A digital twin model of urban utility tunnels and its application

Background Multiple pipelines in utility tunnels may lead to various accidents and serious social impact. In the era of digitalization, how to better model the operation of a utility tunnel, dynamically predict the accident evolutions, and support corresponding decision-makings are essential issues. Methods In this study, a CFD-based digital twin framework for accidents in utility tunnels is proposed. First, Kalman filtering is applied to correct the parameter drift of sensors used for long-term monitoring. A data interaction system is then developed based on Internet of Things (IOT) and OPC Unified Architecture (OPC UA) to comprehensively manage data transmission within the utility tunnel. Subsequently, a natural gas leakage prediction model is developed to enable the efficient prediction of the spatial and temporal distribution in the case of leakage. Finally, these components are integrated for visualization in a digital twin platform for natural gas leakage in utility tunnels. Additionally, numerical simulations are employed to validate of the proposed method. Results The utility tunnel data transmission system based on IoT and OPC UA proposed in this paper is case-validated. By comparing the simulation results at 10 s, 20 s, 30 s, and 40 s, the model accurately predicts the methane concentration at the leak position after 10 seconds and maintains acceptable accuracy thereafter. The simulation results of different cases are introduced to verify the reliability of the risk indicator proposed in this paper, which increases with the leakage rate. Finally, A process for visualizing numerical simulation is proposed into a digital twin. Conclusions The proposed predictive digital twin technology facilitates the rapid risk assessment of and emergency management of natural gas accidents in utility tunnels. Based on the results of predictive model, a risk indicator is introduced to evaluate the natural gas accidents.