The study on forecasting the surface rupture width under strong earthquake based on information diffusion methodology

The surface rupture triggered by strong earthquakes will extensively damage the buildings on the ground. It is significant for choosing construction sites to forecast accurately and calculate reasonable building lien distance. A method is proposed aiming at forecasting the surface rupture width under strong earthquake based on information diffusion methodology in this paper. Firstly, magnitude and the thickness of overburden soil were taken as main evaluation indexes, and then fuzzy relation between two evaluation indexes and surface rupture zone width was established based on the information diffusion principle and fuzzy approximate deduction. Secondly, cross-validation was used to evaluate the model’s stability and accuracy. The performance of information diffusion was compared with multiple linear regression model, quadratic regression model, and BP neural network model. The results show that information diffusion methodology can deal with nonlinear relationship of every index. Finally, the validity and feasibility were proved by instance analysis. Copyright Springer Science+Business Media Dordrecht 2015