Geology Reference
In-Depth Information
The optimization routine is carried out in two
step: a first random design generation is used to
explore all the design space obtaining at least a
number of 30 feasible solution; in a second time
the best two feasible solutions are used as starting
points of sub-problem approximation runs. In this
manner, it is possible to reduce the risk of obtaining
a local optimum solution with a relatively small
increase of computational time.
The described optimization routine took into
account some constraints in order to meet the
space provided in the frame and to avoid failure.
The analysis proposed considered four differ-
ent constraints:
on the dimensions of the plates for mounting
the dissipaters in the 3D frame.
4. Transversal load, corresponding to the
maximum horizontal displacement of 4 mm,
which was initially limited to 20 kN, so that
the resulting panel is compatible with the
test frame where it will be mounted for next
shaking table tests (see par. 4).
The fourth constraint was added to obtain
several panel configurations with different stiff-
nesses, so that they can be adapted to a structure
with different behavior from one another. The
stiffness of the panel to be mounted in a structure,
in fact, must be appropriate to the stiffness of the
structure to be protected.
At the conclusion of this process, the opti-
mization routine provides the optimal geometric
configuration corresponding to that for which the
plastic strain energy takes the maximum value
over all those examined.
With reference to the choice of the geometrical
parameters that allow the description of the con-
figuration of the panel, Table 2 shows the variation
of the parameters ranges and the final configuration
chosen. The optimized panel obtained for 20 kN
transversal load is represented in Figure 7.
1. The stress in the aluminum, whose maxi-
mum value was assumed coincident with
the ultimate tensile stress, equal to 90 MPa
(cf. Table 1);
2. The overall height of the panel, which was
allowed to vary in the range 420-470 mm
on the basis of the considerations relative
to the maximum acceptable size for the test
frame, as shown in Figure 6;
3. The width of the panel, which was allowed
to vary in the range 100-250 mm, depending
Table 2. Variability range of the geometrical parameters and the optimization results
Variability
Range
Optimal
Configuration
Final Configuration
Geometrical Parameters
n
x
Horizontal window
2-6
3
3
n
y
Vertical window
1-4
4
4
b
1
Width of the lateral steel stiffener [mm]
5-10
9.5
8
b
2
Width of the aluminium window [mm]
30-200
46.4
40
b
3
Width of the internal steel stiffener [mm]
5-10
8.4
8
h
1
Height of the external steel stiffener [mm]
5-10
9.8
10
h
2
Height of the aluminium window[mm]
30-200
110.8
100
h
3
Height of the internal steel stiffener [mm]
5-10
10.9
10
t
1
Thickness of the steel plate [mm]
1-4
1
1
t
2
Thickness of the aluminium plate[mm]
1-4
2.5
3
t
a
Projection of the aluminium plate [mm]
1-4
2.7
3
Search WWH ::
Custom Search