A class of distributed optimization methods with event-triggered communication

Authors:
Martin Meinel;Michael Ulbrich;Sebastian Albrecht
Affiliations:
Lehrstuhl für Mathematische Optimierung, Fakultät für Mathematik, Technische Universität München, Garching b. München, Germany 85748;Lehrstuhl für Mathematische Optimierung, Fakultät für Mathematik, Technische Universität München, Garching b. München, Germany 85748;Lehrstuhl für Mathematische Optimierung, Fakultät für Mathematik, Technische Universität München, Garching b. München, Germany 85748
Venue:
Computational Optimization and Applications
Year:
2014

Citing 6
Cited 0

Parallel and distributed computation: numerical methods

Parallel and distributed computation: numerical methods
Smooth minimization of non-smooth functions

Mathematical Programming: Series A and B
On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming

Mathematical Programming: Series A and B
Distributed network utility maximization using event-triggered augmented Lagrangian methods

ACC'09 Proceedings of the 2009 conference on American Control Conference
Decision making as optimization in multi-robot teams

ICDCIT'12 Proceedings of the 8th international conference on Distributed Computing and Internet Technology
A tutorial on decomposition methods for network utility maximization

IEEE Journal on Selected Areas in Communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a class of methods for distributed optimization with event-triggered communication. To this end, we extend Nesterov's first order scheme to use event-triggered communication in a networked environment. We then apply this approach to generalize the proximal center algorithm (PCA) for separable convex programs by Necoara and Suykens. Our method uses dual decomposition and applies the developed event-triggered version of Nesterov's scheme to update the dual multipliers. The approach is shown to be well suited for solving the active optimal power flow (DC-OPF) problem in parallel with event-triggered and local communication. Numerical results for the IEEE 57 bus and IEEE 118 bus test cases confirm that approximate solutions can be obtained with significantly less communication while satisfying the same accuracy estimates as solutions computed without event-triggered communication.