Applied Sciences,
Journal Year:
2023,
Volume and Issue:
13(4), P. 2612 - 2612
Published: Feb. 17, 2023
The
brittleness
of
rock
is
known
to
be
an
important
property
that
affects
the
fragmentation
characteristics
in
mechanized
cutting.
As
interaction
between
cutting
tool
and
(i.e.,
cutter
forces,
efficiency,
s/p
ratio,
abrasivity)
during
mechanical
strongly
influenced
by
fragmentation,
tools
disc
pick
cutter)
experience
different
behaviors
depending
on
brittleness.
In
this
study,
relationships
abrasivity
rock,
efficiency
a
Tunnel
Boring
Machine
(TBM)
were
investigated
for
Korean
types.
was
calculated
mathematical
relations
uniaxial
compressive
Brazilian
tensile
strengths
rock.
evaluated
forces
specific
energy
from
linear
machine
(LCM)
test
Cerchar
index
(CAI)
test,
respectively.
results
show
significantly
correlated
with
CAI
values.
Consequently,
some
prediction
models
energy,
proposed
as
functions
Tunnelling and Underground Space Technology,
Journal Year:
2024,
Volume and Issue:
148, P. 105745 - 105745
Published: April 10, 2024
Monitoring
the
wear
status
of
cutters
is
important
for
safe
and
sustainable
shield
construction
cost
management.
In
this
paper,
an
innovative
stratal
slicing
method
proposed
to
convert
segmented
discrete
uniaxial
compressive
strength
(UCS)
test
data
into
a
sequential
dataset
by
combining
it
with
geological
profile.
The
not
only
accurately
represents
changing
strata
conditions
but
also
differentiates
working
disc
in
various
cutterhead
areas
on
excavation
face.
Its
sequence
characteristics
can
be
better
combined
operational
parameters
time-series
models
real-time
prediction.
Furthermore,
particle
swarm
optimization
(PSO)
algorithm
was
improved
adding
variable
inertia
weights
elimination
mechanisms,
which
effectively
optimised
hyperparameters
long
short-term
memory
(LSTM)
model.
applied
field
tunnelling
case
collected
from
Guangzhou
Metro
Line
18
railway.
results
show
that
UCS
obtained
using
improve
prediction
accuracy
compared
traditional
methods
models.
particular,
IPSO
+
LSTM
horizontal
summation
obtain
most
accurate
has
capability.
With
method,
modelling
approach
generally
applicable
more
complex
ground
larger
diameters.
Sensors,
Journal Year:
2025,
Volume and Issue:
25(6), P. 1650 - 1650
Published: March 7, 2025
Disc
cutters
are
essential
for
shield
tunnel
construction,
and
monitoring
their
wear
is
vital
safety
efficiency.
Due
to
position
in
the
soil
silo,
it
more
challenging
observe
of
disc
directly,
making
accurate
efficient
detection
a
technical
challenge.
However,
existing
methods
that
treat
problem
as
classification
task
often
overlook
issue
data
imbalance.
To
solve
these
problems,
this
paper
proposes
an
end-to-end
method
cutter
state
called
Multivariate
Selective
Attention
Prototype
Network
(MVSAPNet).
The
introduces
attention
prototype
network
variable
selection,
which
selects
important
features
from
many
input
parameters
using
specialized
selection
network.
address
imbalance
data,
used
learn
centers
normal
classes,
achieved
by
detecting
high-dimensional
comparing
distances
class
centers.
performs
better
on
collected
Ma
Wan
Cross-Sea
Tunnel
project
Shenzhen,
China,
with
accuracy
0.9187
F1
score
0.8978,
yielding
higher
values
than
experimental
results
other
models.
