Research

Itasca has participated in numerous research efforts that have led to new knowledge and techniques that have become standards in the geotechnical industries. These include: a geomechanical risk-based cost analysis procedure for optimizing mining methods in deep hard rock mines; new instrumentation and detection methods for methane control in coal mines; micromechanical numerical imulation of fracture development around underground excavations to increase understanding of rock fracture mechanics; a study of caving sponsored by an international mining consortium — a project intended to find new methods for prediction of rock mass caving and from these, to identify optimum layouts and production in block and panel cave mines; development of numerical methods for prediction of rock fracture around tunnels as well as hydraulically fractured wellbores; development of methods for assessment of risk of fault-slip seismicity; field testing of methods for reducing seismicity via high-pressure water injection; and simulation of creep and fracture in viscous rocks.

Itasca accepts the challenge of undertaking research, recognizing that the collective knowledge gained across the world-wide span of our offices is an asset we can use for the benefit of our clients. The ability of our engineers to compare and exchange information about the different circumstances that they have encountered creates a climate of innovative thinking and development.

Extradisciplinary Work

In early 1997, Itasca was contacted by a farm implement manufacturing company seeking assistance on a possible application of our particle flow codes (PFC) to simulate grain harvesting. Design of farm implements is not, by any standard, a part of the geomechanics-based fields where Itasca does its work. We could have declined, but what we said was "Yes." Such inquiries are no longer rare.

Itasca offers two special qualities to prospective clients with unusual problems. We have a broad, international pool of engineering talent covering a wide range of practical experience. Also, as developers of the codes that we use, we know better than anyone what they can do — and what they cannot do. This special expertise is in demand both in geotechnical and in other engineering fields.

Some of these special projects include: an effort to preserve 40 Stone Buddhas in Japan that were at risk of irreparable damage from seismic activity; study of the flow and packing of granular materials in silos and bins; packing and sintering of powders in manufacture of products such as pills and semiconductors; and, fracturing of reinforced concrete beams and columns subjected to earthquake and impact loading. Although Itasca does not actively seek such projects, we welcome the challenges that they present and the opportunity to verify the applicability of our codes and the ingenuity of our engineers. We welcome your enquiries.

ASC (Applied Seismology Consultants) and Itasca Consulting Group, Inc. (the Minneapolis office of Itasca) are continuing a long tradition of collaboration as joint research partners in the Mass Mining Technology (MMT) Project, an industrially-funded project supporting fundamental research into the mechanics of caving, blasting and flow in underground mass mining. The strength of collaboration between ASC and Itasca lies in the ability to study fundamental modes of rock fracture, using the Particle Flow Code (PFC) to create and test "synthetic rock" and comparing the predicted spatial and temporal trends in fracturing directly with microseismic data obtained in the laboratory or field. This represents a powerful combination of tools that can be applied to a wide range of rock mechanics problems. The technique has been successfully employed within pioneering research projects to reproduce in situ seismicity measured in massive brittle rock (Hazzard and Young, 2004) and has been applied to the analysis of damage around underground excavations (Young et al., 2004).

The primary objectives of the MMT research are to extend the technique to the prediction of jointed rock mass behavior in three dimensions, to use this in the development of improved tools for prediction of cave growth, fragmentation, and subsidence and to extract more value from microseismic data in caving prediction and monitoring. At the core of the new methodology is the construction and testing of "Synthetic Rock Mass" (SRM) samples for predicting rock mass behavior. SRM samples are three-dimensional and simulate rock as an assembly of bonded spheres (intact rock) with an embedded discrete network of discshaped flaws (joints).

The Lift 2 block cave at Rio Tinto's E26 Mine in Southeast Australia has been employed as a case study to test the potential for application of SRM technology and advanced microseismic analysis to caving prediction. ASC employed a number of novel microseismic data techniques that enhance the information currently retrieved from microseismic catalogues to develop an understanding of the fracturing and yield that accompanied undercutting and caving at a case study block cave. Through a series of analyses, cave-induced microseismicity showed its potential to provide further insight into the fracture network (interpreting existing seismic parameters in terms of fracture clustering, preferred orientation, size and spacing) and its correlation with different factors present in the production environment.

Itasca constructed a number of SRM samples for the various geomechanical domains at the mine and subjected them to representative cave-induced stresses. The tests provide a significant volume of information including fracture orientations, rock mass modulus, strength, brittleness and fragmentation. Comparison of the predicted and in situ fracture modes and orientations derived from the microseismic catalogue indicated very similar trends and represented a significant validation of the SRM approach.

ASC and Itasca thank the sponsors of the MMT project for the opportunity to conduct this fundamental research. We see significant potential for the application of Synthetic Rock Mass testing and linked microseismic analyses to other mine design problems and fields of engineering and look forward to future opportunities for collaboration in consulting, design and research.

The new material generation procedures available in the FishTank for PFC2D and PFC3D 4.0 have been used to analyze the behavior of the wall slopes of a mine as it transitions from open pit to underground cave mining. The PFC2D system is a rectangular region which has been excavated that encompasses the region from the surface to over 250m below the final pit bottom. It consists of bonded clumped material and contains multiple material zones representing each of four rock lithologies. In order to model such a large model and maintain high-definition in the area of interest, grain-size refinement regions were used, which is akin to mesh refining used in continuum modeling. The enhanced material genesis procedures used are illustrated below.

