Primjena strojnog učenja u vođenju elektroenergetskog sustava

master's thesis

• Title (english): An application of machine learning in power system control
• Author: Slavko Boldin
• Mentor: Ivica Pavić (mentor)
• Committee member: Ivica Pavić (predsjednik povjerenstva)
• Committee member: Juraj Havelka (član povjerenstva)
• Committee member: Goran Grdenić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu je predstavljen Grid2Op, radni okvir koji omogućava modeliranje
vođenja i kontrole elektroenergetske mreže na temelju podržanog učenja. U prvom
dijelu opisane su osnove strojnog, odnosno podržanog učenja te kratka povijest
strojnog učenja u energetici. Potom je opisan problem kontrole elektroenergetske
mreže na temelju modela mreže i topologije (broj sabirnica i spajanje elemenata u
trafostanici). Predstavljene su glavne odrednice i moduli Grid2Op radnog okvira,
dan je popis članskih varijabli u kojima su sadržani podaci o odabranom tipu
elektroenergetske mreže. U dijelu implementacije, opisana je usporedba
performansi dva odabrana agenta u djelovanju u mreži te su u zaključku izneseni
njihovi rezultati i mogućnost poboljšanja.

Abstract (english)

Theme of this paper is an application of machine learning in power system control
in Grid2Op framework. In the first part, basics of machine and reinforcement
learning and brief history of machine learning in Energetics is described. After that,
description of power system control problem based on network model and
topology is given. Then, main characteristics and modules of Grid2Op framework
are showcased, as well as attributes which contain data on network model type. In
the implementation part, comparison of two chosen agents and their performance
in algorithm of power network control is given. Conclusion contains results and
remarks on possible further adjustments.

• Keywords: strojno učenje; podržano učenje; elektroenergetska mreža; agent; Grid2Op
• Keywords (english): machine learning; reinforcement learning; power network; agent; Grid2OP
• Language: croatian
• URN:NBN: urn:nbn:hr:168:248298
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 14:35:46