Reactive power optimization of an active distribution network including a solid state transformer using a moth swarm algorithm

Abstract

This paper presents a reactive power optimization model for an active distribution network (ADN), including a solid state transformer (SST). In this model, the SST, as a power electronic device, compensates the system reactive power by adjusting the modulation coefficients and phase angles of the voltage source converter. A sequential multiscenario technique is implemented to simulate the transient characteristics of distributed generators (DGs). By optimizing the continuous variables of the SST and the reactive power output of a controllable DG, voltage deviation and power loss of the ADN are minimized. Meanwhile, although the cost of an SST is higher than that of a traditional transformer, using this model reduces the total cost of the system through reactive power optimization. In addition, a moth swarm algorithm is proposed. By adopting a position updating strategy of Lévy mutation, lateral positioning, and celestial navigation, the convergence speed and global optimization capability of the algorithm are improved. Finally, the rationality and validity of the proposed model and algorithm are verified by simulation of the IEEE 33-bus system.