Search

In recent years, fractional order proportional-integral-derivative (FOPID) controllers have been applied in different areas in the academy due to their superior performance over conventional proportional-integral-derivative (PID) controllers. When the literature is reviewed, it has been observed that lack of studies that use swarm-based and multi-objective optimization algorithms together with FOPID controllers in frequency control of micro-grid. The load frequency control (LFC) problem is considered as two objectives in order to eliminate the complications that occur when only the frequency deviation is minimized. In our study, a method called MOGOA-FOPID in which both the frequency deviation and the control signal are minimized together for the frequency control in the microgrid is proposed. By using the multi-objective grasshopper optimization algorithm (MOGOA), both the frequency deviation and the control signal are minimized together, and thus, it is aimed to limit the battery capacity, reduce the flywheel jerk and avoid high diesel fuel consumption as well as an effective frequency control. In order to obtain a more realistic system, not only the photovoltaic (PV) solar and wind power but also the load demand is taken in a stochastic structure. Then, the results of the proposed MOGOA-FOPID are compared with the results of multi-objective genetic algorithm (MOGA)-based PID/FOPID and MOGOA-PID and its superiority is demonstrated. Finally, robustness tests of the system are performed under the perturbed parameters and outperform of MOGOA-FOPID over other methods is seen. [ABSTRACT FROM AUTHOR]

In this study, energy demand of a faculty was aimed to supply with a hybrid energy system (HES) consisting of photovoltaic (PV) panels, wind turbine (WT) and bomass (BM) system with optimum power usage distribution and sized to reach a lowest cost and a reliable system. In this optimization, total net present cost (TNPC) for economic analysis, loss of power supply probability (LPSP) for reliability, and localized cost of energy (LCOE) for determining the unit energy cost were considered and an effective control algorithm was developed to decide the power source for improving system reliability. We used genetic algorithm (GA) and simulated annealing (SA), which are commonly used in the literature. On the other hand, we utilized the Grey Wolf Optimizer (GWO), which was recently found out and inspired by the hierarchy and hunting instincts of grey wolves. The results of GWO algorithm were also compared with GA and SA and confirmed that GWO is satisfying. GWO achieved better results to solve problems by setting LPSP to both 0.02 and 0.01 upper limits. When LPSP set to 0.02 maximum point, GWO suggested PV system at 86.39 kW power and BG at 50 kW power. Consequently, the energy requirement of a faculty was supplied by an optimized and designed PV/WT/BM HES. In addition, by the installation of optimized system, 144.29 tons of CO2 emissions per year will be reduced. [ABSTRACT FROM AUTHOR]