A novel cloud-based framework for optimal design of stand-alone hybrid renewable energy system considering uncertainty and battery aging

This paper performs a new optimal framework for a hybrid photovoltaic-wind system design integrated with battery storage (PV/WT/Battery), considering cloud-based uncertainty modeling and battery degradation based on real meteorological data from the Sarein-Ardabil region in Iran. The objective function is presented as minimizing the total net present cost (NPC), load loss, and battery degradation cost. The decision variables include the number of PVs, WTs, batteries, inverter power, and the angle of PVs installation, which is optimally determined via a new meta-heuristic optimization algorithm named opposition-based learning and Gradientbased optimizer (OBLGBO). In the proposed framework, the cloud theory method based on combining fuzzy theory and probability statistics has been applied for modeling the energy resources and load demand uncertainties. The simulation results indicate that considering battery degradation costs increases the overall cost of designing hybrid systems. As the cost of degradation increases, reliability indices improve due to an increase in the number of wind turbines and a decrease in the number of batteries. Also, the results demonstrated that incorporating the uncertainties based on the cloud theory increases the design cost, and the system reliability is weakened. Therefore, the proposed optimal framework has presented a real and accurate approach to the optimal design of energy systems with more accurate knowledge of electricity generation costs and the cost of improving reliability in conditions of uncertainties. Moreover, the superior capability of the OBLGBO compared with traditional GBO and the well-known particle swarm optimization (PSO) and grey wolf optimizer (GWO) is proved to achieve lower design cost and better reliability indices.

Automated summary

This paper proposes a new optimal framework for designing a hybrid photovoltaic-wind system integrated with battery storage, taking into account cloud-based uncertainty modeling and battery degradation. The framework uses an optimization algorithm called opposition-based learning and Gradientbased optimizer (OBLGBO) to determine the optimal values for decision variables such as the number of PVs, WTs, batteries, inverter power, and PV installation angle. The cloud theory method is applied to model energy resources and load demand uncertainties. Simulation results show that considering battery degradation costs increases the overall cost, while incorporating uncertainties based on the cloud theory increases the design cost and weakens system reliability. The OBLGBO algorithm proves to achieve lower design cost and better reliability indices compared to traditional optimization algorithms.