Comprehensive Degradation Modeling and Vine Copula-Based Stochastic Optimization for Hybrid Renewable–CHP Microgrid with Electrical and Thermal Energy Storage Using Enhanced Glider Snake Optimizer

Hybrid microgrids featuring renewables, combined heat and power (CHP), and storage are highly promising systems; however, they suffer from several optimization difficulties, such as degradation, uncertainty, and multi-objective problems. In this paper, a novel optimization framework of a degradation-based stochastic nature considering a photovoltaic (PV)–wind turbine (WT)–CHP–battery energy storage system (BESS)–thermal energy storage (TES) microgrid is presented. The proposed optimization framework has three main novelties: (1) a comprehensive degradation model that includes all forms of degradation, namely, calendar, cycle, electro-thermal, and operation-based degradation; (2) a Multi-Objective Improved Glider Snake Optimizer (MOIGSO) along with Dynamic Gliding Structure Variability based on Movement Entropy (DGSV-ME); and (3) vine copula-based modeling of the non-Gaussian dependency among five different random variables: solar irradiation, temperature, wind speed, electrical load, and thermal load. The presented optimization framework solves a two-objective problem by minimizing both economic and emission costs through the MOIGSO. Three configurations (PV + WT + BESS; PV + WT + CHP + BESS; PV + WT + CHP + BESS + TES) are considered under three cases. According to the results, a degradation-aware design increases the annual cost by 9.8–13.1% but increases the battery lifespan up to 20 years, thereby providing lifecycle sustainability. The MOIGSO obtains a hypervolume index of 0.978 and an inverted generational distance (IGD) of 0.0074 and therefore beats the Multi-Objective Glider Snake Optimizer (MOGSO), Multi-Objective Particle Swarm Optimization (MOPSO), Non-Dominated Sorting Genetic Algorithm II (NSGA-II), and Multi-Objective Whale Optimization Algorithm (MOWOA). A two-layer predictive model (MPC)-based energy management system (EMS) saves 2.9–5.5% of cost, 12.4–15.8% of battery cycling, 6.8–8.2% of fuel, and 6.5–8.0% of emissions and increases renewable usage by 6.5–8.3%. The vine copula decreases the cost expectation by 5.6%, the cost standard deviation by 31.5%, and the computation time by 42.2% compared to Monte Carlo simulation. The proposed framework provides a tangible approach to the implementation of sustainable microgrids, emphasizing degradation-aware designs, prediction-driven energy management, and complex uncertainty modeling.