Modelagem de um sistema multiagente para aplicação em ambientes industriais.

Abstract

In this work a proposal of a decentralized strategy to aid industrial systems operators in the decision making process after the occurrence of contingencies in the system is detailed. This strategy is based on the multiagent approach. Those agents are responsible for evaluating the state of the system and perform a set of corrective actions based on a set of rules and negotiations between themselves. The modeling process of the multiagent system that implements the strategy is presented in the text. The application context chosen to exemplify the applicability of the strategy was that of an electrical power system, more specifically the subtransmission part. The work is related to the investigation of a decentralized strategy adequacy to the supervision of industrial processes, where traditionally it is employed a centralized approach. As result of this work it is proposed the architecture of a multiagent-based computational tool that supports the strategy's application in industrial environments.