Session 3

Predicting short-term precursors of underground structural failures is essential for minimizing human and economic losses. Among various sensing techniques, electromagnetic radiation (EMR) has shown potential as a non-contact method for detecting precursors to rock failure. While its behavior under quasi-static loading is relatively well understood, EMR characteristics under dynamic conditions remain unclear. To address this, a Split Hopkinson Pressure Bar (SHPB) system was developed with integrated EMR monitoring for dynamic rock fracture testing. Dynamic Brazilian tensile tests on granite successfully captured EMR signals near the fracture point. Fast Fourier Transform (FFT) analysis revealed peaks at 70–220 kHz, which are typical of rock failure, and additional high-frequency components at 0.7–1 MHz. These findings validate the system’s capability and provide new insights into EMR generation mechanisms under dynamic loading conditions.

"Rock blasting is widely used due to its economic and efficient nature in hard rock tunnel construction. However, ""overbreak,"" which refers to excavation beyond the design specifications, is recognized as a problem that leads to increased construction costs and reduced safety. Multiple factors contribute to overbreak, including geological conditions, explosive charge, and drilling status, however a clear causal relationship has not yet been elucidated. Therefore, there is a need to manage overbreak and optimize blast design in tunnel excavation sites. In recent years, researchers have proposed the Overbreak Resistance Factor (ORF) as a quantitative indicator to evaluate the relationship between geological conditions and overbreak. They are also developing a model that aims to predict overbreak volume and understand the influence of geological data using an Artificial Neural Network (ANN). However, collecting data on crack attack angle against wall, which is considered a crucial factor in ORF, is difficult in Japanese tunnels. For this reason, conventional research has employed a method of estimating angles from cross-sectional images, which has led to challenges such as data dispersion and impact on ANN model learning accuracy. This study proposes a method for extracting cracks from point cloud data of 3D models created by Structure from Motion (SfM) to improve the versatility and practicality of the overbreak prediction model. This approach is expected to enable the stable collection of geological data and enhance the learning accuracy of the ANN model."

Weathering is an irreversible naturally occurring phenomenon that negatively affects the structural integrity and geotechnical parameters of rocks. This study aims to understand how weathering influences the physical parameters of tropical metamorphic rocks, by testing three types of gneissic rocks — Granitic Gneiss, Quartzo-Feldspathic Gneiss and Biotite Gneiss — collected along the Pothuhera – Galagedara Section of the Central Expressway, Sri Lanka. Rocks were tested under three main degrees of weathering for each rock type, using UCS, Brazilian tensile strength, slake durability, AIV, and LAAV. Results show that Biotite Gneiss deteriorates the fastest under weathering and is the most vulnerable to wear. Consequently, although Granitic Gneiss has the highest strength initially, it declines rapidly. Comparatively, Quartzo-Feldspathic Gneiss has lower initial strength than the other two rock types, but retains it more effectively as weathering progresses.

Metakaolin-based geopolymers incorporating boron carbide (B4C) as a neutron-absorbing filler show promise for radioactive waste remediation. However, B4C’s weakly polar, negatively charged surface causes poor interfacial bonding, leading to reduced mechanical and chemical stability. In this study, cetyltrimethylammonium bromide (CTAB) was introduced to modify the B4C surface, reversing its charge and improving dispersion. Although CTAB slightly hindered metakaolin dissolution, it preferentially interacted with B4C, mitigating negative effects on geopolymerisation. CTAB also promoted gel formation in the interfacial transition zone, forming a dense, stable microstructure. The synergistic effect of B4C and CTAB enhanced interfacial bonding, mechanical strength, and neutron shielding performance, offering a viable pathway to develop multifunctional geopolymer composites for nuclear applications.

Investigation of geochemical changes of different rock types is essential for identifying the degree and intensity of chemical weathering on different rock types. In this research three major rock types were identified (i.e. – granitic gneiss, quartz feldspathic gneiss, biotite gneiss) and the effect of chemical weathering on those rock types were measured. In each rock three weathering stages were identified (i.e. – residual soil, weathered rock, fresh rock). After that, they were subjected to AAS and ICP-MS analysis for elemental concentration identification. After that different chemical weathering indices like CIA, CIW, WIP, V were calculated according to the depth of rock profiles and weathering trends were developed. After studying those weathering trends, as the most weathered rock type, quartz feldspathic gneiss was identified, as it shows a rapid leaching of mobile elements even in the its rock. The LAAV values of different rock types were compared concerning their weathering indices, and a linear relationship was identified

Accurate pore pressure prediction is critical for safe drilling operations. Conventional prediction methods, which rely on simplified empirical assumptions, often fail to capture the multivariate and non-linear relationships present in complex geological settings. Machine learning (ML) provides a data-driven approach that can model these complexities directly from well log data without relying on predefined physical equations. However, the practical application of ML is often inconsistent due to a lack of systematic understanding of how data preprocessing choices impact final model performance. This study aims to resolve this uncertainty by identifying the optimal combination of preprocessing strategy and ML algorithm for this task. A comparative analysis was conducted across four scenarios: raw data, outlier-capped data, feature-selected data, and combined preprocessing (outlier capping and feature selection) using six ML algorithms to systematically evaluate the effects of outlier capping and the removal of multicollinear features. The findings identify a tuned XGBoost model as the top performer (R² = 0.9789), achieving this optimal result on the raw, unprocessed dataset. This result demonstrates that removing linearly correlated features can be detrimental to advanced models and that the necessity of outlier treatment is algorithm dependent. This study concludes that while the data preparation strategy is universal, it is closely tied to algorithm choice, offering a context-aware framework to enhance model reliability and support interpretability in future research.

This final segment invites reflections from presenters and attendees, synthesizing key insights from the session. The session chair will formally conclude the discussion by summarizing thematic threads, highlighting interdisciplinary contributions, and outlining potential collaborative directions.