Theses and Dissertations

Fall 2017

Yuan, Yuan

reconcile the data. Among various approaches for optimization with uncertainty, chance constraint problem is a natural way to quantify the reliability of the solutions by setting a restriction on the level of the probability that the constraints are satisfied. In the case that multiple constraints should be

satisfied simultaneously, joint chance constraint is appropriate to model the uncertainties. However, joint chance constraint problem is generally intractable and a variety of methods are available to approximate it into tractable forms. Robust optimization with the distribution-free property is an approach

models. This thesis develops a novel robust optimization framework to consider the uncertain nonlinear optimization problems. The thesis provides practical applications as well. An economic optimization problem is investigated for steam generation and water distribution for SAGD (steam-assisted-gravity

Fall 2020

Fan, Lei

features can be extracted and the impact of outliers is alleviated by the latent variance scale. The next contribution of this thesis is to develop a semi-supervised model based on probability slow feature analysis to include the information from quality variables in the extracted latent features while

information of the process. With a probabilistic formulation, dynamic latent variable models, based on extracting slowly varying features, are developed in this thesis to address the aforementioned data irregularities, thus give reliable prediction results of quality variables that are otherwise difficult to

-distribution that has heavier tails, more weights can be assigned to the outliers thus they can be properly accounted for during modeling process. In feature extraction phase, a weighted Kalman gain is proposed since it violates the Gaussian assumption of the traditional Kalman filter. Smoother and slower

Spring 2022

Puliyanda, Anjana Thimmaiah

from the reactant to the product configurations. A self-supervised 3D convolutional neural network autoencoder is trained to extract features from the reactant and product simulation trajectories, the probability distributions across the difference between which is used to assess if the solvent

Processing of complex feedstocks for the production of value-added chemicals and fuels is industrially important. The lack of a priori knowledge of the innumerable species and the reaction pathways governing their conversion, has posed challenges to monitoring these processes. Although, data-driven

models have been used, their lack of interpretability and an end-to-end modeling framework has limited the efficiency of diagnostic decisions in process monitoring. On the other hand, systems where the mechanistic knowledge of the species and their reactions are arrived at from first-principles

Fall 2019

Kammammettu, Sanjula

uncertainty realized and presents a conservative solution to the problem that would be valid for any realization of uncertainty it was solved for. In contrast, stochastic optimization deals with uncertainty in an optimization problem by assuming that the probability distribution of the uncertainty is known

presents an additional layer of complexity owing to the presence of uncertainty in the operation of the system. This uncertainty may come from a variety of sources, such as effluent flow rate, contaminant concentration, and treatment unit removal efficiency. Therefore, the need to focus on developing a

The optimal design and operation of effluent treatment system networks poses a significantcant challenge in the present time, with the imposition of stricter environmental regulations and an increased demand for resources exacerbated by a diminishing resource pool. In practice, this problem