SOIL LIQUEFACTION AND ITS EVALUATION BASED ON SPT BY SOFT-COMPUTING TECHNIQUES

Abstract

Damages due to earthquakes in the 21st century have attracted many researchers and engineers on the seismic safety of densely populated cities. One of the fastest growing city Allahabad in Uttar Pradesh (India) is situated on the bank of river Ganga and Yamuna. Most part of river Ganga carries alluvial soil which is an important parameter that influences liquefaction potential of soil. Several factors that also can affect liquefaction behavior of soil that are local site condition, water table location etc. Present research work motivates on liquefaction potential of soil calculated by semi-empirical methods (like Modified Seed's method, Idriss and Boulanger (I&B) method & Tokimatsu and Yoshimi (T&Y) method). The calculated liquefaction value had been used as output/target value for developing the soft computing models. Artificial Neural Network (ANN) & Adaptive Neuro Fuzzy Inference System (ANFIS) techniques have been adapted to the development of soft computing models. I&B method gives more conservative results than other one semi empirical methods.