The active development of nuclear power is an important strategic choice to achieve carbon peak and neutrality goals. It is an essential study to realize the economic operation of nuclear plants on the premise of ensuring the safety and reliability of nuclear power plants. Intelligent monitoring and maintenance technology is an important development tendency that can effectively improve the operation safety and maintenance economy of nuclear power plant. In this article, an intelligent monitoring and maintenance system of AP1000 significant equipment in Sanmen Nuclear Power Plant is developed based on the equipment operation mechanism and failure mode analysis by using advanced pattern recognition technologies including intelligent sensing, big data analysis, multidimensional spatio-temporal correlation condition monitoring and fault diagnosis technologies. It can realize online real-time equipment performance monitoring, automatic trend analysis, automatic health condition evaluation, early anomaly identification and automatic fault diagnosis of equipment in nuclear power plant and promote the transformation of maintenance mode from time-based maintenance to condition-based maintenance. The intelligent monitoring and maintenance system has been applied in various mechanical and electrical equipment of Sanmen Nuclear Power Plant such as reactor coolant pump, main steam turbine, main transformer, heat exchanger and achieved certain technical achievements and economic benefits. The construction of intelligent monitoring system has certain demonstration and popularization.