Cheetah optimization based ANN maximum power point tracking for PV and wind in hybrid AC-DC microgrid

Abstract The integration of multiple Distributed Generations (DGs) within microgrids introduces complexities in power management due to the variable nature of non-conventional sources and the need for efficient energy distribution among diverse loads. To address these challenges, this work proposes an MPPT method for PV & Wind in microgrids by implementing a Cheetah Optimization Algorithm (COA)-based Artificial Neural Network (ANN). The COA, inspired by the hunting strategy and agility of cheetahs, is used to optimize the weights and biases of the ANN, enabling precise prediction of maximum power points under fluctuating environmental conditions and varying load demands. This approach ensures highly efficient power extraction from renewable sources such as photovoltaic (PV) systems and wind turbines integrated into the microgrid. The COA-ANN based MPPT method significantly improves the microgrid’s operational stability and reliability by dynamically adapting to changes in energy availability and load profiles in real time. Notably, the proposed method achieves exceptional efficiency across different conditions reaching 98.6% at 1000 W m −2 irradiance for PV and 97.4% at 12 m s −1 wind speed for wind power systems. Even under low-irradiance conditions (400 W m −2 ), it maintains high efficiency at 98.7%, surpassing traditional MPPT approaches. These results underscore the superior performance of the COA-ANN controller in maximizing energy harvest, minimizing power losses, and maintaining balanced load distribution. Overall, this research contributes a scalable solution for effective utilization of non-conventional resources and intelligent energy management in microgrids, promoting the effective integration of multiple DGs and supporting the broader adoption of sustainable, non-conventional energy technologies in modern power systems.