A multi-agent petri net model power management strategy for wind–solar-battery driven DC microgrid

Abstract

In recent years, autonomous direct current microgrid has been widely investigated to improve its performance in terms of integrating distributed energy resources and power balance issues. In this paper, a multi-agent hybrid petri net model is developed to ensure power management in wind–solar-battery driven low-voltage direct current microgrid. The multi-agent system is a discrete-event dynamic process in which petri nets are employed as the modelling tools. Direct current microgrid is distinguished with a simple control and power management with respect to alternating current microgrid and poses several advantages such as higher conversion as well as transmission efficiency, higher reliability even in remote locations, convenient control, reduced cost and lesser filter effort due to absence of reactive power, phase synchronization, high inrush current etc. To evaluate the behaviour of the direct current microgrid with several distributed task, a multi-agent hybrid petri net adapting two-layer control strategy has been considered. The first layer deals with modelling and development of hybrid petri net for collecting all the possible information by coordinated switching control in between generating units. This information is communicated to the local control structure to select the mode intelligently. Finally, this coordinated control approach is experimentally verified in the direct current microgrid test bench. The test results validate effective mode transition and power sharing among all distributed energy resources and load.