This work developed two smart greenhouses for the protected cultivation of banana and orchid seedlings. The greenhouses feature control and automation systems, bioclimatic architecture and strategies, and energy systems. A sensor network monitors indoor environmental variables, and a weather station monitors outdoor environmental variables. A computational system based on artificial intelligence is responsible for the autonomous control of the greenhouses. This article presents the development of an artificial lighting system for light supplementation in smart greenhouses. The artificial lighting system was defined from modeling and simulation in a computational tool. The proposed system was installed in two smart greenhouses in southern Brazil. The article also presents the development of methods for estimating electricity consumption based on assumptions and operational profiles of the system and the experimental measurement of the artificial lighting system performed according to the procedure described in technical standards. The article presents the energy performance of the agrivoltaic system applied to smart greenhouses. From August 2022 to July 2023, the final yield of the agrivoltaic systems of the smart greenhouses was 1418.91 kW h/kWP and 1490.68 kW h/kWP. The annual electricity consumption of the artificial lighting systems of the smart greenhouses ranges between 3290 kW h and 3825 kW h.