This paper presents the design of Economic Dispatch (ED) system that minimizes costs and reduces emissions in power networks, focusing on the integration of renewable energy sources such as solar and wind power. The Importance of these clean energy sources has grown alongside economic expansion and the increasing demand for energy, while fossil fuels have become costly and contribute to pollution through combustion processes. This has driven interest in clean energy production. However, renewable energy sources are inherently uncertain in their power generation. Therefore, this research aims to develop a combined model for solar power generation and an ED system for thermal and solar power plants, with the goal of minimizing total production costs and reducing emissions. The design incorporates appropriate tools, such as simulation software, to offer efficient solutions for managing and developing power networks. The study found that while machine learning can be used to predict solar irradiance, clustering solar irradiance data by time of day and season proved effective approach. In terms of energy for ED, the results showed that ignoring emissions resulted in the lowest costs, whereas considering emissions led to higher costs. This study employs machine learning techniques, including Sequential Quadratic Programming (SQP), Genetic Algorithm (GA), and Artificial Neural Networks (ANNs), to optimize dispatch schedules. Experimental results indicate that considering emissions increases overall costs by 40–45% and fuel cost 1–5%, while reducing CO₂ emissions by 7–10%. Additionally, clustering solar irradiance data by time of day and season enhances forecasting accuracy.