This paper examines the application of adaptive estimation and control techniques to reduce the amount of communication required between subsystems in a distributed control implementation. Rather than require a large amount of communications to broadcast the outputs or the states of each of the subsystem nodes to all of the other nodes at every sampling instant, local estimators in each subsystem are used to predict the state vectors for all of the other subsystems. These estimates are then used in the calculation of the controller outputs for each of the subsystems. Prior work in the literature has focused on static estimation schemes to achieve such reductions in communications. However, such schemes typically require very accurate models of the plant in order to maintain the desired reduction in communications. Poorly modeled dynamics or systems whose dynamics change slowly over time (due to aging of components, changes in plant parameters such as a robot picking up a heavy object, etc.) can cause a substantial increase in the amount of communications required to maintain the desired system performance. In order to avoid this, this paper presents an adaptive estimation and control scheme for each subsystem in the distributed implementation. The stability of the state estimators and the convergence of the state tracking errors to within a desired threshold is proven. The performance of the system using perfect communication at every sampling instant, using a static estimation scheme, and using the proposed adaptive estimation scheme are then compared in simulation.
Library of Congress Subject Headings
Electric power systems--State estimation; Electric power systems--Control; Electric power systems--Communication systems; Electric power system stability
Department, Program, or Center
Electrical Engineering (KGCOE)
Burry, Aaron, "An Adaptive estimation scheme for reducing communications in a distributed control implementation" (2004). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus