A sneak peak into the future of geomechanics
Krishna Kumar, krishnak@utexas.edu
4th March 2021
Civil Engineering: From the past to the present
Tacoma narrow bridge collapse
Millenium bridge
London Bridge Station bridge
Victoria station upgrade
Structural monitoring
Post Office tunnel
Distributed Brillouin Sensing
Monitoring of piles
Monitoring of tunnels
Monitoring of arches
Monitoring of shaft
Post Office tunnel
Agent Based Modelling of cities
Soga et al., (2017)
Agent Based Modelling of Tokyo
Zhao et al., (2019)
Network analysis of Herpes Simplex Virus (HSV2)
Paranthropus boisei is the intermediary species that gave humans herpes!
(Underdown et al., 2017)
(Underdown et al., 2017)
The usual suspects
Cellular Automata modelling road network degradation
Cellular Automata modelling of disease
Behaviour of soil under different conditions
Everyone is a geotechnical engineer!
Where to build the perfect sand castle?
Building the perfect sand castle
Can we build an underwater sand castle?
Using hydrophobic sand: The force between beads remain constant, but the effective weight of the sandcastle is reduced by a factor of 3.
LBM modeling of unsaturated soil behavior
LBM modeling of unsaturated soil behavior
What happens when you squeeze the balloons?
What happens when you squeeze: soil vs water?
Shear-induced dilation
Saw-tooth model
Visualising effective stresses in soil
https://www.youtube.com/watch?v=Z6rvdkI4T8I
Realistic dilation model model
Granular trip
https://www.youtube.com/watch?v=sjjq5z0qLlI
Why do dogs dig shallow pits?
Scales in modelling soil
Discrete Element Method
Grain-scale simulation of soil
Applications of DEM
Lattice Boltzmann - MRT
\[f_{i}(x + dx, t +\Delta t) - f_{i}(x, t) = -S_{\alpha i}(
f_{i}(x, t) - f_{i} ^ {eq}(x, t))\]
- $S_{\alpha i}$ is the collisional matrix.
- Probability density of finding a particle : $f(x,\varepsilon, t) $, where, x is position, $\varepsilon$ is velocity, and t is time.
LBM-DEM fluid-solid coupling
$$\Delta t_{s}=\frac{\Delta t}{\mathit{n}_{s}} \qquad (\mathit{n}_{s}=[\Delta t/ \Delta t_{D}]+1) $$
- At every fluid iteration, $\mathit{n}_{s}$ sub-steps of DEM iterations are performed using the time step $\Delta t_{s}$.
- The hydrodynamic force is unchanged during the sub-cycling.
Submarine run-out
Credit: Amanda Murphy (2016)
LBM - DEM a = 0.8 & 10,000 particles
- LBM Nodes = 50 Million : DEM grains = 10000 discs
- Run-time = 4 hours
- Speedup = 125x on a Pascal P100
Collapse on an inclined plane
aspect ratio 'a' of 6 on a slope of 5*
Loose v dense: Initiation phase
Pore-pressure distribution along the failure plane during initiation.
Loose v dense: Runout phase
Water entrainment front (~15d length) at a slope of 5*
Fluid structure interaction
Oso landslide (2014)
Oso landslide (2014)
Think hard, and go watch Disney movies!
Disney's Frozen: Snow simulation
Mesh-based vs Mesh-free techniques
Material Point Method
Oso landslide simulation
MPM soil-water interaction
MPM submarine landslide
Depth-averaged Material Point Method (Taka et al., 2012)
MPM rainfall induced landslide
Underground worldview
Underground worldview: Virtual Reality
Multichannel Analysis of Surface Waves (MASW)
Machine Learning with Convolution Neural Network
Machine Learning with CNN
Machine Learning with CNN
Machine Learning with CNN
Machine Learning with CNN
CNN Full Waveform Inversion: Architecture
CNN Full Waveform Inversion: Training
CNN Full Waveform Inversion
Learning to Simulate
Learning to Simulate
Modelling soil is out of this world!
Behaviour of martian soil
Modelling soil is out of this world!
CalTech
Building the cities of the future!
