With the increasing adoption of renewable energy sources, researchers are actively engaged in the development of smart grid technologies. This study introduces a home energy management system (HEMS) designed to optimize home microgrid (HMG) operation by integrating electric vehicles (EVs) through a hierarchical three-level distributed control approach. The proposed system governs the power transfer of distributed energy sources through converter control at the primary level, while employing an intelligent optimal power exchange technique for EVs and energy storage (ES) at the secondary level. To achieve optimal energy demand management, the system employs a two-layer strategy, comprising offline and online scheduling, utilizing particle swarm optimization and artificial neural network to reduce computing time. The offline scheduling formulates a deterministic optimization model to minimize HMG operation costs, thereby extending ES lifetime through optimal load-sharing. The online scheduling approach enables real-time performance under uncertainty and accommodates plug-and-play renewable energy sources, HMGs, loads, and batteries. Furthermore, the proposed algorithm is tested using MATLAB/Simulink simulations over a 24-h period, affirming its ability to respond promptly to HEMS status changes and adapt decision-making methods to new conditions with improved training time and reduced mean square error. In addition, experimental studies on two laboratory-based HMGs demonstrate the feasibility of the proposed HEMS.