Economic power sharing using droop strategy for parallel converters in a low-voltage AC island microgrid

Abstract

Distributed generation units participating in a microgrid may be employing different primary energy sources such as fossil fuels, renewable resources or battery systems. Given the diversity of such sources, it is natural that the operational costs will probaly vary between such units. When a microgrid is disconnected from the main electrical system, its distributed generators must be able to maintain adequate voltage and frequency levels to meet local loads. In addition, at this condition it is also necessary to coordinate the power dispatch of each generator based on their respective operational costs in order to minimize the overall islanded system operational cost. The microgrid local control in island operation can be accomplished by two methods: utilizing a central control system, which relies on communication technologies to coordinate the operation of local power sources, or by means of a distributed economic power-sharing management system, which performs sequential power dispatch of the distributed generators based on each unity operational cost, without a paralell communication system being required. The coordination of this strategy is based on the variation of grid parameters, such as voltage amplitude and frequency. This control method is known as Droop strategy and has the advantage of avoiding the use of secondary communication channels between converters, increasing system reliability, simplicity and speed response performance. In this context, this study consists of adapting a power sharing strategy based on economic aspects inherent to the primary energy source of the distributed generation units for a single-phase low-voltage microgrid, aiming system overall operational cost reduction. This strategy was implemented using finite-control-set model predictive control as local control method for the power-electronic converters, which added adequate ability to follow reference signals with dynamic adjustment, disturbance rejection and to deal with nonlinearities in the system, with performance comparable to traditional control techniques. The validation of the adapted economic strategy is performed in computer simulations using the MATLAB/Simulink software for a low-voltage microgrid with distinc loads and variable power demand over time.