Hybrid intelligent multiagent model of heterogeneous thinking for solving the problem of restoring the distribution power grid after failures

Abstract

Problems arising in such dynamic environments as regional power grid have the following features: partial observability, high dimensionality of the state space, interconnection and dependence of solutions on each other, which do not allow correcting the erroneous solution in the future. Abstract-mathematical models are limited and irrelevant to such dynamic environments. For this reason teams of experts of various specialties or their computer models are involved, but even they do not always successfully solve emerging problems. The success of the team depends largely on the ability of the decision maker to organize the process of heterogeneous collective thinking, the diagnosis of collective effects, the problems of group behavior and to choose corresponding model of collective reasoning. Under the conditions of time constraints in practice, it is not possible to organize such a comprehensive collective problem solving process. In this regard, the paper proposes the formalized model and the basic algorithms of a new class of intelligent systems, namely hybrid intelligent systems of heterogeneous thinking. Their main feature is modeling of the discussion management in the expert team by the facilitator with heterogeneous collective thinking methods. These methods will allow the agent modeling facilitator’s actions to organize communication and diagnostics of collective effects, problems and adjustment of group behavior, ensuring the relevance of the system to conditions of dynamic directly unobservable environments.