Programska knjižnica za stvaranje znanja zasnovanog na eksperimentima unutar platforme za simuliranje trgovanja električnom energijom

undergraduate thesis

• Title (english): A program library for experiment-driven knowledge creation within a power trading simulation platform
• Author: Josip Jozić
• Mentor: Vedran Podobnik (mentor)
• Committee member: Vedran Podobnik (predsjednik povjerenstva)
• Committee member: Jurica Babić (član povjerenstva)
• Committee member: Darko Štriga (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Kao dobar pristup za modeliranje liberaliziranog tržišta pokazao se model zasnovan na programskim agentima u kombinaciji sa simulacijom. Jedna takva simulacija pod nazivom Power TAC obrađena je u ovom završnom radu. Glavni akter je programski agent koji simulira brokera na tržištu električne energije. On kupuje električnu energiju od velikih proizvođača na veleprodajnom tržištu, te pomoću odgovarajućih tarifa pojedinim potrošačima prodaje na maloprodajnom tržištu. Cilj je u natjecanju s drugim agentima ostvariti najveći profit, što je ujedno i kriterij pobjede.
Jedan od dosadašnjih nedostataka kostura programskog agenta bio je nemogućnost učenja kroz prethodne simulacije. Naime, po završetku jednog eksperimenta sve informacije o njemu bile bi zaboravljene, odnosno nisu se pohranjivale. Zadatak programske knjižnice koja je razvijana u sklopu ovog Završnog rada bio je omogućiti pohranu ovih podataka te pristup istim u nekim budućim s ciljem izrade programskog agenta koji je u stanju stvarati i koristiti znanje o prošlim simulacijama.

Abstract (english)

The agent-based model in combination with an simulation has proved to be good approach to modeling liberalized market. One of those simulations named Power TAC has been examined in this thesis. The main actor is an agent who acts as a broker on the electrical energy market. The agent is buying the energy from generating companies on the wholesale market, and through appropriate tariffs he is selling it to customers on the retail market. The main goal of the competition is to achieve the highest profit among other agents, what is at same time the criteria for win.
At the moment the structure of the agent had some shortcomings, one of them was the inability of learning through previous simulations. At the finish of an experiment all information regarding the experiment are lost, or better said they do not save. The task of the program library that was in this thesis discussed was to enable storage of this data and access to the data in the case of some future simulations with the end goal of constructing a more competitive agent.

• Keywords: platforma za simuliranje trgovanja električnom energijom; Power TAC; pohrana znanja; pristup znanju
• Keywords (english): power trading simulation platform; Power TAC knowledge storage; knowledge access
• Language: croatian
• URN:NBN: urn:nbn:hr:168:768889
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 09:28:07