Microgrids, with their ability to integrate renewable energy sources, play a crucial role in achieving sustainable and resilient energy systems. Effective planning and optimization of microgrids, particularly considering the inclusion of compressed air energy storage (CAES) systems, are essential for maximizing their benefits. This study proposes a novel approach, the Hybrid Artificial Neural Network-Modified Dragonfly Algorithm (HANN-MDA), for determining the optimum capacity of CAES in microgrid planning. The HANN-MDA method combines the learning capabilities of artificial neural networks with the optimization power of the modified dragonfly algorithm. The proposed method aims to minimize the overall cost of microgrid operation while considering the integration of renewable energy sources and the storage capabilities of CAES. Simulation results demonstrate the effectiveness of the HANN-MDA method in accurately determining the optimal CAES capacity, leading to improved microgrid performance and cost savings. The findings highlight the importance of considering CAES in microgrid planning and the potential of the HANN-MDA method for achieving efficient and economically viable microgrid designs. Manuscript profile