Safe operation plan for high-speed rail based on dynamical prediction model of short-term rainfall
Minjia Tan, Yifei Yang, Lishuang Bian, Jinbo Zhang, Susumu Ohno · KSCE Journal of Civil Engineering · 2025
To ensure the safe operation of high-speed rail in rainstorm conditions, this paper explores Cloud Model as a framework for predicting rainfall distribution. Recognizing the urgency required for high-speed rail operations, we propose a dynamical short-term rainfall prediction model based on Cloud Model. This model differentiates between historical precipitation data by dividing it into a historical cloud model and a trend cloud. These components are integrated into a unified predictive cloud model using a blank qualitative concept. Subsequently, the maximum entropy principle is employed to determine the proportion of corresponding rainfall durations within the total precipitation period, thereby enabling the calculation of short-term precipitation at any given time. Moreover, this paper establishes an operational mechanism linking rainfall predictions to safe operation plan. A case study of super typhoon "Lekima" illustrates that the proposed model exhibits high predictive accuracy and a strong multi-correlation coefficient, thereby assisting station managers in developing effective operational plans under adverse rainstorm conditions. The research contributes novel insights to the knowledge base by integrating advanced predictive modeling with practical applications in railway safety management, ultimately enhancing the robustness and reliability of high-speed rail operations during extreme weather events.