Agent based State Estimation in Smart Distribution Grid

A novel agent based static state estimation strategy for a specific class of physical systems is proposed. Physical systems, (e. g. smart distribution grid) which are well modeled using decentralized measurements and distributed state-space formulations are considered. In such systems, sensor nodes acting as agents estimate only a subset of states, instead of evaluating local estimates of global states. In general, for each agent, the measurement model reduces to an underdetermined nonlinear system and in many cases, the state elements associated with an agent may overlap with neighboring agents. A classic example of such a physical system is a radial power distribution grid. We propose a state estimation strategy, which effectively integrates the principles of local consensus and least squares solution and illustrate its potency using the power distribution grid. We also present rigorous analysis of convergence of the proposed approach to motivate its application to other multi-agent systems.