International audience

International audience

In this paper, we present a robust nonlinear de-centralized control scheme for an islanded DC microgrid (MG) where the main control goals are to achieve a sustained stability for the DC bus voltage at a certain desired value, to maintain the power balance in the system and to insure robustness against disturbances and perturbations. The proposed control strategy uses the Model Free Super-Twisting Sliding Mode (MFSMC) as it needs no accurate representation of the environment in order to be effective which makes it a suitable choice for system with nonlinear model prone to parameters variation. The studied microgrid is composed of a solar photo-voltaic (PV) unit and a hybrid energy storage system including a battery and a supercapacitor (SC) along with DC loads. To attain the intended objectives, a hierarchical cascaded control strategy is designed with two levels: a high-level control that stabilizes the DC bus voltage at a reference value by generating a current reference to be tracked, and a low-level control composed of an energy management system EMS based on a passive filtration to distribute the reference current between the storage system units according to their dynamic specifications. Simulations on MATLAB/Simulink are carried out to evaluate the effectiveness and robustness of the proposed control scheme under various operating conditions created by random variations of power generation and consumption. Noise sensitivity test is also carried out for this controller and for a model-based one that is the Feedback linearization control technique (FL).

This paper presents a new Energy Management System (EMS) strategy for the Energy Storage System (ESS) of a Microgrid (MG) based on the proper time scale of each ESS component. The theoretical framework allows to achieve more efficient use of the different ESS components’. While almost all existing methods rely on filtering load profile signals to achieve power separation, our proposed approach achieves a better performance on components’ ageing rate with less processing complexity using a simple limiter in algorithm. The proposed method is first evaluated as an EMS only and then extended to the whole MG. Results show an ageing rate reduction of 93% for fast component at the price of 24% of ageing rate increase for slow component.