Multi-objective optimization of a hybrid photovoltaic–thermal energy system under load demand uncertainties using Bat algorithm

In this study, an optimally configured hybrid photovoltaic–thermal (PVT) energy system is developed to meet variable electrical, thermal, and cooling demands in a generic building. The system integrates a PVT collector, absorption and electric chillers, a heat pump, thermal storage, and a bidirectional grid connection. Real-world load demand uncertainties are represented through probabilistic scenarios. A multi-objective optimization strategy, based on the Bat algorithm, is applied to simultaneously minimize annual total cost and carbon dioxide emissions. The optimization process, implemented in MATLAB, evaluates multiple system configurations under varying operating conditions. Results show that integrating renewable components such as PVT can substantially reduce environmental impacts, though capital and maintenance costs remain high. Selling surplus electricity to the grid, particularly during peak demand, proves more economically advantageous than internal consumption. The optimal configuration comprises a gas microturbine, single-effect absorption chiller, thermal storage, and bidirectional grid connection, achieving the lowest operational costs and environmental emissions. This research highlights the importance of incorporating uncertainty modeling and renewable integration in designing efficient, sustainable energy systems.