A cooperative game theory algorithm for distributed reactive power reserve optimization and voltage profile improvement

Modern power systems are continuously pushed to operate close to their limits due to economic and environmental constraints. As a result, the system voltage profile may seriously degrades, which can lead to voltage instability. The improvement of system voltage is thus of paramount importance for the reliability and stability of power grids. Several optimization techniques such as particle swarm optimization, artificial neural networks and non-cooperative game theory have been proposed to address the system voltage improvement and the related problem of reactive power maximization with varying levels of success. The framework of cooperative game theory is utilized in this work to propose novel models and algorithms for the joint optimization of reactive power reserve and voltage profile. We formulate a cooperative game with transferable utility to depict the interactions between system components during the operation of a power grid. First, a Shapley-value-based algorithm is developed to optimize system components locally (i.e., at the regional level). Second, we propose a cooperative game that adequately models the cooperation between independent regions within an interconnected power grid during power system operations. An algorithm is then developed to arrive at a global strategy that strives to optimize reactive power reserve and voltage profile at the regional level in a distributed manner. The proposed algorithms have been successfully validated using the IEEE-39 bus system and very encouraging results are reported.