Risk analysis of fractured rock mass underground structures

ABSTRACT

Recent accidents in underground structures have raised the risk awareness of the geotechnical engineering community. Geotechnical design is subject to significant uncertainties in load and strength parameters as well as in engineering models. However, engineering models which objectively address such uncertainties in design are still scarce. This paper presents an objective framework for the quantification of the risks involved in underground structures excavated in fractured rock masses, where structural failures may occur due to block falls. The framework considers the structure as a distributed system, where falling block probabilities are integrated over the main structural dimension. Random block size and geometry, arising from random joint orientation, are taken into account, as well as uncertainties in joint strength and geometrical parameters. A cost function is used to quantify failure consequences in terms of the block size. The framework is demonstrated in an application to a case study involving a real structure: the Paulo Afonso IV power station cavern. Results of the case study show that the studied cavern presents high reliability and very low risk. The framework proposed herein is shown to be a practical tool for the risk evaluation of underground structures constructed in rock masses, such as caverns and tunnels.

KEYWORDS:

• Risk quantification
• risk analysis
• structural reliability
• probability of failure
• fractured rock mass
• underground structure

Acknowledgements

The authors would like to thank Giacomo Re (Themag Engenharia) and Aurelio Vasconcelos (CHESF) for providing invaluable material for this paper. Also, RocScience is gratefully acknowledged for the software loan of Unwedge.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Gian Franco Napa-García http://orcid.org/0000-0002-3724-4952

André Teófilo Beck http://orcid.org/0000-0003-4127-5337

Tarcisio Barreto Celestino http://orcid.org/0000-0002-5062-6779

Additional information

Funding

This work was supported by National Council for Scientific and Technological Development: [Grant Number 306373/2016-5].