A Risk evaluation and control method of highway based on digital twin

With the aim of enhancing the level of safety management during the operation phase of highways, the application of digital twin technology platform is employed. Following the principles of intelligent perception of information, risk assessment, and intelligent decision-making, a set of intelligent risk assessment and control methods for highway traffic is established based on the integration of static and dynamic traffic infrastructure information. Firstly, IoT (Internet of Things) technology is utilized to achieve intelligent perception of dynamic information such as traffic and environment based on the foundation of static data. Multiple-source fusion cloud databases are built in the digital twin system. Secondly, a road segment risk identification using KNN-SVR (K-Nearest Neighbor Support Vector Regression) neural network is proposed. Additionally, a vehicle risk coupling model is introduced to determine the form of traffic risk. Lastly, a rule-based reasoning (RBR) and case-based reasoning (CBR) technology are employed to establish decision support libraries for system and case decisions, aiming to provide specialized support for control decisions. The research outcomes can offer safety improvement decisions and suggestions for transportation safety management departments.