Hybrid NSGA-III and Genetic Programming (GP) Integrated with Game Theory for Energy Trading and System Optimization
Minu Mary JoyResearch Scholar, Department of Electrical and Electronics Engineering, Noorul Islam Centre for Higher Education, Kumarakoil, Tamil Nadu, India. minujoy3@gmail.com0009-0003-5403-1064
Dr.H. VennilaAssociate Professor, Department of Electrical and Electronics Engineering, Noorul Islam Centre for Higher Education, Kumarakoil, Tamil Nadu, India. vennilarajesh@yahoo.co.in0000-0001-6516-3124
Keywords: Hybrid Renewable Energy Systems (HRES), Multi-objective Optimization, Hybrid NSGA-III, Genetic Programming (GP), Game Theory, Energy Trading, Energy Storage, Market Dynamics, System Optimization, Sustainability.
Abstract
Hybrid Renewable Energy Systems (HRES) have become an important pathway for developing sustainable means of electricity generation using renewable energy technologies and storage. This occurs while responding to multiple and conflicting objectives, including cost minimizing, reliability maximization, minimizing losses and meeting environmental sustainability. The complexity of incorporating a variety of HRES constraints is compounded through also needing to participate in energy trading between micro grids, HRES units and utility grids, that are reliant on unpredictable markets and demand-supply uncertainty. To study how to deal with progressed modelling uncertainty, to propose a new method for the optimization of sustainability frameworks, through Hybrid NSGA-III, Generation Programming, and Game Theory. Hybrid NSGA-III ensures the multi-objective optimization of system sizing and operations, Generation Programming allows for evolving simulation of the uncertainty in the operating system solutions, and game theory approaches provides fair and efficient energy trading. A full statistical comparison with benchmarks (NSGA-II and MOEA/D) has shown that the suggested framework is the best. Specifically, the proposed framework showed a 11% ± 2.3 cost reduction and 9% ± 1.8 emissions reduction in 2 for 1 trading efficiency increase, 9-10% in reliability improvements compared to benchmarks, and developed a diverse Pareto front has shown and illustrated (helps to develop manage the trade-off between conflicting objectives). Applying an analysis of variance (ANOVA) and 95% confidence interval also indicators that the proposed approach improves robustness and represents statistically significant improvements throughout the presented development efforts. Overall, the proposed integration provides a scalable, resilient, and intelligent optimization of Hybrid Renewable Energy Systems (HRES), and contributes to the conversational development toward more sustainable and economically viable decentralized energy markets.