Application of ANN to Predict Liquefaction Potential of Soil Deposits for Chandigarh Region, India

Abstract

The phenomenon of liquefaction generally caused by dynamic factors where there is a mass of saturated soil sand. To prevent probable destruction of structures in such areas, prediction of liquefaction potential seems to be necessary. For the purpose of data collection we need to do boreholes at various locations and carry our many experiments, each of which requires a vast expenditure of time and money. Therefore, prediction of liquefaction by existing data leads us to decreasing cost of time and money. Neural networks are intelligent systems that uses specific processing characteristics of the brain The present study attempt to the prediction of liquefaction potential of soil deposits by artificial neural network approach in the Chandigarh region of India. To meet the objective 670 datasets from different boreholes were collected for the development of ANN models. ANN models were trained with seven input parameters by optimum number of hidden layers, epochs and suitable transfer function. Out of total datasets 70 % (470 datasets) of data were used for development of models and 30 % (200 datasets) of datasets were used for testing and validation. The predicted value of liquefaction potential by ANN models were compared with method [1], which shows that ANN method could predict with 95 % accuracy in Chandigarh region of India.