RocSupport Introduction Manual

RocSupport
Introduction Manual
Topics Covered:
• Rock Support Interaction Solution Method
• Ground Reaction Curve
• Support Reaction
• Solution Methods
• Deterministic Analysis
• Probabilistic Analysis
RocSupport Introduction Manual

Introduction
RocSupport is a quick and simple to use program for estimating the deformation of
circular tunnels in weak rock, and visualizing the tunnel interaction with various support
systems.
The analysis method used in RocSupport is often referred to as “rock support
interaction” or “convergence-confinement” analysis. This analysis method is based on
the concept of a “ground reaction curve” or “characteristic line,” obtained from the
analytical solution for a circular tunnel in an elasto-plastic rock mass under a hydrostatic
stress field.

Applicability of Method
The main assumptions in the analysis method are as follows:
• tunnel is circular
• in-situ stress field is hydrostatic (i.e. equal stress in all directions)
• rock mass is isotropic and homogeneous. Failure is not controlled by major structural
discontinuities.
• support response is elastic-perfectly plastic
• support is modeled as an equivalent uniform internal pressure around the
entire circumference of the circular tunnel
This last assumption in particular (that support is uniform around the entire circumference of
the tunnel), should be carefully considered by the user, when comparing actual tunnel
behavior, and calculated results using RocSupport.
The assumption of uniform support pressure implies that:
• shotcrete and concrete linings are closed rings
• steel sets are complete circles
• mechanically anchored rockbolts are installed in a regular pattern which completely
surrounds the tunnel.
Because this will not usually be the case, actual support capacities will be lower, and
deformations larger, than those assumed in RocSupport.
The idealized model used for a RocSupport analysis is not intended to replace detailed
final design and analysis requirements for tunnel support. In general, this will require
numerical analysis (e.g. finite element), particularly for tunnels with large strain.
However, a great deal can be learned about the interaction of tunnels in weak rock, with various
support systems, by carrying out parametric studies using RocSupport, in which different
combinations of in-situ stress levels, rock mass strengths and support characteristics are
evaluated.

2
RocSupport Introduction Manual

Methods of Support Design

Although there are no clearly defined rules for tunnel support and lining design at the present
time, three general methods have emerged over recent years. These can be described as:
1. Closed form solution methods that are based upon the calculation of the extent of “plastic”
failure in the rock mass surrounding an advancing tunnel, and the support pressures required
to control the extent of the plastic zone and the resulting tunnel deformation.
2. Numerical analysis of the progressive failure of the rock mass surrounding an
advancing tunnel and of the interaction of temporary support and final lining with this
failing rock mass.
3. Empirical methods based upon observations of tunnel deformation and the control of
this deformation by the installation of various support measures.
RocSupport belongs to the first category of solution methods, i.e. “rock support interaction”
or “convergence-confinement” methods.
A good example of a numerical analysis program which belongs to the second category, is
RS2, a finite element stress analysis and support design program for underground
excavations, also available from Rocscience.
Each of these methods has advantages and disadvantages, and the optimum solution for
a given tunnel design, may involve a combination of different methods, at different stages
of the design. For example, a preliminary analysis of temporary support requirements
could be
carried out with RocSupport, and detailed final design, including plastic failure of the rock
mass, and yielding support, can be carried out with RS2.
In spite of the limitations discussed above, rock support interaction analysis has many
attractions, and when used in conjunction with numerical analyses, it can provide valuable
insights into the mechanics of rock support, and reasonable guidelines for the design of this
support.

Rock Support Interaction

A starting point for a discussion of the “rock support interaction” method, is to discuss
the deformation which occurs in the vicinity of an advancing tunnel face, for an
unsupported tunnel. This is illustrated in the following figure.

3
RocSupport Introduction Manual

Figure 1-1: Radial displacements around an advancing tunnel face (not to scale). (Hoek, 1998).

Note that the radial displacement:

• begins a certain distance ahead of the tunnel face (about two and one-half
tunnel diameters)
• reaches about one third of its final value AT the tunnel face
• reaches its maximum value at about four and one-half tunnel diameters behind the face
It is important to note that even for an unsupported tunnel, the tunnel face provides an
“apparent support pressure.” It is this apparent support pressure that provides the stability to
give sufficient stand-up time for the actual support to be installed.