After the model is generated, the pre-mining in-situ stress field observed at the site is added using the stress-initialization procedure assuming an elastic behavior. A specified isotropic stress is obtained through an iterative approach whereby a set of boundary (and possibly also interior) particles is moved, then the boundary particles are fixed, the interior particles are freed and static-equilibrium conditions are allowed to develop. During each stage, the applied particle displacements are computed from the strain increment that is related by linear elasticity to the stress increment needed to reach the target stress. The stages continue until the all target-stress components are within a given tolerance of the target stress. If the grain shape is circular or spherical, then floating particles are reduced. The final specimen is produced by adding contact and/or parallel bonds (if desired). Once this is achieved, the model is permitted to behave non-linearly (i.e. damage/bond cracking and boundary forces are allowed to respond naturally) and the pit is excavated in a series of stages. Once the excavation sequence is finished the system is prepared for underground cave draw-point excavation.

This model was used to estimate the stability of the pit slope due to underground mining and the degree of dilution that could be expected along two mining horizons.

The design of open pits normally involves several scales of analysis where different acceptability criteria apply. The smallest scale is the bench scale, where the parameters defining the geometry are the bench height, the bench face angle and the berm width. These parameters also provide a first attempt at establishing the interramp angle, which is in turn assessed with different analysis tools at the interramp and overall scales.

Currently, there is no unique methodology to perform the bench/berm design process and there are several approaches proposed by different practitioners. The methodology that Itasca uses is based on the statistical analysis of the interaction between the structural sets identified in the wall to be designed and the application of a catchment criterion to decide the appropriate bench geometry.

While this type of analysis can be performed using commercially available wedge and planar failure analysis packages, encouragement from the mining industry to improve the efficiency of calculations, led Itasca S.A. to develop a procedure that automatizes the calculation process, in order to make it faster and more reliable. As a result, the codes SWISA (for wedge failures) and PFISA (for planar failures) have been created.

The codes are a series of Excel macros which operate jointly with the @Risk add-in to perform a statistical analysis of the volumes of unstable wedges formed by the joint sets in a slope wall. The basic input data (see Figure 1) is the statistical information (average, standard deviation, maximum and minimum values) of:

• Wall orientation
• Joint sets (dip and dip direction)
• Bench face angle
• Bench height
• Cohesion and friction angle for the joints
• Unit weight of the rock

Figure 1. One of the parameter input spreadsheets for SWISA

The main characteristics of the analysis are:

• Identification of unstable wedges
• Construction of cumulative distribution curves for the unstable volume and other geometrical parameters (Figure 2)
• Calculation of berm widths based on a comparison between the spill length of the critical volume and the catchment criterion

Figure 2. Curve showing the spill length determined from the cumulative volume distribution for an analysis with SWISA

Currently, both SWISA and PFISA are run as Excel macros and are only used internally by the Itasca staff in open pit and rock cut projects. Eventually the two tools will be made commercially available to the geotechnical community. Some developments have been conceived to improve the data pre and post processing through a graphical interface, but they are still under development (see Figures 3 and 4).

Figure 3. Initial graphical interface to input data into SWISA

Figure 4. Initial graphical interface to post-process SWISA output

The behavior of a jointed rock mass is strongly governed by the detailed micromechanics of joint slip and new fracture growth. However, consideration of all joints on the scale of an engineering problem is often prohibitive. Itasca Consulting Group has developed a methodology called Synthetic Rock Mass Modeling (SRM) that allows for detailed consideration of the rock mass joint fabric on the scale of 10-100m. The approach can be used to derive rock mass properties such as modulus, strength and brittleness for later use in larger-scale continuum or discrete-element models.

A Synthetic Rock Mass Model (SRM) – Itasca's PFC2D/PFC3D is used to create an assembly of bonded particles representing a large intact rock sample. A Discrete Fracture Network is then generated which honors the joint measures derived from drilling and mapping on site (e.g. spacing, trace length, orientation). The entire network is then embedded within the bonded assembly, which is subjected to stress changes expected in the field. Joints are inserted into the PFC particle assembly using a newly developed smooth joint Model which allows slip and opening on internal planar surfaces.

Three increasingly detailed views of a 2-D SRM made of 330,000 particles and 40,225 joints, resulting in 38,656 clusters.

Validation of the SRM Approach – Validation of the approach has been achieved through back analysis of caving at a case study mine. Applied Seismology Consultants (ASC), an Itasca technology partner, performed spatial analyses on bond breakages from the SRM sample and microseismic events from the mine and showed that the inferred new fracture orientations are consistent. Predictions of the size distribution of the fragments that are generated were also consistent with mine observations. When SRM-derived material properties are employed in a larger-scale FLAC3D model, the rock volumes exhibit microseismicity, yield and flow that are in good agreement with in situ observations.

A slice through a FLAC3D model of block cave mining employing SRM-derived properties. The predicted extent of microseismicity (within the dark gray isosurface) compares favorably with the location of recorded events (black dots). The predicted extent of flow (light gray isosurface) and yield (dark blue) are also shown.

Increasing fragmentation with increasing applied strain on a planar slice through a spherical (3-D) SRM sample.

References

Pierce, M., Mas, D., Cundall, P. and Potyondy, D. (2007) "A Synthetic Rock Mass Model for Jointed Rock," in Rock Mechanics: Meeting Society's Challenges and Demands (1st Canada-U.S. Rock Mechanics Symposium, Vancouver, May 2007), Vol. 1: Fundamentals, New Technologies & New Ideas, pp. 341-349, E. Eberhardt et al., Ed. London: Taylor & Francis Group.

Reyes-Montes, J. M., Pettitt W. S. and Young R.P. (2007) "Validation of a Synthetic Rock Mass Model Using Excavation Induced Microseismicity," in Rock Mechanics: Meeting Society's Challenges and Demands (1st Canada-U.S. Rock Mechanics Symposium, Vancouver, May 2007), Vol. 1: Fundamentals, New Technologies & New Ideas, pp. 341-349, E. Eberhardt et al., Ed. London: Taylor & Francis Group.