The demand for sensory information is expanding in various places, namely, IoT-based smart cities, automated street lights, green agriculture, etc. The energy-consuming rate is generally too difficult to exchange data between two sensor nodes. Moreover, we cannot change the batteries of sensor nodes at regular intervals due to the harassing area or unreachable places. However, the solar energy harvesting technique provides unlimited energy resources to sensor nodes at the least cost. However, one general problem comes with solar energy harvesting systems, which is called energy conversion rate, due to the weather fluctuations in various time slots. Regarding this issue, the artificial intelligence (AI) approach estimates accurate performance to predict external energy sources. These methods include artificial neural networks, Q-learning, and linear regression. Our findings demonstrate that AI-driven methods significantly improve the adaptability and performance of energy harvesting under varying environmental conditions, marking a step forward in sustainable energy solutions.