Tackling Variability in Renewable Energy Production and Electric Vehicle Consumption with Stochastic Optimization - The Benefits of Using the Stochastic Quasi-Gradient Method compared with Exact Methods and Machine Learning
dc.contributor.advisor | Korpås, Magnus | |
dc.contributor.advisor | Gaivoronski, Alexei | |
dc.contributor.advisor | Hu, Zechun | |
dc.contributor.author | Harbo, Sondre Flinstad | |
dc.date.accessioned | 2018-08-06T14:01:13Z | |
dc.date.available | 2018-08-06T14:01:13Z | |
dc.date.created | 2018-04-09 | |
dc.date.issued | 2018 | |
dc.identifier | ntnudaim:20052 | |
dc.identifier.uri | http://hdl.handle.net/11250/2507657 | |
dc.description.abstract | The work presented in thesis investigates different applications for implementing the Stochastic- Quasi Gradient (SQG) model to solve stochastic multistage AC-OPF problems, and com- pares it with a Stochastic-Dynamic Programming (SDP) approach and an Evolutionary algorithm. Where the SDP quickly becomes too cumbersome to solve, the thesis also shows the other two as more appropriate tools, where the SQG method works better in larger cases, the Evolutionary algorithm in smaller. Hence, to analyze how energy storage may optimally be used for incorporating variable renewable energy sources to bigger grid networks, the SQG method may be of academic and practical interest. | |
dc.language | eng | |
dc.publisher | NTNU | |
dc.subject | Energi og miljø, Energianalyse og planlegging | |
dc.title | Tackling Variability in Renewable Energy Production and Electric Vehicle Consumption with Stochastic Optimization - The Benefits of Using the Stochastic Quasi-Gradient Method compared with Exact Methods and Machine Learning | |
dc.type | Master thesis |
Tilhørende fil(er)
Denne innførselen finnes i følgende samling(er)
-
Institutt for elkraftteknikk [2401]