Distributed Optimization for Economic Dispatch Over Time-Varying Directed Networks With Row-Stochastic Matrices

Abstract

In recent years, with the continuous development of power grid, the power system is also transitioning to a more intelligent power grid. However, the economic dispatching problem of power system needs effective coordination in order to achieve greater benefits. In this paper, a distributed projection gradient algorithm based on row-stochastic over time-varying directed graph is proposed to solve the economic dispatch problem of power system. The algorithm uses row-stochastic condition instead of double-stochastic condition and column-stochastic condition, so that each intelligent power system can independently assign the weight value of neighboring nodes according to the situation, which improves the autonomous controllability of the system. The convergence of the algorithm is analyzed theoretically, and it is proved that the algorithm can match the optimal solution. Finally, the validity and accuracy of the algorithm are verified by numerical experiments.