Modelling dynamic behaviour of sand–waste tires mixtures using Neural Networks and Neuro-Fuzzy

Abstract

This investigation describes the results of a series of cyclic triaxial tests on sand–waste tires mixtures, and applications of Neural Networks (NN) and Neuro–Fuzzy (NF) for the prediction of damping ratio and shear modulus of the mixtures were tested. In the cyclic triaxial testings, shear modulus and damping ratio of the sand–waste tires mixtures at various ratios have been measured for a strain range of .0001% up to .04%. Test results show that the shear modulus and damping ratio of the mixtures are strongly influenced by the waste tire inclusions. It is seen that the greater the proportion of waste tire crumbs or tire buffings on the sand, the greater is the damping ratio and the less is the shear modulus, regardless of confining pressure. The input variables in the developed NN and NF models are the (1) waste tires contents which are 0, 10, 20 and 30, (2) waste tires types which are tire crumbs and tire buffings, (3) confining pressures which are 40, 100 and 200 kPa and (4) strain level and the outputs are (1) damping ratio and (ii) shear modulus. The performance of proposed NN models (R ² = .99 for shear modulus, and R ² = .98 for damping ratio) is observed to be more accurate than the NF models (R ² = .96 for shear modulus, and R ² = 0.97 for damping ratio).

Keywords:

• sand
• waste tire
• cyclic triaxial testing
• Neural Networks
• Neuro-Fuzzy