Apsey, Jason.
Abstract:
In the South African mining industry, the design of tunnel support systems is
generally based on empirical methodologies that consider rockmass characteristics as well as the type of loading (e.g. seismic) that the excavation experiences. The design methodologies are by no means infallible, and work is continually being conducted to improve the classification of excavation conditions and thereby improve the selection of a suitable support system.
This study is concerned with finding a means to monitor the installed support units
rather than with improving the classification methodologies. It is postulated that
with the extraction of accurate information describing the state of any support unit
at any given time, areas of instability in the tunnel can be readily identified and
strengthened~ Also, the information gathered as to the behaviour of the support
units in a particular region can be used to assist in understanding the
environmental characteristics of that region (rockmass, loading, etc.).
A material survey was conducted to identify suitable candidates that could feasibly
be used in either a passive (feedback when interrogated) or active (constant
feedback) structural health monitoring system. The preferred candidates identified
in this study are the group of passive smart materials referred to as TRIP steels,
which are a subset of strain memory alloys. TRIP steels exhibit microstructural
changes from paramagnetic austenite to ferromagnetic martensite as a function of
increasing deformation at a given temperature.. The strength of the magnetic field
at critical locations provides an indication as to the health state of the component.
Because of their high strengths and ductility, TRIP steels can be used as what
amounts to a self-monitoring support unit (interrogation apparatus required).
Finite element methods are a practical means of predicting the mechanical and
magnetostatic behaviour of TRIP steel structural members once material equations
have been established by experiment.
