Caltech Logo

Rigorous Systems Research Group (RSRG) Seminar

Thursday, May 10, 2018
12:00pm to 1:00pm
Add to Cal
Annenberg 213
A linearly convergent primal-dual gradient algorithm for optimal distributed feedback voltage control
GUANNAN QU, Graduate Student, School of Engineering and Applied Sciences, Harvard Uninversity,

This talk will be divided into two self-contained parts: a distributed voltage control method and the primal-dual gradient algorithm underlying it. In the first part of the talk, we will present a distributed voltage control method in power distribution networks through reactive power compensation. The proposed control can (i) operate in a distributed fashion where each bus makes its decision based on local voltage measurements and communication with neighbors, (ii) always satisfy the reactive power capacity constraint, (iii) drive the voltage magnitude into an acceptable range, and (iv) minimize an operational cost. In the second part of the talk, we focus on the optimization technique underlying the voltage control. We propose a variant of the primal-dual gradient algorithm (dynamics) that has a global linear convergence (exponentially stability) guarantee under some regularity conditions. We will present the proof of the linear convergence (exponentially stability), which relies on a specially constructed quadratic Lyapunov function. Based on joint work with Na Li.

For more information, please contact Yanan Sui by email at [email protected] or visit Seminars & Events.