Multi-agent approach to power distribution network modelling

Power supply reliability has become a question of ever growing importance. Due to its geographical distribution and the high costs of maintaining electric power systems (EPS), the necessary quality cannot be achieved simply by redundancy. It is necessary to search for new and inexpensive yet effective means of ensuring the required reliability. Emergent systems and intelligent agents are in the focus of great attention in many areas of computer science and artificial intelligence. Intelligent autonomous systems are seen as systems that can react intelligently and flexibly upon changes in operating conditions and demands from the surrounding processes. The Rice multi-agent system, which is presented in this paper, is designed as a distributed environment with elements that use communication standards for local interactions. The task of maintenance, control and simulation of EPS is handled by modelling the network with agents that resemble the behaviour of real EPS components. The overall design and main architecture goals of Rice are described here and interesting Rice simulation use-cases and acquired outputs are demonstrated.