A novel liquefaction study for fine-grained soil using PCA-based hybrid soft computing models

Earthquake-induced liquefaction is an unpredicted phenomenon that causes catastrophic damages and devastation to the environment, structures, human life. The assessment of soil behavior a decisive work for geotechnical engineers especially during designing phase any civil engineering projects. These decisions implicate tedious costly experimental procedures extensive evaluation. Considering these facts, present study aims simplify process evaluating soil’s in broader domain involving least datasets. Three PCA (principal component analysis)-based advanced hybrid computational models, namely PCA-ANN, PCA-ANFIS, PCA-ELM were developed predict soils. dimension reduction technique, i.e. PCA, was used avoid multicollinearity effect course development said models. Geotechnical parameters, plasticity index, SPT blow count, water content liquid limit ratio, bulk density, total stress, effective fine along with other seismic input variables, such as ratio peak ground acceleration due gravity, magnitude earthquake develop predictive accuracy proposed models evaluated via several fitness parameters. In end, best model determined using novel tool called Rank Analysis. Based on results, it has been established can be considered new alternative assist task assessing potential preliminary design stage project.