Solar energy, regarded as an essential renewable energy source, is increasingly becoming a sustainable solution in the global energy transition. However, many solar power plants currently face challenges in identifying key factors that impact the balance between electricity production and consumption. These challenges can significantly reduce the efficiency and reliability of the power grid. This study integrates Explainable AI (XAI) with econometric methods to predict solar power generation and analyze critical influencing factors. Using historical weather data, energy prices, and predicted weather attributes, we developed and evaluated a range of machine learning and econometric models. These models not only demonstrated higher accuracy in forecasting, but also provided insight into the impact of various factors on electricity production. The results indicate that the proposed models play a crucial role in mitigating grid imbalances, thereby enhancing efficiency and reliability in renewable energy integration. Moreover, this research contributes to optimizing energy consumption and minimizing environmental impact. In future work, the research team aims to focus on integrating real-time data and employing deep learning techniques to further improve predictive accuracy and effectiveness.