Blar i NTNU Open på forfatter "Grimstad, Bjarne"
-
A spatial branch-and-bound method for ReLU network-constrained problems
Masdal, Even (Master thesis, 2020)Motivert av evnen kunstige nevrale nettverk har til data-dreven black-box modellering og bruken av spatial branch-and-bound (sBB) algoritmer til å løse ulineære optimaliseringsproblemer presenterer denne opgaven en metode ... -
A Study in MINLP-class Optimization Problems for Simulated Petroleum Production
Ausen, Håvard (Master thesis, 2012)To aid in faster and better decision making it is interesting to couple advanced simulators with optimization tools.Most simulators however does not offer gradients, therefore derivative-free methods must be used. In this ... -
Ablating a Graph Neural Network for Branching in Mixed-Integer Linear Programming
Sandberg, Lars Lødemel (Master thesis, 2021)Denne oppgaven evaluerer ablasjoner av et grafkonvolusjonelt nevralt nettverk for maskinlæringsassistert forgrening foreslått av Gasse et al. (2019) for mer effektiv løsning av blandede heltallsproblemer (MILP). Effektive ... -
An exploration of sequence models using multi-task learning for multiphase flow rate estimation in oil and gas wells
Heggland, Morgan Feet; Kjærran, Patrik (Master thesis, 2022)Å estimere mengden flerfasestrøm som strømmer gjennom olje- og gassbrønner er avgjørende for å kunne ta informerte beslutninger angående operasjonelle aktiviteter relatert til offshore olje- og gassproduksjon, som for ... -
Application of Bayesian neural networks for petroleum optimization
Baugstø, Sondre Wangenstein (Master thesis, 2019)Denne masteroppgaven undersøker bruken av Bayesianske nevrale nett for prediktiv modellering i oppstrøms olje- og gassproduksjon. Hypotesen er at Bayesianske nevrale nett kan brukes til å lage prediktive modeller som har ... -
Automatic Detection of Poorly Calibrated Models in State Estimation Applied to Oil and Gas Production Systems
Skibeli, Håkon (Master thesis, 2015)In modern oil and gas industry, there is an increasing use of instrumentation. This lead to a huge flow of information, which typically is not utilized to its full potential. By the use of increasingly more complex Virtual ... -
Daily Production Optimization for Subsea Production Systems: Methods based on mathematical programming and surrogate modelling
Grimstad, Bjarne (Doctoral thesis at NTNU;2015:275, Doctoral thesis, 2015) -
Data driven analysis in oil and gas operations - Datadrevne analysemetoder i olje- og gassproduksjon
Nordmo, Michael Helland (Master thesis, 2016)Knowledge about the production system and models of relevant parts of the production network can improve the decision-making process in offshore oil and gas production. This thesis investigates how multivariate projection ... -
Data Driven Real-Time Petroleum Production Planning Using Optimization and Neural Networks
Malvik, Arnt Gunnar; Witzøe, Bendik (Master thesis, 2018)Real-time optimization (RTO) of petroleum production is concerned with maximizing daily oil production by suggesting minor adjustments to the system and frequently re-solving the optimization model. A central part of the ... -
Dynamic Estimation for Controlling a Subsea Production System - Virtual Flow Metering using B-spline Surrogate Models
Robertson, Patrick Michael (Master thesis, 2014)Knowledge of the flow rates from individual wells in a subsea production system can greatly improve decision-making processes. This thesis investigates flow estimation in the subsea template Tilje in the BP-operated Skarv ... -
Mixed-Integer Nonlinear Programming Heuristics Applied to a Shale Gas Production Optimization Problem
Sharma, Shaurya (Master thesis, 2013)Mixed-Integer nonlinear programs(MINLPs) are a general class of nonlinear optimization problems that have a wide array of real-world applications. These problems are in general notoriously difficult to solve, and it is ... -
MLOps - challenges with operationalizing machine learning systems
Kjetså, Tor Istvan Stadler (Master thesis, 2021)Det er en økende etterspørsel etter maskinlæringsapplikasjoner innenfor flere industries. Det finnes gode maskinlæringsmodeller, men det er utbredte vanskeligheter i sammenheng med å operasjonalisere dem. Mangel på verktøy ... -
On a hybrid approach to model learning applied to virtual flow metering
Hotvedt, Mathilde (Doctoral theses at NTNU;2022:165, Doctoral thesis, 2022)Process modeling using first-principle equations has existed for centuries as a methodology to represent and analyze real-world processes. In time with increasing computing power and sensor data availability, data-driven ... -
Predictive modeling with applications in decision support systems for oil and gas production
Cenar, Ugur Alpay (Master thesis, 2017)In the past decade, artifical neural networks (ANN) has given us self-driving cars, practical speech recognition and more effective web search. It is in interest of petroleum industry to research the applicability of ANNs ... -
Real-Time Data-Driven and Hybrid Modeling of Two-Phase Flow in Oil and Gas Wells - A study on how data quality affects model accuracy and the application of neural networks for flow estimation
Sjulstad, Christine Foss; Almås, Ingeborg Victoria Aarsvold (Master thesis, 2020)Denne avhandlingen forsøker å estimere to-fase-flyt for olje- og gassbrønner i sanntid ved bruk av den nyeste teknologien innenfor datadrevne og hybride modeller. Siden fokuset er på enkeltbrønner, modellerer vi brønnspesifikke ... -
Stochastic Gradient Optimization of Petroleum Assets: Towards Reinforcement Learning
Grepperud, Jakob Eide (Master thesis, 2022)Et petroleumsproduksjonssystem byr på flere unike utfordringer som gjør sanntidsoptimering vanskelig. Modellutvikling basert på fysiske lover er komplisert og kostbart på grunn av svært kompleks dynamikk og mangel på ...