Foods,
Journal Year:
2024,
Volume and Issue:
13(21), P. 3523 - 3523
Published: Nov. 4, 2024
Falling
damage
is
the
most
common
form
of
sustained
by
kiwifruit
during
process
picking
and
post-processing,
it
difficult
to
conduct
a
quantitative
analysis
this
phenomenon
through
traditional
experimental
methods.
In
order
deeply
understand
sensitivity
falling
collision
damage,
finite
element
numerical
simulation
method
was
used
evaluate
predict
harvesting.
First,
we
obtained
appearance
characteristics
reverse
engineering
technology
determined
geometric
mechanical
property
parameters
physical
mechanics
experiments.
Then,
according
fruit
tissue
structure,
multiscale
model,
including
skin,
pulp,
core,
constructed
simulate
effects
different
heights,
angles,
contact
surface
materials
on
accuracy
model
verified
Finally,
based
results,
Box-Behnken
design
employed
within
response
methodology
establish
prediction
for
drop
across
materials.
The
results
showed
that
maximum
relative
error
between
speed
change
using
high-speed
camera
5.19%.
high
rationality
in
energy
distribution,
with
value
hourglass
not
exceeding
0.08%
internal
energy.
On
material
large
elastic
modulus,
higher
height
larger
angle
will
significantly
increase
risk
bruise.
When
steel
plate,
1
m,
90°;
bruise
reached
6716.07
mm
Foods,
Journal Year:
2024,
Volume and Issue:
13(5), P. 785 - 785
Published: March 3, 2024
In
the
handling
or
processing
process,
fruits
are
easily
crushed
by
external
loads.
This
type
of
damage
in
fruit
often
leads
to
internal
pulp
browning
and
rotting,
with
severity
largely
dependent
on
tissue’s
geometric
mechanical
properties.
kiwifruits,
their
thin
skin
dark-colored
flesh,
it
is
particularly
challenging
observe
analyze
caused
extrusion
through
traditional
experimental
methods.
The
objective
this
research
construct
a
multi-scale
finite
element
model
encompassing
skin,
core
measuring
properties
kiwifruit,
assess
predict
characteristics
under
compression,
verify
accuracy
experiments.
results
indicated
that
kiwifruits
demonstrated
different
compressive
strengths
directions
during
compression.
strength
axial
direction
was
higher
than
radial
direction,
there
little
difference
between
long
short
directions.
flesh
tissue
most
vulnerable
followed
core.
At
strain
levels
below
5%,
no
noticeable
kiwifruit.
However,
when
exceeded
began
manifest
some
tissue.
To
maintain
quality
storage
transportation,
stacking
height
should
not
exceed
77
48
53
direction.
analysis
showed
established
can
effectively
simulate
behavior
compression
loads,
which
helpful
for
deeper
understanding
provides
theoretical
basis
technical
guidance
minimizing
handling.
Horticulturae,
Journal Year:
2025,
Volume and Issue:
11(4), P. 440 - 440
Published: April 20, 2025
The
bruising
of
fruits
occurs
at
various
stages,
including
picking,
transportation,
and
sale.
For
with
large
kernels
that
occupy
a
significant
portion
their
overall
volume,
considering
the
impact
kernel
is
crucial
in
elucidating
mechanisms
controlling
bruise
formation.
This
study
employs
reverse
engineering
to
develop
composite
finite
element
model
loquat
peel,
flesh,
kernels.
Bruise
formation
during
collisions
analyzed
from
perspectives
contact
force,
equivalent
stress,
energy,
susceptibility,
aiming
reveal
role
fruit
core
process.
In
this
paper,
we
propose
use
3D
printing
technology
accurately
quantify
measurement
for
results
showed
maximum
internal
energy
between
steel/wood
were
essentially
consistent,
but
all
exceeded
those
observed
when
using
rubber.
Due
blocking
stress
transmission
by
kernel,
susceptibility
loquats
increases
height
before
decreasing.
elucidates
mechanism
provides
methods
ideas
research
precise
complex
characteristics.
This
study
is
a
pioneering
endeavor
to
investigate
the
capabilities
of
Large
Language
Models
(LLMs)
in
addressing
conceptual
questions
within
domain
mechanical
engineering
with
focus
on
mechanics.
Our
examination
involves
manually
crafted
exam
encompassing
126
multiple-choice
questions,
spanning
various
aspects
mechanics
courses,
including
Fluid
Mechanics,
Mechanical
Vibration,
Engineering
Statics
and
Dynamics,
Mechanics
Materials,
Theory
Elasticity,
Continuum
Mechanics.
Three
LLMs,
ChatGPT
(GPT-3.5),
(GPT-4),
Claude
(Claude-2.1),
were
subjected
evaluation
against
faculties
students
with/without
background.
The
findings
reveal
GPT-4’s
superior
performance
over
other
two
LLMs
human
cohorts
answering
across
topics,
except
for
signals
potential
future
improvements
GPT
models
handling
symbolic
calculations
tensor
analyses.
performances
all
significantly
improved
explanations
prompted
prior
direct
responses,
underscoring
crucial
role
prompt
engineering.
Interestingly,
GPT-3.5
demonstrates
prompts
covering
broader
domain,
while
GPT-4
excels
focusing
specific
subjects.
Finally,
exhibits
notable
advancements
mitigating
input
bias,
as
evidenced
by
guessing
preferences
humans.
unveils
substantial
highly
knowledgeable
assistants
both
pedagogy
scientific
research.
