Abstract: in order to solve the security threats faced by the wind farm monitoring system in the Internet of things environment, as well as the problems of lagging response and high false alarm rate of the traditional static protection scheme. A security protection method of wind farm monitoring automation system based on Internet of things technology is proposed. By building a multi-dimensional trust quantification model of terminal devices, real-time dynamic access control is realized; Based on the federal learning framework, the local detection model is trained distributed in each wind turbine group, and the global anomaly recognition model is aggregated to improve the overall defense capability of the system under the premise of ensuring data privacy. Experimental results show that the attack response time of the proposed method is improved by more than 40% compared with the traditional method, and the false positive rate is reduced to 5.2%, and it still maintains stable performance in high load environment. The proposed method effectively improves the security, real-time and adaptability of the wind farm monitoring system, and provides a reliable security protection solution for the smart energy system.