New empirical relationship between probabilistic and deterministic procedures using a genetic algorithm

Abstract

This article analyses the reliability based on the deterministic method and acceleration attenuation in the Chalus City, as well as, obtaining the density function of probability distribution and statistics for cyclic stress ratio (CSR). The density function of probability distribution and the cycle resistance ratio (CRR) can be concluded from the possibility of the cycle of resistance curves. The proposed first-order, second-moment procedure is used to determine the relationship between three factors including, the probability of liquefaction, the safety factor and the reliability index. In this study, the numerical approach of genetic algorithm is utilized to minimize the function of the reliability index. The usage of genetic algorithm model provides a reliable mechanism suitable for a computer program. Two empirical relationships based on PL, $$ N_{\text{SPT}} $$, and CSR with a correlation coefficient and mean error of 95 and 10%, respectively, and the relationship between PL and Fs with a correlation coefficient of 0.887 are provided. Accuracy is measured by comparing the results with those of previous studies. The results of the present study show that a safety factor greater or smaller than 1 does not mean safety and/or liquefaction; to assure liquefaction probability, a reliability-based method should be used for analyses. Moreover, the designed algorithm in this study can be extended by keeping its generalities and having required information from other regions.