Modified Social Group Optimization—a meta-heuristic algorithm to solve short-term hydrothermal scheduling

Abstract

Social Group Optimization (SGO), developed by Satapathy et al. in the year 2016, is a class of meta-heuristic optimization inspired by social behavior. It has two phases: improving phase and acquiring phase. In the improving phase, each individual improves its knowledge by interacting with the best person/solution and in acquiring phase, the individuals interact with randomly selected individuals and the best person simultaneously to acquire knowledge. Modified Social Group Optimization (MSGO) is the improved version of SGO, where the acquiring phase is modified. A self-awareness probability factor is added in the acquiring phase, which enhances the learning capability of an individual from the best-learned person in the societal setup. It is observed that this modification has improved both exploration and exploitation abilities in comparison with the conventional SGO. To analyze the performance of the MSGO, an exhaustive performance comparison is made with GA, PSO, DE, ABC, and a few newer algorithms of the years 2010–2019. The results are tabulated in six experiments. Later, MSGO is applied to solve the short-term hydrothermal scheduling (HTS) problem. The central objective of the HTS problem is to ascertain the optimal plan of action for hydro and thermal generation minimizing the fuel cost of thermal plants and, at the same time satisfying various operational and physical constraints. The valve point loading effect related to the thermal power plants, transmission loss, and other constraints lead HTS as a complex non-linear, non-convex, and non-smooth optimization problem. Simulation results clearly show that the MSGO method is capable of obtaining a better solution.