The rapid penetration of renewable energy sources such as solar photovoltaic (PV) and wind power into modern power grids introduces significant operational challenges, including intermittency, voltage instability, harmonic distortion, and fault vulnerability. Conventional control strategies are often insufficient for handling dynamic grid disturbances and nonlinear system behavior. This study proposes an Artificial Intelligence (AI)-enhanced control framework for grid-connected renewable energy systems to enable adaptive control, predictive fault detection, and resilient operation. The proposed architecture integrates machine learning-based fault classification, adaptive inverter control, and real-time grid condition monitoring. A hybrid dataset composed of simulated grid disturbances and real operational parameters is used to train and validate the AI model. Results demonstrate improved fault detection accuracy, reduced system recovery time, enhanced voltage stability, and improved power quality under dynamic grid conditions. The proposed AI-driven framework enhances grid reliability, supports high renewable penetration, and contributes to resilient and sustainable energy infrastructure.