Investigating Advanced Metaheuristic Algorithms for Economic-Emission Load Dispatch of a Diesel-Solar-Wind Microgrid

The microgrid system can be identified as a localised power generation infrastructure designed to serve a small region for continuous and reliable power supply. The growing emphasis on environmental sustainability has led to the optimal sizing of all connected renewable and conventional microgrid units such that the power is allocated to non-emitting renewable energy sources primarily. Economic Load Dispatch (ELD) involves determining the optimal sizing of generation units, with the objective of minimizing generation costs. The inclusion of the emission minimization objective into the ELD problem transforms it into the bi-objective EconomicEmission Load Dispatch problem. Metaheuristic approaches are found to be the most efficient to solve the EELD problem. Consideration of realistic aspects such as VPL, transmission losses and hourly dynamic load conditions makes EELD a challenging optimization problem. This article explores advanced metaheuristic methods to address EELD problem and proposes the application of the African Vulture Optimization Algorithm (AVOA) to solve EELD problem of the diesel-wind-solar microgrid. AVOA is then used to solve EELD of an islanded diesel-wind-solar microgrid modelled on field data of a location in Jaisalmer, India. The African Vulture Optimization Algorithm (AVOA) mimics the vulture’s food-seeking behaviour and navigation phenomena with efficient exploration and exploitation capabilities. AVOA’s ability to solve EELD problem is initially confirmed on three different-sized test systems of 10, 6 (IEEE30-bus), and 40 units prior to applying it to the microgrid case. To establish the efficacy of AVOA, its performance is compared with other established and advanced methods like SSA, GA, GWO, WOA, MFO, SAR, QTLBO, MOEA/D etc. Apart from the performance criteria like cost, emission, convergence etc; the feasibility of the approaches is further assessed by a novel index named as Viability Score formulated using three main selected criteria. Results of statistical data tests like ANOVA, Wilcoxon and robustness establish the statistical reliability of AVOA solving EELD in microgrid. Further, a multi-comparison post-hoc TukeyHSD test is introduced for statistical comparison of AVOA with other methods more explicitly. Results find AVOA most efficient method with 5.25% and 33.09% reduction in cost and emission respectively as compared to its next competitive method in solving EELD on microgrid (all sources), which is remarkable.