Distributed predefined-time economic dispatch algorithm for microgrid

Abstract

This paper investigates the economic dispatch problem (EDP) concerning a cluster of distributed generators (DGs) within microgrid (MG) interconnected via a connected graph. The objective of each individual DG is to collaboratively minimize the total cost of generation while adhering to power balance constraints and individual generator capacity constraints. A novel distributed predefined-time algorithm is proposed to solve the EDP within a predefined time frame by utilizing distributed weighted gradient method, polynomial feedback control technique and $ϵ$-exact penalty function method. The minimum upper bound of the convergence time of the algorithm as an explicit parameter can be preset within any physically allowable range, which remains independent of the initial conditions of generators and easily adjustable. The power balance under the proposed algorithm is always satisfied. Rigorous proof of predefined-time convergence for the proposed algorithm is provided via convex optimization and Lyapunov theory. The effectiveness and superiority of the proposed algorithm are verified by IEEE 14-bus system simulation and comparison results.