Soil Nail Analysis Program (SNAP) Value Engineering (VE) Design Methodology
SNAP can obtain optimal to near-optimal soil nail wall designs in minutes. The program can adjust:
Wall
Height and batter angle
Nail
length
inclination angles
fan angles
horizontal and vertical spacing
cantilever distance
bar size
Drillhole diameter
Backslope cuts and downslope cuts
The user specifies valid ranges and resolutions for each parameter, as well as minimum factors of safety for seismic loading and static loading at
each staging cut.
The VE design minimizes total cost while meeting all global stability safety factors. VE designs are obtained using the scientific method. The program:
Collects observations
Postulates causal relationships and proposes theories
Generates a hypothesis
Predicts results
Runs an experiment to acquire measurements
Compares predictions to the measurements
Re-evaluates theories, if necessary
Generates a new hypothesis
This process continues until proposed theories accurately predict any given design.
An artificial intelligence technique incorporating a specialized neural network is used to study observations, analyze parameters, propose theories,
generate hypotheses, and predict results. The neural network simulates both the structure and function of the brain. It is trained on a set of
observations (simulated wall designs) to predict results of future experiments (simulated wall designs) by analyzing causal relationships and
interactions among parameters.
The neural network goes through a process of logical reasoning and thought to make an educated guess, or hypothesis, of the optimal wall design.
The neural network predicts the price and minimum factor of safety of this wall design. The program then calculates a cost and a factor of safety for this
wall design. If the neural network prediction is incorrect, the neural network re-analyzes parameter relationships to generate a new hypothesis for the
best wall design. This process repeats until the neural network prediction of the hypothesis is correct.
The neural network then goes through a sleep stage, where the synapses connections release energy from the neurons. The neural network is then presented with a new random wall design, which the refreshed neural network uses in an attempt to think of a better wall design from a different perspective. The process concludes after no better design is discovered after several thought and sleep cycles.
This process is far more efficient than alternative minimization search algorithms, such as conjugate gradient, simulated annealing, or Monte Carlo techniques. Conjugate gradient descends through valleys in the search space until reaching a minimum, requiring numerous function evaluations,
and must be repeated for numerous random locations in the search space to increase the probability of finding a good solution. Simulated annealing
and other Monte Carlo techniques also require numerous function evaluations, but do not need to be repeated. Simulated annealing was implemented
for soil nail wall design as a comparison. Simulated annealing was able to find a good solution half the time after 12,000 wall design evaluations,
requiring about 10 hours. The neural network was able to find a good solution every time after 120 wall design evaluations, in about 10 minutes.
For example, a recent wall design from an existing project at a total cost of $35.4 M with a minimum factor of safety of 1.35 was analyzed by SNAP,
varying 9 parameters. SNAP generated 127 wall designs out of a possible 672 billion to obtain a VE design at a total cost of $21.0 M with a minimum
factor of safety of 1.35, in 34 minutes. This was a cost savings of $14.4 M, or 41%, at the same factor of safety. The time required for the neural network training, logical analysis, and sleep cycles was 15 minutes. The time required to exhaustively search all 672 billion possible designs would be 345,000 years.
This artificial intelligence technique has a wide range of applications beyond the design of soil nail walls. It has been successfully used for:
First arrival picking for geophysical surveys (patent, 1990)
Program implemented on IBM 3090, replacing work of 80 PhD's
Trained by observing human experts
Terrain Topology
Lidar point cloud processing
Used in RockVision-3D (ground imaging), TRT (tunneling), CRSP-3D (rockfall), and SNAP-3 (slope stability)
Well log interpolation
Used in RockVision-3D and TRT
Electric warfare
Implemented in optical components
Decreased equipment weight from 2000 lbs to 200 lbs
Correctly identified 1 million signals with no errors or false positives
Significant improvement, allowing location of sources before detection
Installed on F-16
Vision systems for robotic welding
Installed on MTS systems and robotics for welding 2 foot thick steel, in air and under water
Worked in 90% image corruption from smoke and spatter, outperforming human operators by 50%
Weld parameter modeling
Used to determine mil-specs for US Navy and welding specs for Chicago Bridge and Iron
Determined optimal current, voltage, and weld speed
Predicted bead penetration, cap, strength, quality, and slag removal
Used in robotic welding applications
Concrete curing models in drilled shafts
Used to construct empirical models of heat of hydration, expansion, strength, stiffness, and heat transfer within reinforced concrete
Submitted as FHWA publication: Velocity Variations in Cross-hole Sonic Logging Surveys
Concrete curing models in drilled shafts
Automated mining vehicle operation
Controlled speed and steering based on ultrasonic range sensors at corners of vehicle and high level commands for direction
Trained by observing human operator
Could turn vehicle around in tight conditions
Credit card customer service scheduling
Predicted number of customer calls on any given day using previous 3 years of data, using a laptop in 20 minutes (90% accuracy)
Out-performed a US national lab, non-linear mathematics division, using 25 Sun workstations running 6 months (85% accuracy)
Weather prediction
Matched performance of meteorologists for prediction of temperature and precipitation given previous weather data
Horse race prediction
Predicted race time of a horse based on previous history, track conditions, gate, etc.
Writing music
Determined guitar chords from melody
However, the value of applying this technique to design of soil nail walls alone could have a significant positive impact on engineering,
construction, and transportation industries.
Our Specialties:
Commercial
Slope stabilization (SNAP, SNAP-2, SNAP-2 Premium)
Rockfall mitigation (CRSP-5, CRSP-3D)
3D Numerical model development
Distinct Element Method
Large and small deformation
Solids, fluids, gases
Finite Difference
Ground water seepage
Multi-dimensional apex parameterization
Multi-dimensional minimization
Artificial Intelligence
Neural-Networks
Geophysics
Full waveform tomographic and holographic inversion
Joint seismic, seismoelectric, and electrical methods
Geotechnical engineering
Slope stability and reinforcement design
Rockfall Mitigation