R-CCS Cafe 第270回（2024年10月4日）

Title:
Development of Disaster Management Digital Twin through Data Transformation for Automated Hazard Simulation System
Abstract:
Japan has many natural disaster hazard potentials, such as the Nankai Trough mega earthquake, earthquakes directly under the Tokyo metropolitan area, wide-area flooding in the Tokyo metropolitan area and storm surge flooding in three major port cities. The Disaster Prevention Digital Twin, which is necessary for physics based numerical simulations to predict the impact of hazards on an entire city, breaks through the limitations of conventional damage estimation and realizes more scientific damage estimation. The objective of this research and development is an automatic generation system for the digital twin. As outputs using the automatically generated digital twin, we propose a probabilistic hazard map, a dynamic hazard map, and a superimposed earthquake, storm surge, and flood hazard map. Probabilistic hazard maps use simulations based on a large number of disaster hazard scenarios. The probabilistic hazard maps clearly distinguish between localized severe damage with very low frequency and small to medium-scale damage with high frequency in various locations, which is impossible with conventional methods that have low detail and resolution, and contribute to more efficient and effective disaster prevention planning. A dynamic hazard map is a dynamic visualization of simulation results based on disaster hazards that have taken place in other regions. By visualizing "what would happen if the same disaster takes place" just after the occurrence of a disaster in another area, the system fosters residents' awareness of disaster prevention and encourages them to take concrete measures, resulting in a sense of security. The superimposed earthquake-surge flood hazard map utilizes a physical simulation of the superposition of a large-scale earthquake and a large-scale flood disaster. This contributes to the study of the superposition, which is literally a national disaster, although it has not yet been experienced. Immediately after a disaster, critical structures such as nuclear power plants require careful safety confirmation. In some cases, it took a considerable amount of time before they were put back into service. It is necessary to use a digital twin of highly detailed structures to significantly shorten the time required for safety confirmation. In this presentation, these digital twins and the SaaS that automatically generates disaster prevention and mitigation information using these digital twins for hazard simulation will be explained.