Predicting groutability of granular soils using adaptive neuro-fuzzy inference system

TEKİN E., AKBAŞ S. O.

NEURAL COMPUTING & APPLICATIONS, vol.31, no.4, pp.1091-1101, 2019 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 31 Issue: 4
  • Publication Date: 2019
  • Doi Number: 10.1007/s00521-017-3140-3
  • Journal Name: NEURAL COMPUTING & APPLICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1091-1101
  • Keywords: Groutability, Adaptive neuro-fuzzy inference system (ANFIS), Granular soil, Microfine cement, PERMEABILITY, ANFIS, MODEL
  • Gazi University Affiliated: Yes

Abstract

In this paper, the applicability of adaptive neuro-fuzzy inference system (ANFIS) for the prediction of groutability of granular soils with cement-based grouts is investigated. A database of 117 grouting case records with relevant geotechnical information was used to develop the ANFIS model. The proposed model uses the water-cement ratio of the grout, the relative density and fines content of the soil, the grouting pressure, and the ratio between the particle size of the soil corresponding to 15% finer and that of grout corresponding to 85% finer as input parameters. The accuracy of the proposed ANFIS model in terms of the corresponding coefficient of correlation (R) and root mean square error (RMSE) values is found to be quite satisfactory. Furthermore, a comparative analysis with existing groutability prediction methods indicates that the ANFIS model demonstrates superior performance.