BedrockLauri Uotinen

The Big Picture

The Big Picture

6.4.2017

BEDROCK Control Group 3

KARMO Mikael Rinne, Aalto

Mechanical properties of rock joints

ROSA Eeva-Liisa Laine, GTKFracture simulator, which respects the measured fracture length and orientation distributions

KARMO

Outline

1. Problem

2. Continuum

3. Method

4. Results

5. Publications

6. Status

7. Cooperation

6.4.2017

Problem

The research problem is obtaining the mechanical parameters for

over 10 m long rock joints. Natural scale experiments are expensive

and unrepresentative. Laboratory experiments are affordable and

quantitative, but the scale effect is a factor to be taken in account.

The goal is to produce a method to obtain

the mechanical parameters of rock joints.

6.4.2017

Continuum

6.4.2017

KARMO I

KARMOII

KARMOIII

KYT2014

2011 2012 2013 2014 2015 2016 2017 2018

Pilot

Photogrammetry

Modelling

Body Scanning

Mine Wall Scanning

SAAS Scanning

Invention disclosures:

KYT2018

KARMO I

KARMOII

KARMOIII

Shear

Str

ength

Joint length

Johansson & Stille 2015

1-2 m100-200 mm 10-20 m

Rock joint length:Continuum

Method

6.4.2017

Results

- 5 theses (2 MSc, 3 BSc), 3 conference papers, 1 working report

- 2 laboratory experiments: 2 m x 1 m & 0.50 m x 0.25 m

- The developed photogrammetric method reaches point

spacing accuracy of 0.150 … 0.250 mm, which is sufficient

- 3 invention disclosures, 1 TUTL application

6.4.2017

Publications

1. Rakogeometrian jäljentäminen 3D-teknologialla

2. Quantifying loss of geometrical features in downscaling of rock joint surfaces using shear box

replica series

3. A method to downscale joint surface roughness and to create replica series using 3D printed

molds

4. Rakopinnan karkeuden mittaaminen fotogrammetrisesti

5. Itsetiivistyvä betoninen jäljennysmateriaali rakopinnoille

6. Requirements for initial data in photogrammetric recording of rock joint surfaces

7. Determination of joint mechanical parameters for stability analysis in a low stress open pit

mines

8. Tilt testing experiment to validate the photogrammetrically predicted friction angle

9. Determination of joint mechanical parameters for stability analysis in low stress open pit mines

10. Photogrammetric calculation of JRC for rock slope support design

6.4.2017

Status

KARMO III 2017-2018 goal is to verify the photogrammetric

method using numerical modelling and shear test experiments.

Sivi Kivivirta is doing 2D numerical modelling in her BSc thesis

Retest of 2 m x 1 m shear-push configuration in autumn.

KTH (Fredrik Johansson and Francisco Ríos Bayona) will visit

Aalto on 21st of April to discuss numerical modelling of rock joints.

6.4.2017

Cooperation

KTH, Sweden Geomechanica, Canada ROSA, Finland

200 mm x 200 mm Irazu FEM-DEM

6.4.2017

ROSA

6.4.2017

ROSA• KYT-puoliväliseminaari 7.4.2017

Contents (ROSA)

• Motivation

• The aim of the project

• Methods and workflows

• WP1

• WP2

• Results

Motivation

• Building a freeware, ROSA code, for statistical fracture simulationwith a possibility to add own advanced spatiostatistically simulatedfracture properties – or linked with any codes or scripts which offermore geologically realistic fracture property distribution –especially fracture dimensions (length)

• ROSA will have the link with a scientific statistical R code and Irazu FEMDEM – with a possibility to physically simulate possiblefracture networks typical for the Finnish bedrock – giving an alternative for validation of fracture networks

• By using C- and other codes for Linux workstations, we’ll getcodes which can calculated faster using parallel programming bye.g. CSC computing environment

The aim of the ROSA project

• WP1: Statistical fracture simulation code ROSA with a special emphasis on the fracture length distributions

• WP2: Statistical, geological and physical background for fracture network modelling with a special emphasis on the fracture data and the study of validation possibilities of the final fracture network.

• WP3: A suggested Master Thesis about fracture network modelling. (Helsinki, Turku or Aalto university)

• WP4: 3D and 4D visualizations

ROSA code

Ouput: Fracturenetworkmodel

Finite elementmesh to Irazu

FEMDEM, UDEC, 3Ddec,

Groundwatermodelling software,

thermal energyevaluation, …

FEMDEM simulating

fracturing usingpalaeostresses

Spatial and advanced

statistical toolslinked with

ROSA code(e.g. Julia)

Multipointstatistical

tools linkedwith ROSA

code

3D modellingtools to use this

fracturing to analyze statistical

distributions –possible spatial

and propertydistributions

Fracturedata &

statisticalmodels as

input

Workflow

3D AND 4D VISUALIZATIONS (WP4)

Paraview

VALIDATION (WP2)

Fracture orientations Palaeostresses Simulated fractures Statistical distributions/locations

FRACTURE NETWORK REALIZATIONS

C program for building fracture network realizations (WP1) Linking WP2 results and MPS code into the C program

FRACTURE PROPERTIES (orientation, surface, length, filling, ….)

