In West Africa, particularly in rural areas, plant oil production development is limited by poor access to affordable energy and processing technologies. While solar energy is widely available, its use for powering oil extraction remains limited. Existing systems are often unsuited to the conditions of small-scale producers, and energy supply is often unreliable. Mechanical pressing stands out as a widely adopted method for plant oil extraction in rural areas due to its simplicity and lower initial investment. At the same time, the performance of oil extraction varies depending on the type of oilseed and how it is processed. Most studies focus only on oil recovery, without looking at energy demand. This makes it difficult to design solar systems and to apply smart control methods. The aim of this work was to optimize a solar-powered peanut oil extraction system by combining photovoltaic computational sizing, multi-criteria process optimization, and AI-based control. The research was divided into three steps: solar system sizing, process optimization, and controller development. Overall, this study introduced a method for modeling the energy demand of plant oil production using a synthetic load profile, optimized the extraction process for both oil yield and energy use, and developed a smart controller using reinforcement learning. The system was tested under real conditions and showed improved oil yield and operational efficiency. While the control strategy focused on rotational speed, it could be extended to other parameters such as pretreatment. The methods used in this work can be adapted to other oilseeds, equipment, and regional contexts to support the wider use of solar-powered oil extraction in rural areas.