Conventionally, liquefaction-induced settlements have been predicted through numerical or analytical methods. In this study, a machine learning approach for predicting the liquefaction-induced settlement at Pohang was investigated. In particular, we examined the potential of an artificial neural network (ANN) algorithm to predict the earthquake-induced settlement at Pohang on the basis of standard penetration test (SPT) data. The performance of two ANN models for settlement prediction was studied and compared in terms of the R² correlation. Model 1 (input parameters: unit weight, corrected SPT blow count, and cyclic stress ratio (CSR)) showed higher prediction accuracy than model 2 (input parameters: depth of the soil layer, corrected SPT blow count, and the CSR), and the difference in the R² correlation between the models was about 0.12. Subsequently, an optimal ANN model was used to develop a simple predictive model equation, which was implemented using a matrix formulation. Finally, the liquefaction-induced settlement chart based on the predictive model equation was proposed, and the applicability of the chart was verified by comparing it with the interferometric synthetic aperture radar (InSAR) image.

Casein is often used as an eco-friendly wood adhesive. In this study, we used casein for soil cementation by mixing it with Jumunjin sand, sodium hydroxide (SH), and calcium hydroxide (CH) as a standard casein formula. The modified casein binder with different proportions of SH and CH was applied to improve water resistance. Furthermore, a blast furnace slag (BFS) was additionally mixed and reacted with alkalinity of modified casein binder. Thus, three types (standard, modified, and modified + BFS, referred to as STD, MOD, and MBS, hereafter) of casein binders were tested for durability and strength of casein-cemented sand. A piezoelectric sensor was installed within each sample to determine the curing time of the casein-cemented samples. The samples were air-cured at room temperature for seven days and some were repeatedly immersed in water thrice. Unconfined compression and jar slake tests were carried out to evaluate the strength and durability of the casein-cemented sand. Also, the microstructure was analyzed using a scanning electron microscope (SEM). We observed variations of peak conductance and corresponding frequency converged as the curing time increased. It was most significant for the MBS samples, which developed strength early. The unconfined compressive strength (UCS) of the air-cured samples was higher than those repeatedly immersed in water due to wash-off of the casein binder. The UCS of the dry MBS sample was 9900 kPa while that of the immersed sample was 430 kPa, which gradually decreased to 60 kPa upon repeated immersion. The samples with STD and MOD had no resistance to durability and showed cracks on the surface, while the MBS sample exhibited significantly improved durability and no cracks. We found that the MBS binder had a positively significant effect on the durability and strength of casein-cemented sand.

The use of nontraditional soil stabilizers increases. Various new soil binding agents are under study to augment renewability and sustainability of an earth structure. However, despite increasing interest involved in red clay, there is minimal research investigating the stabilizing red clay with polymer. This paper presents the findings obtained by applying the acrylic polymer and epoxy emulsion as binding agent for red clay and that for sand. The epoxy–hardener ratio, amount of epoxy emulsion, and amount of polymer aqueous solution were manipulated to quantify their effects on red clay and sand, respectively. After compacting a pair of cylindrical samples of which diameter and height are 5 cm and 10 cm, respectively, it is cured for 3 and 7 days in a controlled condition. Each pair is produced to represent the engineering performance at each data point in the solution space. An optimal composition of the binding agents for red clay and that for sand mixture are identified by experimenting every data point. In addition, given lime into each sample, the maximum unconfined compressive strength (UCS) endured by red clay sample and that by sand sample are 2243 and 1493 kPa, respectively. The UCS obtained by the sample mixed with clay and sand reaches 2671 kPa after seven days of curing. It confirms that the addition of lime remarkably improves the UCS. When the clay–sand mixture, of which the ratio is 70:30, includes 5% lime, the UCS of the mixture outperforms. Indeed, these findings, i.e., the optimal proportion of components, may contribute to the increase of initial and long-term strength of an earth structure, hence improving the renewability and sustainability of the earth construction method.

Annual coffee consumption has increased to 10 million tons. Of the coffee consumed, 65% is discarded as spent coffee grounds (SCG). However, most SCG are buried in the ground as organic waste. The more coffee consumption increases, the more land is used for disposing of spent coffee. SCG recycling has gotten considerable attention as a solution involved in these issues. The construction community has studied means and methods to recycle SCG as construction materials, such as bricks, subgrade fillers, thermal insulators, etc. This paper presents a new method, which recycles SCG as a construction material, maximally using its acidity. The SCG were hardened with natural binders (i.e., animal glue (AG) and starch (S)) and red clay (RC). The SCG mixtures were pressed with 2 MPa in a cylindrical mold and cured for 7 days. Then, the strength, durability, and pH tests were measured. The AG- and RC-treated SCG sample, which outputs 1933 kPa of strength and a 4.9 pH value, is identified as the optimal sampling method among the acid materials produced in this study. The optimal sample decreases the pH to approximately 7 of water where 68% weight of Ordinary Portland cement was soaked in.

This paper presents findings obtained by evaluating the compressive strength, thermal conductivity, and durability of sand cylinder specimens stabilized with either epoxy emulsion (EM), acrylic polymer aqueous solution (APAS), EM-APAS mixture, or EM-APAS-lime mixture. Given the data obtained from the laboratory test, simulation analysis that uses a heat transfer fluid model of a ground-coupled heat pump (GCHP) system confirms the EM-APAS-lime binder performs best in the compressive strength and thermal conductivity; EM-APAS mixture performs best in the durability. However, the slake durability index of specimens containing EM-APAS-lime is equal to or greater than 80%. In addition, the compressive strength of sand stabilized with the EM-APAS-lime mixture is more than three times that of sand stabilized with cement. The thermal conductivity of sand stabilized with cement and that of sand treated with EM-APAS-lime mixture are 0.1 W/m·K and 0.9–1 W/m·K, respectively. It is confirmed that the heat collection of sand stabilized with EM-APAS-lime outperforms five times over that of sand stabilized with cement. These findings provide admissible evidence that the EM-APAS-lime mixture, which outperforms cement in compressive strength and thermal conductivity, is most suitable for ground improvement binder for GCHP systems.