A Two-Stage Optimization Framework for Power Systems Containing Virtual Power Plants with Distributed Generators and EV Charging Demand

Abstract

Abstract Smart Cities bring people the benefits of advanced technologies such as the Distributed Generators (DGs) by using renewable energy sources and the emission-free Electric Vehicles (EVs). However, the increasing integrations of DGs and EVs in electricity distribution networks bring challenges to the centralized power system operators, because of the stochastic generation and demand levels and the incomplete information at the operators’ side. One potential way to effectively accommodate the large number of DGs and EVs is to use the concept of Virtual Power Plants (VPPs). This paper presents a two-stage optimization framework for systems with VPPs containing various types of entities including the DGs and EV charging demand. At the first stage, the system operator optimizes the total output levels of each VPP by using the aggregated parameters provided by the VPP aggregator; at the second stage, each VPP aggregator distributes the total output level among the local entities. Optimization of the two stages in the above framework are both formulated as linear programming and are solved by using existing software tools. Numerical examples are presented to demonstrate that the proposed two-stage VPP model leads to better results than the traditional centralized power system dispatch method.