Hierarchical energy management system with multiple operation modes for hybrid DC microgrid

This paper presents a developed energy management system (EMS) with a hierarchical three-level distributed control approach proposed for a photovoltaic/wind turbine/diesel generator with energy storage in an islanded hybrid DC microgrid (MG). In the hierarchical three-level control structure, the power to be transferred by the distributed energy sources (DER) is provided by controlling the converters in the primary control layer. The secondary control layer is used to achieve DC bus voltage restoration and increase current-sharing accuracy simultaneously. In the tertiary control layer, it is aimed to minimize the production cost by performing optimum load-sharing. For the operation of this control structure, a strategy is proposed to optimally manage the energy demand of the hybrid system autonomously under variable load conditions. The EMS with three different operating modes (Mode-I, Mode-II, Mode-III) is proposed to share power between DC MG sources optimally. In Mode-I, renewable energy sources (RES) are enabled to operate with MPPT operation. Energy consumption minimization strategy (ECMS)-based linear programming (LP), genetic algorithm (GA) and pattern search (PS) algorithms are used to minimize the fuel consumption and carbon emissions of the diesel generator (DG) in Mode-II operating Mode. In Mode-III operating mode, DG is not needed, and the power flow between DERs is optimized by using the fuzzy logic controller (FLC) and state machine control strategy (SMCS). The amount of fuel consumed, the cost of fuel consumption and ESS's SOC level are used as the EMS's performance criteria for the MG. The minimum fuel consumption of the DG is calculated as 2.328 L for the operation of LP-based ECMS, and the fuel cost, in this case, is found to be $3.26. Considering the charging of the ESS, the highest SOC level is achieved with 64.87% for the PS-based ECMS-SMCS combination. By considering fuel minimization and SOC level together, it can be seen that the best result is achieved with the LP-based ECMS-SMCS combination.