Estimation of physical, mechanical and hydrological properties of permeable concrete using computational intelligence approach

This present work estimates the physical, mechanical and hydrological properties of PC using computational intelligent technique on the platform of support vector regression (SVR) due to excellent generalization and predictive ability of SVR in the presences of few descriptive features.

Abstract

Permeable concrete (PC) has gained a wide range of applications as a result of its unique properties which result into highly connected macro-porosity and large pore sizes. However, experimental determination of these properties is intensive and time consuming which necessitates the need for modeling technique that has a capability to estimate the properties of PC with high degree of accuracy. This present work estimates the physical, mechanical and hydrological properties of PC using computational intelligent technique on the platform of support vector regression (SVR) due to excellent generalization and predictive ability of SVR in the presences of few descriptive features. Four different models were built using twenty-four data-points characterized with four descriptive features. The estimated properties of PC agree well with experimental values. Excellent generalization and predictive ability recorded in the developed models indicate their high potentials for enhancing the performance of PC through quick and accurate estimation of its properties which are experimentally demanding and time consuming.