R script for statistical distributions (WP1) Marked point processes, theoretical property models (WP2)

FRACTURE LOCATIONS

Random locations using C (WP1) Physical, mathematical or geostatistical models (WP2, Julia)

CLUSTERING

Matlab script (Markovaara-Koivisto and Laine 2012) Julia script (WP2)

DATA (WP1 and WP2)Fracture data (GTK

database)Fracture network (photo orgeological interpretation)

Line/area fracture data Lidar dataDrone and

stereophotography

2015-2016

2017

2017

2016

2018

2018 +

2018

DATA

• Kopparnäs drill core, drone phographs and detailed elevation models

• Palmottu drill core data, detailed elevation model, outcrop observations

• Dense data sets from future nuclear waste sites

• Detailed fracture length data from Helsinki area

• In addition for testing scripts and codes data from Mäntsälä and Uunimäki

DTM made from drone images from Kopparnäs with the high-resolution images draped onto the model.

Fractures were mapped with scanline technique and plotted with Structural Suite by Surpac. Red colour of

the surfaces indicate rough fracture surfaces, yellow slightly rough and green smooth.

WP1 ROSA: Fracture simulator which respects the

measured fracture length and orientation distributions• Test data used in coding was acquired from a dimension stone quarry in Mäntsälä. Scanline

mapping was carried out in 2007 and 2009 for Mira’s Doctoral Thesis: (KYT-Kalliolaatu)

• Rock fracture data analysis

• Number of fractures in the modelling volume (R)

• In defining fracture density

• Squared fracture lengths (normal, exponential, log-normal and gamma)

• Angle between the normal of fracture surface and mean orientation of thefracture set (normal, exponential, log-normal and gamma)

• Fitting theoretical probability distributions to the fracture data (R)

• Fracture length per fracture sets

• Normal, exponential, log-normal and gamma

• Fracture orientations

• Truncated normal distribution (0-90°) for dip

• von Mises for dip direction

• Percentages for surface properties Ja ja Jr

• Discrete Fracture Network simulation – possibility for high performance computing

• Random location of the fractures (C)

WP2 Statistical, geological and physical background for fracture network modelling and the study of validation possibilities of the final fracture network

• Extracting planar surfaces from Lidar data: possiblelarger fracture surfaces

• Julia scripts for analysing observed fractureproperties

• The use of Multipoint statististics (MPS) in fracturenetwork modelling

• Physical, mathematical and geostatistical tools to predict fracture locations and properties

• Application of marked point processes and theoretical property models

• Linking these with C program for fracture simulation

• Validating the fracture property distributions usingphysical simulation of fractures (Irazu FEMDEM)

S & SE dipping structures

10 m

Lidar data cpompared with the fracture measurements

ResultsPublications:

• Laine, E. and Markovaara-Koivisto M, 2015. 3D visualization and analysis of fracturing in the Precambrian bedrock characterized by complicated structures, two case studies from southern Finland. (laajennettu abstrakti 10 s.) IAMG2015 , Freiberg, 5th-13th September 2015.

• Laine, E and Markovaara-Koivisto M, 2016. New 3D modeling approaches in the study of Palmottufracture patterns. Bulletin of The Geological Society of Finland, Special Volume. Abstracts of the 32nd

Nordic Geological Winter Meeting 13th-15th January 2016, Helsinki, Finland. pp. 94.

• Markovaara-Koivisto, M., Laine, E. ja Rasilainen, K., 2016. Kallioperän rikkonaisuuden mallinnus Suomessa. ATS Ydintekniikka, Vol. 45. No 1, sivut 12-14.

• Laine, E. ja Markovaara-Koivisto M., 2016. The 3D visualization and analysis of Laser scanning data together with geological and geophysical data applied for fracture network modeling – examples studies from southern and eastern Finland. The 2nd Virtual Geoscience Conference (VGC 2016). Abstrakti ja posteri.

• Laine, E. and Markovaara-Koivisto M, 2016 Rakoverkkomallinuksen (DFN) lähtötiedot. Louhinta- ja kalliomekaniikan päivät 13.-14.10., Hotelli Haaga, Helsinki, poster presentation.

• Laine E., 2017 Rakoverkkomallinnus, menetelmät ja ohjelmistot. Geological Survey of Finland, Workreport 17/2017 (31.3.2017)

Scripts:

• R script for statistical analysis of fracture property distributions (Markovaara-Koivisto)