A microgrid is an integration of distributed energy sources, loads and energy storage systems. Indeed, energy storage systems are required in order to ensure reliability and power quality because of the intermittent nature of renewable energy sources and changes of load demand. Apart from that, the use of distributed energy storage units provides redundancy to the system and support possible increments in load consumption. In consequence, the control strategy used in the microgrid must take into account the stored energy balance between distributed energy storage units in order to avoid over-charge or deep-discharge in one of the energy storage units. Primary control in a microgrid is responsible for power sharing among units; and droop control is typically used in this stage. This paper proposes a modular and decentralized gain-scheduling control strategy based on fuzzy logic that ensures balanced stored energy among distributed energy storage units, as well as low voltage deviation in a DC microgrid. Hardware in the loop simulations show the performance of the proposed control strategy.