Publications

Type of Publication: Article in Collected Edition

A Framework for Agent-Based Simulations of Hybrid Energy Infrastructures

Author(s):
Derksen, Christian; Branki, Cherif; Unland, Rainer
Editor:
Ganzha; Science, Maciaszek -; Computer, Federated Conference On
Title of Anthology:
Federated Conference on Computer Science 2012
pages:
1293-1299
Publication Date:
2012
ISBN:
978-83-60810-51-4
Language:
englisch
Link to complete version:
https://fedcsis.org/proceedings/2012/pliks/338.pdf
Citation:
Download BibTeX

Abstract

In many countries, future energy supply, management, and consumption change completely compared to today’s status. One reason is that sustainable energy production will rely on a large number of small and medium sized, de-centralized energy production units. The aim is to replace a small number of large power plants. The energy supply strategy change puts completely different demands on the energy grid as well as on the business models to run those energy grids efficiently and, especially, securely. The future energy grid will have to deal with an autonomous, communication-driven, highly diversified and dynamic, and continuously changing environment. Additionally, while today’s energy grids are run as independent systems the interaction between different energy networks, like natural gas or district heat, will strongly increase in the future. Inspired by ideas and techniques like Mini-CHP (combined heat and power) or Power-to-Gas, it is expected that future energy networks have to be seen as an entire hybrid energy network, where quantities of energy and energy forms will be dynamically exchanged and substituted as required. As a consequence, a lot of expertise that was gained in the past is no longer valid. On the other hand, little experience exists when it comes to shaping and managing the future energy grid. Due to the nature of multi-agent based simulation this class of simulation tools is well-suited for the simulation of future energy networks. The main advantages are the autonomy of agents, their highly developed interaction skills, the inherent distributed/decentralized approach of control in such a system and the high flexibility in the execution of tasks. This article presents Agent.GUI, which can be seen as a simulation toolkit and framework for complex and diversified energy systems. Based on the well-known JADE platform, the toolkit provides a wealth of specific features tailored to the needs for simulation of hybrid energy networks.