Online Optimization of Capacitor Switching in Electric Power Distribution Systems

This thesis is focused on the development of a new online framework for control and optimization of capacitor switching in automated electric power distribution systems. The presented methodology is intended to address distribution system operation needs which are emerging in response to increased adoption of distributed generation and customer load management. The work is intended to assist in the transition of distribution system operation from pre-determined local control schemes to coordinated online control and optimization.;Time windowing concepts, commonly employed in model predictive control (MPC), are adapted to transform the problem of steady-state optimization of electric power distribution systems into a quasi-steady-state approach. An algorithm is presented for identifying time windows over which online decisions should be computed and implemented, based on forecasted system load and source injection values. The identified time windows are embedded directly within an online capacitor control problem formulation.;Simulation results are presented which demonstrate an implementation of the methodology applied to a 394 bus distribution system. The results demonstrate the feasibility of computing and implementing online capacitor control decisions, subject to both practical computational requirements and capacitor switching limits. The approach provides an online control methodology for addressing emerging system conditions and can further support advanced distribution automation.