Abstract

A soil behavior type (SBT) chart was developed to more reliably identify organic soil deposits at sites in the state of Michigan based on piezocone (CPTu) data. Organic soils are often highly compressible organic silts, clays, and peats. Many of these soils are fluvial with high void ratios and large compressibilities. These soils are typically removed prior to the construction of roadways and shallow bridge foundations due to the risk of excessive settlement. CPTu soundings were strategically performed alongside companion soil borings in which standard penetration testing (SPT) was performed and from which split-spoon and Shelby tube samples were recovered and tested. While many of the widely used SBT charts characterize the inorganic soils with reasonable agreement to the soil descriptions presented in the soil boring logs, the organic soils are often mischaracterized as inorganic clays. A hybrid SBT approach was therefore developed that yields more accurate characterization of the organic soils. The inorganic soils are reliably characterized by plotting the normalized tip resistance versus the normalized friction ratio. This SBT approach, however, ignores the valuable piezometric data provided by the CPTu. For organic soils, piezometric data helps to more reliably distinguish between inorganic clays and organic soils. As such, a screening tool was implemented to flag likely organic soils by plotting a parameter that accounts for both the tip resistance and pore pressure versus the normalized friction ratio before characterizing the likely-inorganic soils using existing approaches.

Abstract

The Szigetköz (Hungary) is a hotbed of sand boil formation, owing to the combination of a 100-250 m thick gravel layer beneath a relatively thin covering of poor soil with varying thickness. Soil behavior is critical for flood protection in this region. This work proposes a novel way to predict Soil Behaviour Types (SBT) based on detailed CPT data collected from 29 sites in the Szigetköz area using an artificial intelligence (AI) model. The study follows a methodically planned approach that includes data collecting, preprocessing, SBT categorization based on the SBT chart developed by Robertson et al. (1986), and AI model building. The CPT dataset contains critical metrics like cone resistance and friction ratio, which are essential in characterising soil behavior. The AI model, built with powerful machine learning algorithms, is intended to learn complicated associations within data to forecast SBT classifications. Extensive feature selection, hyperparameter tuning, and cross-validation are all necessary steps in model construction to ensure accuracy and generalizability. The results show that the model can accurately forecast SBT classifications for the Szigetköz area, shedding information on the soil's behavior near the Danube River. Spatial distribution visualizations emphasize the region's many SBT categories, giving valuable information for engineering projects, land use planning, and environmental conservation activities. The AI model's interpretability elucidates the major CPT parameters driving SBT forecasts, providing stakeholders with actionable information for decision-making. Furthermore, validation of the model with new, previously unseen CPT data confirms its applicability and robustness in real-world circumstances.

Abstract

The Szigetköz (Hungary) is a hotbed of sand boil formation, owing to the combination of a 100-250 m thick gravel layer beneath a relatively thin covering of poor soil with varying thickness. Soil behavior is critical for flood protection in this region. This work proposes a novel way to predict Soil Behaviour Types (SBT) based on detailed CPT data collected from 29 sites in the Szigetköz area using an artificial intelligence (AI) model. The study follows a methodically planned approach that includes data collecting, preprocessing, SBT categorization based on the SBT chart developed by Robertson et al. (1986), and AI model building. The CPT dataset contains critical metrics like cone resistance and friction ratio, which are essential in characterising soil behavior. The AI model, built with powerful machine learning algorithms, is intended to learn complicated associations within data to forecast SBT classifications. Extensive feature selection, hyperparameter tuning, and cross-validation are all necessary steps in model construction to ensure accuracy and generalizability. The results show that the model can accurately forecast SBT classifications for the Szigetköz area, shedding information on the soil's behavior near the Danube River. Spatial distribution visualizations emphasize the region's many SBT categories, giving valuable information for engineering projects, land use planning, and environmental conservation activities. The AI model's interpretability elucidates the major CPT parameters driving SBT forecasts, providing stakeholders with actionable information for decision-making. Furthermore, validation of the model with new, previously unseen CPT data confirms its applicability and robustness in real-world circumstances.

Abstract

Although gravelly soils have been observed to liquefy in 27 earthquakes in the past 120 years, many engineers believe that gravel cannot liquefy due to its high hydraulic conductivity. Gradations from gravel liquefaction case histories have shown these deposits typically contain 25 to 40% sand, reducing the hydraulic conductivity and enabling excess pore pressures to cause liquefaction. While cone penetrometers (CPT), typically used to evaluate liquefaction resistance in sand, may show increases in penetration resistance due to their small diameter relative to gravel particles, the CPT has successfully predicted gravel liquefaction for looser sandy gravels. Case histories in Wellington, New Zealand demonstrate the successful identification of gravel liquefaction hazards using CPT. Although some layers in the profile indicated high penetration resistance, most of the profile was correctly predicted to liquefy. The Soil Behavior Type (SBT) from the CPT did not consistently indicate a sandy gravel profile but was often classified as behaving like a sand or silty sand; likely influenced by higher sand percentages between gravel particles. To evaluate the ability of the CPT to characterize gravelly soils and their liquefaction potential, additional field case histories are desirable. This paper presents test results from two case histories, one in Wellington, New Zealand, and one in Petrinja, Croatia, where gravels have liquefied. In both cases, the CPT occasionally overestimated liquefaction resistance in gravel layers. The advantages of using a 74 mm diameter Dynamic Cone Penetrometer (DPT) are also highlighted with companion testing.