- Title
- A stochastic approach for the shear strength of rock discontinuities
- Creator
- Casagrande, Davide Donato
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2018
- Description
- Research Doctorate - Doctor Philosophy (PhD)
- Description
- Natural rock discontinuities play a key role in the stability of rock masses but estimating their shear strength is a nontrivial task. Engineers or researchers are often relying on results obtained from small-sized specimens (cored from the rock mass) because of the inherent complexity of accessing discontinuities within a rock mass. As a result, the outputs are often influenced by the well-known scale effect. Some methods have been developed to estimate the shear strength of discontinuities when some traces are visible, but these methods present limitations and to date, there is no consensus on the most appropriate method to estimate the shear strength of large in situ discontinuities. A new approach is proposed in this thesis to predict the shear strength of discontinuities directly at the intended scale, hence avoiding changes of scale and minimizing (if not eliminating) the scale effect. The new approach is based on the idea that for natural rock surfaces, it can reasonably be considered that traces contain information that are representative of the whole surface. It is then proposed to extrapolate the available 2D information (trace) to 3D data (surfaces) via a rigorous random field model. Producing a large number of synthetic surfaces and estimating their shear strength results in predicted mean shear strength with a predicted standard deviation. The first part of the research focuses on the development and validation of an efficient semi-analytical model that can predict peak and residual shear strength of discontinuities. The model is mechanistic in nature and only requires the 3D surface geometry, shearing direction and material strength parameters. The second component of this research is the development of a random field model that allows the generation of synthetic surfaces representative of an original rock surface. An extensive statistical characterization of traces and surfaces has been conducted in order to identify the most appropriate roughness descriptor that will be used to as an input in the random field model. Although intended for large discontinuity, the new approach was validated at the laboratory scale because of the need to conduct and repeat a large number of tests in the laboratory under controlled conditions. The new approach was validated and it was concluded that about 100 simulations are required to obtain a meaningful distribution of shear strength. In addition, a sensitivity analysis revealed that the correlation length and the variance of heights are two key parameters influencing the prediction. The final part of the research deals with the preliminary application of the new stochastic approach to a large in situ surface (2m per 2m). Although it was not possible to experimentally measure the shear strength of the discontinuity (being located in a natural reserve), its shear strength was estimated from the semi-analytical model and compared to the predictions made by the new stochastic approach where only the information from a trace are used. A satisfactory math was obtained hence validating this new approach for the prediction of shear strength of large in situ discontinuities.
- Subject
- rock joint; discontinuity; shear strength; random field; stochastic; scale effect
- Identifier
- http://hdl.handle.net/1959.13/1385870
- Identifier
- uon:32307
- Rights
- Copyright 2018 Davide Donato Casagrande
- Language
- eng
- Full Text
- Hits: 3031
- Visitors: 3055
- Downloads: 810
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 11 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 245 KB | Adobe Acrobat PDF | View Details Download |