Initialization-free Privacy-guaranteed Distributed Algorithm for Economic Dispatch Problem

TL;DR

This paper introduces a simple, initialization-free distributed algorithm for economic dispatch in power networks that guarantees privacy, adapts to network changes, and detects infeasibility, with proven robustness in simulations.

Contribution

It presents a novel, privacy-preserving, initialization-free dual gradient-based distributed algorithm for economic dispatch that handles network changes and detects infeasibility.

Findings

Algorithm is robust in IEEE 118 bus system simulations.

No private information is disclosed during optimization.

The method adapts to online changes in network and demand.

Abstract

This paper considers the economic dispatch problem for a network of power generators and customers. In particular, our aim is to minimize the total generation cost under the power supply-demand balance and the individual generation capacity constraints. This problem is solved in a distributed manner, i.e., a dual gradient-based continuous-time distributed algorithm is proposed in which only a single dual variable is communicated with the neighbors and no private information of the node is disclosed. The proposed algorithm is simple and no specific initialization is necessary, and this in turn allows on-line change of network structure, demand, generation constraints, and even the participating nodes. The algorithm also exhibits a special behavior when the problem becomes infeasible so that each node can detect over-demand or under-demand situation of the power network. Simulation results on IEEE 118 bus system confirm robustness against variations in power grids.