Figure 1-2: Support pressure pi at different positions relative to the advancing tunnel face (not to scale). (Hoek,
1999a)
4
RocSupport Introduction Manual
Observe that the apparent support pressure:
• is equal to the in-situ stress (i.e. pi = po ) at a certain distance (about two and one-half
tunnel diameters) within the rock mass, ahead of the advancing face
• is equal to about one-quarter of the in-situ stress, at the tunnel face
• gradually reduces to zero at a certain distance behind the face.
Note that plastic failure of the rock mass surrounding a tunnel does not necessarily mean
that the tunnel collapses. The failed material can still have considerable strength, and
provided that the thickness of the plastic zone is small compared with the tunnel radius, the
only evidence of failure may be a few fresh cracks and a minor amount of raveling or
spalling.
On the other hand, when a large plastic zone is formed and when large inward
displacements of the tunnel wall occur, the loosening of the failed rock mass can lead to
severe spalling and raveling and eventual collapse of an unsupported tunnel.
The primary function of support is to control the inward displacement of the walls and to
prevent the loosening, which can lead to collapse of the tunnel. The installation of support
(e.g. rockbolts, shotcrete lining, or steel sets) cannot prevent failure of the rock surrounding
a tunnel subjected to significant overstressing, but these support types do play a major role
in controlling tunnel deformation (Hoek et. al. 1995).

Ground Reaction Curve

At the heart of the “rock support interaction” analysis method used in RocSupport, is the
“ground reaction curve” or “characteristic line,” which relates internal support pressure to
tunnel wall convergence. The general derivation of the ground reaction curve, is as follows.
Assume that a circular tunnel of radius ro is subjected to
hydrostatic in-situ stress po and a uniform internal support
pressure pi , as illustrated in the margin figure.
Failure of the rock mass surrounding the tunnel occurs when
the internal pressure provided by the tunnel lining is less than
a critical support pressure pcr .
If the internal support pressure pi is greater than the critical
support pressure pcr , no failure occurs, and the behaviour of
the rock mass surrounding the tunnel is elastic. The inward
radial elastic displacement of the tunnel wall is given by:

ro (1  ) Eqn. 1.1
uie  ( po  pi )
E

5
RocSupport Introduction Manual
When the internal support pressure pi is less than the critical support pressure pcr , failure occurs
and a plastic zone of radius rp is formed around the tunnel. The inward radial plastic displacement
uip is then defined by the ground reaction curve between pi = pcr and pi = 0.
A typical ground reaction curve is shown in Figure 1-3.

Figure 1-3: Ground reaction curve showing relationship between support pressure and tunnel wall convergence
(Hoek et. al. 1995).

This plot shows:

• zero displacement when the support pressure equals the hydrostatic stress (pi = po )
• elastic displacement uie for po > pi > pcr
• plastic displacement uip for pi < pcr
• maximum displacement when the support pressure equals zero
For a given tunnel radius and in-situ stress, the shape of the ground reaction curve depends on
the rock mass failure criterion which is assumed and the specific rock mass characteristics.
The following are dependent on the rock mass failure criterion and characteristics:
• the critical support pressure pcr
• the radius of the plastic zone rp
• the shape of the ground reaction curve in the plastic region (pi < pcr )
See the Solution Methods topic, later in this Introduction, for an overview of the different
solution methods used in RocSupport. These correspond to Mohr-Coulomb or Hoek-Brown
rock mass failure criteria, and have been derived for the rock support interaction problem.

6
RocSupport Introduction Manual
Support Reaction

In order to complete the rock support interaction analysis, the reaction curve for the rock
support must be determined. This is a function of three components:
1. The tunnel wall displacement that has occurred before the support is installed.
2. The stiffness of the support system.
3. The capacity of the support system.
Referring back to Figure 1-1, remember that a certain amount of deformation takes place
ahead of the advancing face of the tunnel. At the face itself, approximately one-third of the
total deformation has taken place, and this cannot be recovered. In addition, there is almost
always a stage of the excavation cycle in which there is a gap between the face and the
closest installed support element. Therefore, further deformation occurs before the support
becomes effective. This total initial displacement will be called uso and is shown in Figure 1-4.

Figure 1-4: Response of support system to tunnel wall displacement, resulting in establishment of equilibrium
(Hoek et.al. 1995).

Once the support has been installed and is in full and effective contact with the rock, the
support starts to deform elastically as shown in Figure 1-4. The maximum elastic
displacement which can be accommodated by the support system is usm and the maximum
support pressure psm is defined by the yield of the support system.
Depending upon the characteristics of the support system, the rock mass surrounding the
tunnel and the in-situ stress level, the support system will deform elastically in response to
the closure of the tunnel, as the face advances away from the point under consideration.

