Emended snake optimizer to solve multiobjective hybrid energy generation scheduling

Abstract

This paper proposes an emended snake optimizer (ESO) for solving hydrothermal, pumped hydro, and solar generators’ non-convex, highly constrained, and non-linear power generation scheduling problem. The generation scheduling problem aims to reduce thermal generator operating costs and pollutants by maximizing hydro volume and utilizing solar power generation. The minimization of operating costs and pollutants is subjected to various constraints, like meeting load demand, active power generation violations, water volume utilization, etc. The conflicting objectives of the multiobjective generation scheduling are handled using the non-interactive approach exploiting the price-penalty method. The direct heuristics search is utilized to satisfy the load demand and water volume constraints. The snake optimization algorithm (SOA) often gets stuck in the local minima while solving complex engineering optimization problems, resulting in sluggish convergence behavior. The basic SOA is emended through simple search and opposition-based learning, enhancing exploitation, convergence behavior, and procuring near to global solutions. The simulation studies involve solving unconstrained standard benchmark problems and electric power system problems. The proposed emended snake optimizer offers significant cost savings for electric power systems ranging from 10-15%. Statistical analysis using Wilcoxon signed-rank test and Friedman’s test justifies the amendment. The rapid convergence behavior and Whisker box plots justify the proposed ESO’s robustness.

Keywords

• coordinated generation scheduling
• renewable energy
• snake optimization algorithm
• simplex search method
• opposition-based learning
• metaheuristics optimization
• optimization problem