CrocodileAgent 2018: Robust Agent-based Mechanisms for Power Trading in Competitive Environments

Demijan Grgic1, Hrvoje Vdovic1, Jurica Babic1 and Vedran Podobnik1

  1. University of Zagreb, Faculty of Electrical Engineering and Computing
    Unska 3, 10000 Zagreb, Croatia
    {demijan.grgic, hrvoje.vdovic, jurica.babic, vedran.podobnik}


Besides the smart grid, future sustainable energy systems will have to employ a smart market approach where consumers are able choose one of many different energy providers. The Power Trading Agent Competition (Power TAC) provides an open source, smart grid simulation platform where brokers compete in power brokerage. This paper presents CrocodileAgent, which competed in the Power TAC 2018 finals as a broker agent. The main focus in the design and development of CrocodileAgent 2018 was the creation of smart time-of-use tariffs to reduce peak-demand charges. CrocodileAgent 2018 was ranked third in Power TAC 2018 Finals, with a positive final profit and a positive result in each of three game types. In addition, CrocodileAgent 2018 had the highest percentage of “profitable games” (91%) from among all competing agents, the second highest level of “net profit per standard deviation” (0.48) and the third highest “net profit per subscriber” (79 monetary units).

Key words

computer simulation, agent-based modelling, electricity trading, tariff design, Power TAC, CrocodileAgent

Digital Object Identifier (DOI)

Publication information

Volume 16, Issue 1 (January 2019)
Year of Publication: 2019
ISSN: 1820-0214 (Print) 2406-1018 (Online)
Publisher: ComSIS Consortium

Full text

DownloadAvailable in PDF
Portable Document Format

How to cite

Grgic, D., Vdovic, H., Babic, J., Podobnik, V.: CrocodileAgent 2018: Robust Agent-based Mechanisms for Power Trading in Competitive Environments. Computer Science and Information Systems, Vol. 16, No. 1, 105-129. (2019),