7
RocSupport Introduction Manual
Rock-Support Equilibrium
Equilibrium is achieved if the support reaction curve intersects the rock mass displacement
curve before either of these curves have progressed too far. If the support is installed too late
(i.e. uso is large in Figure 1-4), the rock mass may have already deformed to the extent that
loosening of the failed material is irreversible. On the other hand, if the capacity of the
support is inadequate (i.e. psm is low in Figure 1-4), then yield of the support may occur before
the rock mass deformation curve is intersected. In either of these cases the support system
will be ineffective, since the equilibrium condition, illustrated in Figure 1-4, will not have been
achieved.

Support Characteristics
In RocSupport, the stiffness and capacity of support is expressed in terms of Maximum
Support Pressure and Maximum Support Strain. In this form it is incorporated directly into
the rock support interaction analysis.
Note that since the support capacity is simply modeled as an equivalent internal pressure, the
reinforcement provided by grouted rockbolts or cables cannot be properly accounted for in
this simple model. However, the radius of the plastic zone calculated from the analysis, can
be used as a guide for the length of bolts or cables – i.e. bolts or cables should always be
anchored in unyielded rock.
The stiffness and capacity of support systems such as rockbolts, steel sets, shotcrete and of
combinations of these elements can be estimated from relatively simplistic analyses
published in Hoek and Brown (1980) and summarized in Hoek (1999b). These estimates have
been used for the pre-defined support types available in RocSupport.

Support Installation
The origin of the support reaction curve in Figure 1-4 (i.e. the value of uso), is the tunnel
convergence which has occurred at the point of support installation. In RocSupport, this value
can be specified in two ways:
• directly (as a convergence or wall displacement), or
• indirectly (a distance from the tunnel face is specified, which is then converted to
tunnel convergence using a longitudinal tunnel deformation profile)

In RocSupport you can choose from a pre-defined longitudinal deformation profile (LDP)
function, or create a user-defined LDP function. An example of one of the pre-defined LDP
functions is shown in Figure 1-5

8
RocSupport Introduction Manual

Figure 1-5: Tunnel wall displacement as a function of distance from face (Hoek, 1999a).

Determination of the tunnel wall displacement that has occurred before the support is
installed is not a trivial problem, since it involves a consideration of the three-dimensional
stress distribution, and propagation of failure surrounding the advancing face. The latest
method published by Vlachopolous and Diederichs (2009) is currently the default LDP
function in RocSupport. Other approaches are found in Chern et al (1998), Panet (1995) and
Unlu and Gercek (2003).

Solution Methods

A number of derivations of the “rock support interaction” analysis method have now been
published. All methods assume a circular tunnel in a hydrostatic stress field, and the main
theoretical efforts have been devoted to the calculation of the size of the plastic zone, and
the shape of the ground reaction curve, for different assumptions on how the failure of the
rock mass progresses as the tunnel is advanced.
The main differences between the various methods used to calculate the ground reaction
curve, are in the choice of the rock mass failure criterion, and in whether or not the rock mass
dilates (changes in volume) during failure.
In RocSupport, the following solution methods are available:
• Duncan Fama (1993)
• Carranza-Torres (2004)
• Vrakas and Anagnostou (2014)
• Lee and Pietruszczak (2008)
• Barbosa (2009)
• Vrakas (2016)
9
RocSupport Introduction Manual
All are closed-form solutions, but with slightly different approaches and assumptions. In the
Project Settings dialog, a small description of each solution method is displayed to the right
of the Solution Method section once a specific method has been selected. See the references
for details about each solution method.

Deterministic Analysis
In the Project Settings dialog, the user can choose either Deterministic or Probabilistic analysis
types.
A Deterministic analysis simply means that all input variables are assumed to be
“exactly” known (e.g. in-situ stress and rock strength parameters).
This results in a unique solution for all program output, including:
• the Ground Reaction curve
• Plastic Zone Radius
• Equilibrium pressure (if support is installed)
• Factor of Safety (for support)

Factor of Safety
A unique Factor of Safety for the support is calculated in a Deterministic analysis. The
definition of the Factor of Safety in RocSupport is as follows:

A Factor of Safety GREATER THAN 1 is calculated as shown in Figure 1-6. In this case the Factor
of Safety is simply the ratio of the Maximum Support Pressure psm to the Equilibrium Pressure peq
(the pressure at the intersection point of the Ground Reaction and Support Reaction curves).

Figure 1-6: Definition of Factor of Safety > 1.

10
RocSupport Introduction Manual

A Factor of Safety LESS THAN 1 is calculated as shown in Figure 1-7. This occurs when
the Ground Reaction curve intersects the Support Reaction curve after the elastic limit of
the support has been exceeded. A “projected” equilibrium pressure p’eq is calculated by
projecting the elastic support reaction curve until it intersects the Ground Reaction curve,
and this value is used in the denominator of the Factor of Safety equation.

