Solving a class of resource allocation problem under dynamic constraints: A predefined-time distributed optimization scheme.

Abstract

In this paper, a predefined-time distributed optimization algorithm is designed to solve the resource allocation problem (RAP) with dynamic constraints. This algorithm updates auxiliary variables in real time through a distributed approach and allocates resources to each node based on dynamic constraints. Its advantages include ensuring all nodes quickly converge to the optimal value within a predefined time, thereby enhancing algorithm efficiency. Moreover, the auxiliary variables exchanged between nodes do not contain any real physical information, effectively preventing privacy data leakage. In addition, the convergence of the algorithm is analyzed strictly by the Lyapunov method, which ensures the accuracy of the algorithm. Finally, application examples in smart grids and multi-UAV dynamic collaboration are provided to demonstrate the effectiveness and advantages of the algorithm in different application scenarios.