Due to factors such as foundation settlement, improper operations, and external load changes, atmospheric storage tank may become deformed or tilted. In view of the problems of heavy workload and insufficient accuracy of conventional monitoring of large atmospheric storage tank methods, which are affected by the field environment and personnel skills, in this paper, a 3D laser scanner is used to scan the tank and acquire point cloud data of the tank. The digital tank model is generated by splicing, noise reduction, filtering, and grid processing. The results show that this method can efficiently monitor the circularity, plumbness, and settlement of storage tanks, improve the inspection accuracy, and expand the inspection range. Additionally, data storage and processing are more simplified compared to traditional methods.