Figure 1-7: Definition of Factor of Safety < 1

Probabilistic Analysis

In the Project Settings dialog, the user can choose either Deterministic or Probabilistic analysis
types.
A Probabilistic analysis allows the user to input statistical distributions for:
• tunnel radius
• in-situ stress
• all rock mass parameters
Using either Monte Carlo or Latin Hypercube sampling, the program will then sample the input
distributions and run the analysis for the specified Number of Samples defined by the user in
the Project Settings dialog.
The user can then view statistical distributions of all output variables (e.g. plastic zone radius,
wall displacement), rather than simply a single number as calculated from a Deterministic
analysis.

11
RocSupport Introduction Manual

Probability of Failure
A Probabilistic analysis results in a distribution of Safety Factor, rather than a single
value. From a Safety Factor distribution, a Probability of Failure can be calculated.
The Probability of Failure in RocSupport is simply the number of analyses with Safety
Factor less than 1, divided by the total number of analyses generated by the Probabilistic
analysis.
For example, if 100 out of 1000 samples in a Probabilistic analysis resulted in a Factor of
Safety less than 1, then the Probability of Failure would be 10 %.

Figure 1-8: Definition of Probability of Failure.

Mathematically speaking, the Probability of Failure is the area under the Factor of Safety
probability distribution to the LEFT of Factor of Safety = 1 (i.e. the black area in Figure 1-
8), divided by the total area under the curve.

12
RocSupport Introduction Manual

References
Barbosa, R.E. (2009). Analytical solution for a deep tunnel excavated in a porous elasto-
plastic material considering the effects of seepage forces. Proceedings of the 3rd CANUS
Rock Mechanics Symposium, Toronto.
Carranza-Torres, C. (2004). Elasto-plastic solution of tunnel problems using the generalized
form of the Hoek–Brown failure criterion. International Journal of Rock Mechanics and
Mining Sciences, Proceedings of the ISRM SINOROCK 2004 Symposium, edited by J.A. Hudson
and Xia- Ting Feng, Volume 41, Issue 3.
Chern, J.C.. Shiao, F.Y. and Yu, C.W. (1998). An empirical safety criterion for tunnel construction.
Proc. Regional Symp. on Sedimentary Rock Engineering, Taipei, Taiwan, Nov 20-22, pp 222-227.
Duncan Fama M. E. (1993). Numerical Modeling of Yield Zones in Weak Rock. In Comprehensive
Rock Engineering, (ed. J.A. Hudson) 2. Oxford: Pergamon, pp. 49-75.
Hoek, E. (1998). Tunnel support in weak rock. Proc. Regional Symp. on Sedimentary Rock
Engineering, Taipei, Taiwan, Nov 20-22, pp 1-12.
Hoek, E. (1999a). Personal Communication.
Hoek, E. (1999b). Support for very weak rock associated with faults and shear zones. Proc.
International Symposium on Rock Support and Reinforcement Practice in Mining, Kalgoorlie,
Australia, 14-19 March.
Hoek, E. and Brown, E.T. (1980). Underground Excavations in Rock. Instn. Min. Metall., London.
Hoek, E., Kaiser, P.K. and Bawden, W.F. (1995). Support of underground excavations in hard rock.
Balkema, Rotterdam.
Lee, Y-K. and Pietruszczak, S. (2014). A new numerical procedure for elasto-plastic analysis
of a circular opening excavated in a strain-softening rock mass. Tunnelling and Underground
Space Technology, Vol. 23, No. 5, 588-599.
Panet, M. (1995). Calcul des Tunnels par la Me´thode de Convergence–Confinement. Presses
de l’Ecole Nationale des Ponts et Chausse´es. Paris. 178p.
Unlu, T. and Gercek, H. (2003). Effect of Poisson’s ratio on the normalized radial
displacements occurring around the face of a circular tunnel. Tunnelling and Underground
Space Technology. 18. 547–553
Vlachopoulos, N. and Diederichs, M.S. (2009). Improved longitudinal displacement profiles
for convergence-confinement analysis of deep tunnels. Rock Mechanics and Rock
Engineering (Accepted - In Press) 16 pgs.
Vrakas A. (2016). A finite strain solution for the elastoplastic ground response curve in
tunnelling: rocks with non-linear failure envelopes. International Journal for Numerical and
Analytical Methods in Geomechanics, Vol. 41(7), 1077-1090.
Vrakas A. and Anagnostou G. (2014) A finite strain closed-form solution for the elastoplastic
ground response curve in tunnelling. International Journal for Numerical and Analytical Methods
in Geomechanics, Vol. 38, No. 11, 1131–1148.

13