Research on Vulnerability Assessment of Rainwater and Flood Disasters in Urban High speed Railway Stations Based on Machine Learning

Research on Vulnerability Assessment of Rainwater and Flood Disasters in Urban High speed Railway Stations Based on Machine Learning

연구 분야: Infrastructure

학회: 2024 International Conference on Image Processing, Computer Vision and Machine Learning (ICICML)

In recent years, extreme rainstorm weather has occurred frequently under the global climate change environment, and urban rainstorm flood disasters have shown a trend of increasing frequency and severity. Urban high-speed rail stations are closed and densely populated. Once a rainstorm flood occurs, it will be extremely dangerous. A large amount of rainwater pouring into the station in a short time will not only affect the travel of urban residents but also cause huge economic losses, and even threaten the life safety of urban residents. This study used a composite model constructed by entropy weight TOPSIS and neural network to evaluate the risk of rain and flood disasters at High-speed railway station. The entropy weight TOPSIS method was used to assess the vulnerability and risk exposure of high-speed railway stations to rain and flood disasters, and the neural network model was used to simulate and evaluate the disaster risk. Overall, Nanjing station belongs to level II. A comparison with metro station shows that there are the most metro stations with a vulnerability of Level II, accounting for more than one-third of the total, while metro stations with a vulnerability of Level III or above account for more than half of the total, and there are certain regional and line vulnerability differences.