An optimal master-slave model for stochastic planning of AC-DC hybrid distribution systems

Abstract

In this study, a novel stochastic planning method is proposed for AC-DC hybrid distribution networks. The proposed approach is based on the graph theory, and the optimal AC-DC structure of the network is selected among the system spanning trees. The presented method is a Mixed Integer Nonlinear Programming (MINLP) problem, which is solved using genetic algorithm. The buses and lines of the network can be either AC or DC to minimize the system investment costs in the master optimization problem. The location and capacity of the Distributed Energy Resources (DERs) as well as the site and size of the Electric Vehicle (EV) charging stations are optimized in the slave problem to minimize the network losses and system costs. The proposed model utilizes Monte Carlo simulation to deal with the stochastic variations of the renewable energy resources power and load demands. Besides, the converter efficiency curve in the proposed planning problem is modeled based on a function of its input current using PLECS software. The proposed approach for network design can be applied to different DG resources and AC-DC loads. The comparison between the simulation results of the proposed approach and the conventional AC planning method demonstrates the efficiency of the proposed model in reducing network losses and system costs

Keywords

• ac-dc distribution systems
• stochastic planning
• graph theory
• converter efficiency curve
• distributed energy resources
• electric vehicle charging stations