Evaluation of Index Properties Affecting Unconfined Compressive Strength of Jet-Grouted Soil Columns for Bangkok Soils-Paper

Paper

Pdf Summary

The study evaluates soil index properties that influence the unconfined compressive strength (qu) of jet-grouted columns in Bangkok. Researchers compiled field and laboratory data from samples collected on the MRTA Blue Line, Blue Line Extension, and Orange Line projects, where large-diameter, double-tube jet grouting was used.<br /><br />An artificial neural network (ANN) was trained and validated to estimate qu from soil properties. The best-performing inputs were unit weight and percentage passing the No. 40 sieve for sandy and silty soils, and unit weight and plasticity index for clayey soils. These models achieved validation R² values of 0.75–0.86, indicating reasonable predictive performance. The authors recommend using the resulting prediction charts for comparable Bangkok soils.<br /><br />The study also assessed the relationship between qu and the modulus of elasticity, E50. Estimated correlations were E50 = 69.32qu for sandy soils and E50 = 80.83qu for clayey soils; the lower-bound relationship across the data was E50 = 50.84qu. This is below the commonly used Japan Jet Grout Association correlation of E50 = 100qu. The authors attribute the difference partly to Thailand’s lower cement content and different nozzle configuration, as well as differences in soil characteristics.<br /><br />They recommend using the lower-bound correlation for jet-grouted soils in Thailand, while noting that the prediction charts and correlations reflect the specific soils and equipment studied. Similar methods could be applied to develop region-specific relationships elsewhere.

Keywords

jet grouting

unconfined compressive strength

Bangkok soils

artificial neural network

soil index properties

modulus of elasticity

sandy soils

clayey soils

predictive model

prediction charts