive distortions and residual stresses in compliant assembled structuresYoshizato, Anderson 26 May 2020 (has links)
The procedure of joining flexible or nonrigid parts using applied loads is called compliant assembly, and it is widely used in automotive, aerospace, electronics, and appliance manufacturing. Uncontrolled assembly processes may produce geometric errors that can exceed design tolerances and induce an increment of elastic energy in the structure due to the accumulation of internal stresses. This condition might create unexpected deformations and residual stress distributions across the structure that compromise product functionality. This thesis presents a method based on nonlinear Finite Element Analysis (FEA), metamodelling, and optimization techniques to provide accurate and on-time shimming strategies to support the definition of optimum assembly strategies. An example of the method on a typical aerospace wing box structure is demonstrated in the present study. The delivered outputs intend to support the production line by anticipating the response of the structure under a specific assembly condition and presenting alternative assembly strategies that can be applied to address eventual predicted issues on product requirements. / Graduate
|
58 |
Elektromagnetická analýza a modelování asynchronního stroje s plným rotorem / Electromagnetic analysis and modeling of a solid rotor induction machineBílek, Vladimír January 2021 (has links)
Tato diplomová práce se zabývá elektromagnetickou analýzou a modelováním asynchronního stroje s plným rotorem. Tato práce tedy zahrnuje literární rešerši na téma vysokootáčkových elektrických strojů s porovnáním s klasickými elektrickými stroji s převodovkou a popisem jejich výhod či nevýhod, rozdělení vysokootáčkových elektrických strojů s plnými rotory a srovnání jejich výhod či nevýhod, kde se tato práce nejvíce soustřeďuje na vysokootáčkové asynchronní stroje s plnými rotory a jejich použití v průmyslu. Dále se tato práce zabývá metodami výpočtu elektrických asynchronních strojů s plnými rotory. Proto jsou zde uvedeny a popsány metody výpočtu stroje mezi které patří analytické metody i metoda konečných prvků. Vzhledem k povaze elektrických strojů s plnými rotory je hlavně kladen důraz v této práci na výpočet stroje pomocí metody konečných prvků ve 2D prostoru s využitím korekčních činitelů konců plných rotorů, které jsou zde velmi detailně popsány a rozděleny. Na základě dostupné literatury je vypočítaný elektrický stroj s plným rotorem pomocí MKP analýzy. Elektromagnetický výpočet stroje je automatizován pomocí skriptu vytvořeného v Pythonu. Dalším hlavním cílem této práce je popis tzv. náhradních modelů, uvedení jejich výhod či nevýhod, použití v jiných průmyslových odvětvích a hlavně použití náhradních modelů na elektrický stroj s plným rotorem. S využitím náhradních modelů je dále optimalizovaný vybraný asynchronní stroj s plným rotorem a to pomocí programů SymSpace a Optimizer. Pro samotnou optimalizaci byly uvažovány 3 návrhy stroje, které byly na závěr mezi sebou porovnány a to hlavně z hlediska jejich elektromagnetického výkonu.
|
59 |
IMAGE SEGMENTATION, PARAMETRIC STUDY, AND SUPERVISED SURROGATE MODELING OF IMAGE-BASED COMPUTATIONAL FLUID DYNAMICSMD MAHFUZUL ISLAM (12455868) 12 July 2022 (has links)
<p> </p>
<p>With the recent advancement of computation and imaging technology, Image-based computational fluid dynamics (ICFD) has emerged as a great non-invasive capability to study biomedical flows. These modern technologies increase the potential of computation-aided diagnostics and therapeutics in a patient-specific environment. I studied three components of this image-based computational fluid dynamics process in this work.</p>
<p>To ensure accurate medical assessment, realistic computational analysis is needed, for which patient-specific image segmentation of the diseased vessel is of paramount importance. In this work, image segmentation of several human arteries, veins, capillaries, and organs was conducted to use them for further hemodynamic simulations. To accomplish these, several open-source and commercial software packages were implemented. </p>
<p>This study incorporates a new computational platform, called <em>InVascular</em>, to quantify the 4D velocity field in image-based pulsatile flows using the Volumetric Lattice Boltzmann Method (VLBM). We also conducted several parametric studies on an idealized case of a 3-D pipe with the dimensions of a human renal artery. We investigated the relationship between stenosis severity and Resistive index (RI). We also explored how pulsatile parameters like heart rate or pulsatile pressure gradient affect RI.</p>
<p>As the process of ICFD analysis is based on imaging and other hemodynamic data, it is often time-consuming due to the extensive data processing time. For clinicians to make fast medical decisions regarding their patients, we need rapid and accurate ICFD results. To achieve that, we also developed surrogate models to show the potential of supervised machine learning methods in constructing efficient and precise surrogate models for Hagen-Poiseuille and Womersley flows.</p>
|
60 |
[en] FAST AND ACCURATE SIMULATION OF DEFORMABLE SOLID DYNAMICS ON COARSE MESHES / [pt] SIMULAÇÃO RÁPIDA E PRECISA DE DINÂMICA DE SÓLIDOS DEFORMÁVEIS EM MALHAS POUCO REFINADASMATHEUS KERBER VENTURELLI 23 May 2024 (has links)
[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.
|
Page generated in 0.0615 seconds