Predicting resilient modulus of compacted subgrade soils under influences of freeze–thaw cycles and moisture using gene expression programming and artificial neural network approaches. (May 2021)

• Record Type: Journal Article
• Authors: Zou, Wei-lie; Han, Zhong; Ding, Lu-qiang; Wang, Xie-qun
• Is Part Of: Transportation geotechnics. Volume 28(2021)
• Publication Date: 2021-05
• Subjects: Pavement subgrade soil -- Resilient modulus -- Gene expression programming -- Artificial neural network -- Environmental influences; Engineering geology -- Periodicals; Soil mechanics -- Periodicals; Rock mechanics -- Periodicals; Transportation -- Periodicals; 624.15105
• DOI: 10.1016/j.trgeo.2021.100520
• Languages: English
• ISSNs: 2214-3912

Abstract

Abstract: This study aims at developing a gene expression programming (GEP) model and an artificial neural network (ANN) model to predict the resilient modulus ( M R ) of compacted pavement subgrade soils based on their physical properties, external stress states, and environmental factors. A database of 2813 M R measurements derived from 12 subgrade soils with different moisture-temperature histories was established for model development and validation. Influencing factors considered in this database include the weighted plasticity index (wPI), dry unit weight ( γ d ), confining stress ( σ c ), deviator stress ( σ d ), moisture content ( w ), and the number of freeze–thaw cycles (NFT ). Sensitivity analysis was conducted to evaluate the importance of each factor. The order of influencing factors by decreasing importance was found to be wPI, γ d, w, NFT, σ d, σ c and the importance of the σ c and σ d is remarkably lower than other factors. This indicates that the M R of compacted subgrade soils is strongly dependent on the soil type (wPI, γ d ) and is more sensitive to environmental factors (NFT, w ) than external stress states ( σ c, σ d ). The developed GEP and ANN models reasonably predicted the M R in the database and achieved better performance compared to several existing empirical models.