Abstract: To improve the capabilities of automatic identification and rapid location of pipe burst accidents for the equipment pipe networks of fire protection, Heating, Ventilation and Air Conditioning (HVAC), water supply and drainage in concealed environments, as well as the overall management level of pipeline distribution, this paper adopted 3D modeling and digital twin technology, and constructed a digital twin detection model for equipment pipe networks based on the Dempster-Shafer (D-S) evidence theory. It generated multiple batches of simulated pipe burst data and completed the effectiveness verification of the pipe burst inference algorithm. Considering the insufficient hardware performance of on-site operators’ terminal devices in rendering 3D models, the paper proposed a lightweight optimization scheme for 3D model rendering. It designed and carried out verification experiments. The experimental results show that the identification accuracy of the proposed model for pipe bursts of equipment pipe networks exceeds 95%, and the model rendering efficiency is significantly improved. The proposed model possesses good engineering application and promotion value.