During the construction of an underground excavation, damage occurs in the surrounding rock mass due in large part to stress changes. While the predicted damage extent impacts profile selection and support design, the depth of damage is a critical aspect for the design of permeability sensitive excavations, such as a deep geological repository (DGR) for nuclear waste. Review of literature regarding the depth of excavation damage zones (EDZs) indicates three zones are common and typically related to stress induced damage. Based on past developments related to brittle damage prediction using continuum modelling, the depth of the EDZs has been examined numerically. One method to capture stress induced damage in conventional engineering software is the damage initiation and spalling limit (DISL) approach. The variability of depths predicted using the DISL approach has been evaluated and guidelines are suggested for determining the depth of the EDZs around circular excavations in brittle rock masses. Of the inputs evaluated, it was found that the tensile strength produces the greatest variation in the depth of the EDZs. The results were evaluated statistically to determine the best fit relation between the model inputs and the depth of the EDZs. The best correlation and least variation were found for the outer EDZ and the highly damaged zone (HDZ) showed the greatest variation. Predictive equations for different EDZs have been suggested and the maximum numerical EDZ depths, represented by the 68% prediction interval, agreed well with the empirical evidence. This suggests that the numerical limits can be used for preliminary depth prediction of the EDZs in brittle rock for circular excavations
Machine learning methods for data processing are gaining momentum in many geoscience industries. This includes the mining industry, where machine learning is primarily being applied to autonomously driven vehicles such as haul trucks, and ore body and resource delineation. However, the development of machine learning applications in rock engineering literature is relatively recent, despite being widely used and generally accepted for decades in other risk assessment-type design areas, such as flood forecasting. Operating mines and underground infrastructure projects collect more instrumentation data than ever before, however, only a small fraction of the useful information is typically extracted for rock engineering design, and there is often insufficient time to investigate complex rock mass phenomena in detail. This paper presents a summary of current practice in rock engineering design, as well as a review of literature and methods at the intersection of machine learning and rock engineering. It identifies gaps, such as standards for architecture, input selection and performance metrics, and areas for future work. These gaps present an opportunity to define a framework for integrating machine learning into conventional rock engineering design methodologies to make them more rigorous and reliable in predicting probable underlying physical mechanics and phenomenon.
22 August 2017 in Geotechnical and Geological Engineering