ESTIMATING THE EARTHQUAKE SOURCE PARAMETERS: SIMULATED ANNEALING VERSUS NELDER-MEAD SIMPLEX ALGORITHM

Authors : Özlem TÜRKŞEN, Ayşen APAYDIN

Pages : 53-66

Doi:10.1501/Commua1_0000000698

View : 21 | Download : 11

Publication Date : 2013-08-01

Article Type : Research Paper

Abstract :The parameter estimation is an important process in many earthscience problems in order to define the features of ground movement. It is clearthat the model nonlinearity makes the estimation of parameters more di¢ cultand more challenging. In this case, metaheuristic algorithms and derivativefree optimization methods are more proper than classical optimization methods. In this study, Simulated Annealing insert ignore into journalissuearticles values(SA);, a well known metaheuristicalgorithm, and Nelder-Mead simplex algorithm, a derivative free optimizationmethod, are used to estimate the earthquake source parameters. The algorithms are applied on a synthetically generated data set. The estimated parameter results show that the SA is better than Nelder-Mead simplex method
Keywords : Simulated Annealing, Nelder Mead simplex algorithm, parameter estimation, fault plane parameters