Using Graph Transformations in Distributed Adaptive Design System

Kotulski, Leszek; Strug, Barbara

doi:10.1007/978-3-642-02345-3_47

Cited by 4 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The graph representations satisfy those demands. They are expressive enough to represent a broad range of systems [17][18][19]. Moreover, they were shown to be a suitable environment for multi-agent systems' deployment [20,21].…”

Section: Graphsmentioning

confidence: 99%

Multi-Agent System Supporting Automated Large-Scale Photometric Computations

Sędziwy

Kotulski

2016

Entropy

Self Cite

View full text Add to dashboard Cite

Abstract:The technologies related to green energy, smart cities and similar areas being dynamically developed in recent years, face frequently problems of a computational nature rather than a technological one. The example is the ability of accurately predicting the weather conditions for PV farms or wind turbines. Another group of issues is related to the complexity of the computations required to obtain an optimal setup of a solution being designed. In this article, we present the case representing the latter group of problems, namely designing large-scale power-saving lighting installations. The term "large-scale" refers to an entire city area, containing tens of thousands of luminaires. Although a simple power reduction for a single street, giving limited savings, is relatively easy, it becomes infeasible for tasks covering thousands of luminaires described by precise coordinates (instead of simplified layouts). To overcome this critical issue, we propose introducing a formal representation of a computing problem and applying a multi-agent system to perform design-related computations in parallel. The important measure introduced in the article indicating optimization progress is entropy. It also allows for terminating optimization when the solution is satisfying. The article contains the results of real-life calculations being made with the help of the presented approach.

show abstract