Aiming at the problems of “many points，long lines，wide areas”，high safety risks in maintenance operations of the public channel of the Shanghai Port，we use on-site investigation，theoretical modeling，and system development to carry out the key technology research on the intelligent management of the channel maintenance and support，and propose a method of abnormal real-time detection of ship operation based on AIS data，which can realize the goal of “one picture full perception，one-click whole situation” for the whole process of channel maintenance and support. The result is fully applied in Yangshan Port，Waigaoqiao Port，Luojing Port and other public channel maintenance and support in Shanghai Port. This method has significant economic and social benefits and broad prospect for promotion and application.