Risk Assessment of Typhoon Storm Surge Disasters in Guangdong Province Based on the Improved Fuzzy Bayesian Network

摘要

To expand insufficient information resource and express fuzzy uncertainty for risk assessment, a new Bayesian network was proposed in this paper. Information diffusion model and triangle fuzzy values were employed to improve the prior probability, using Monte Carlo method to establish the conditional probability tables with the help of AHP. According to the data of typhoon storm surge disasters, economic, cultural and other characteristics, the zoning map of storm surge risk based on Grid and GIS was realized to study disaster risk, vulnerability and capacity of disaster prevention in Guangdong. It showed the results tallied with the practical situation, which may provide the help to reduce the loss caused by storm surge in China.