Solar energy offers a renewable source of power but its fluctuating nature raises concerns about the electrical grid balancing. Network regulators have to estimate the upcoming production to match supply with demand; consequently, power plant operators may be asked to provide accurate forecasts. Planning the thermal or electrical output of solar power plants is thus highly required to ensure a stable power chain supply. This paper presents a solution that couples a meteorological model with a solar power plant performance model. The power output is predicted 24 h ahead in the case of a solar Fresnel power plant. The required Direct Normal Irradiance is inferred from the global horizontal irradiance; the thermal production is evaluated from an optical and thermal model. Our approach has been validated on a 1000 m2 Fresnel power plant, paving the way for model-based storage strategy.