The Neuro-genetic approach for estimating the compression index

Abstract

In the last decade, a number of empirical correlations have been proposed to connect the compression index to other soil parameters, such as liquid limit, plasticity index and the void index. This paper presents a correlation study between the physical properties and compression index which was conducted on normally consolidated clay by the hybridization of two approaches (artificial neuronal networks and genetic algorithms). A comparison was made between the measured experimentally and predictions compression indexes. The obtained results indicate that the Neuro-genetic model has the ability to accurately predict the compression index thus be used in practice by geotechnicians.

Keywords

• Compression index
• Neuro-genetic
• Genetic algorithm
• Artificial neuronal network