UNIVERSIDAD POLITÉCNICA DE MADRID
ESCUELA TÉCNICA SUPERIOR DE INGENIEROS DE CAMINOS, CANALES Y
PUERTOS (MADRID)
Dynamic and Seismic Analysis of Structures
Work done by : William DEBOUTO
Course 2023-24
1
In this study, using the Matlab program, we will perform various analyses on a building. Among
other things, we will conduct a modal analysis and study its behavior in free and forced
response to different seismic inputs.
The building consists of 3 floors and 4 columns.
The parameters of the building are defined through the following code snippet.
We want to introduce a Rayleigh damper into this building such that the damping for the first
and third vibration modes is 2%. Therefore, we first conduct the modal analysis to extract the
vibration frequencies of the structure. We then add the following code snippet to the previous
script.
We obtain the following frequencies: 𝜔1 = 60,28 𝑟𝑎𝑑/𝑠 𝜔2 = 168,89 𝑟𝑎𝑑/𝑠 𝜔3 =
244,06 𝑟𝑎𝑑/𝑠
𝑎 𝑎1 𝜔𝑟
We use the following formula to determine the Rayleigh damping: 𝜁𝑟 = 2𝜔0 +
𝑟 2
2
Here, 𝜁𝑟 represents the damping for mode r, and 𝑎0 and 𝑎1 are coefficients determined from
2𝜔1 𝜔3 2
the following formulas: 𝑎0 = ζ 𝜔 +𝜔 , 𝑎1 = ζ 𝜔 +𝜔 where ζ = 2% .
1 3 1 3
We add the following code to the previous script.
The damping for the second mode is: 𝜁2 = 0.0168.
Here is also presented the Rayleigh damping curve.
Figure 1: Rayleigh damping curve
3
We can calculate the transfer functions of the system based on what we want to study. Recall
that a transfer function is a tool that describes the relationship between the input and output
of a dynamic system.
Therefore, using the state space representation of a system, we appropriately define each
variable according to what we want to obtain as the system output. The following code is used
for this purpose.
4
We obtain the following results.
Output displacement
➢ Transfer functions at floor 3 from forces at floor 3 (H3_3)
0.0002 𝑠 4 + 0.002702 𝑠 3 + 14.68 𝑠 2 + 81.45 𝑠 + 2.019𝑒05
𝐻3_3 =
𝑠 6 + 17.86 𝑠 5 + 9.182𝑒04 𝑠 4 + 8.856𝑒05 𝑠 3 + 2.021𝑒09 𝑠 2 + 6.338𝑒09 𝑠 + 6.173𝑒12
5
Figure 2: Bode diagram, floor 3, forces at floor 3
When w=0, the receptance has a value of approximately -150dB, which represents an
insignificant or practically nonexistent response (displacement) at the third floor. This is normal
because the excitation frequency is zero (w=0), which means there is no excitation or very few
excitations that influence the third floor of the building.
➢ Transfer functions at floor 3 from forces at floor 1 (H3_1)
0.001163 𝑠 2 + 17.69 𝑠 + 6.731𝑒04
𝐻3_1 =
𝑠 6 + 17.86 𝑠 5 + 9.182𝑒04 𝑠 4 + 8.856𝑒05 𝑠 3 + 2.021𝑒09 𝑠 2 + 6.338𝑒09 𝑠 + 6.173𝑒12
Figure 3: Bode diagram, floor 3, forces at floor 1
6
Output acceleration
➢ Transfer functions at floor 3 from forces at floor 3 (acc_H3_3)
0.0002 𝑠 6 + 0.002702 𝑠 5 + 14.68 𝑠 4 + 81.45 𝑠 3 + 2.019𝑒05 𝑠 2 + 7.571𝑒 − 11 𝑠 + 3.459𝑒 − 07
𝑎𝑐𝑐_𝐻3_3 =
𝑠 6 + 17.86 𝑠 5 + 9.182𝑒04 𝑠 4 + 8.856𝑒05 𝑠 3 + 2.021𝑒09 𝑠 2 + 6.338𝑒09 𝑠 + 6.173𝑒12
Figure 4: Bode diagram, floor 3, forces at floor 3
➢ Transfer functions at floor 3 from forces at floor 1 (H3_1)
0.001163 𝑠^4 + 17.69 𝑠^3 + 6.731𝑒04 𝑠^2 + 1.57𝑒 − 09 𝑠 + 8.628𝑒 − 25
𝑎𝑐𝑐_𝐻3_3 =
𝑠 6 + 17.86 𝑠 5 + 9.182𝑒04 𝑠 4 + 8.856𝑒05 𝑠 3 + 2.021𝑒09 𝑠 2 + 6.338𝑒09 𝑠 + 6.173𝑒12
Figure 5: Bode diagram, floor 3, forces at floor 1
7
Now we will study the building's response in terms of acceleration when a sinusoidal force is
applied at the first floor. We use the following code.
We obtain the following results.
Figure 6: Input force
8
Figure 7: Acceleration of the first floor
In response to an excitation force of 10000N, we obtain a maximum steady-state acceleration
of approximately 0.25m/s². Due to the type of damper used, the floor's acceleration is reduced
by approximately 10 times since the maximum acceleration induced by the force is 2m/s². It is
evident that the structure quickly dampens vibrations (transient regime duration of
approximately 1 second). Although an acceleration of 0.25 m/s² may be perceptible to
individuals, we can conclude that the response of this floor is satisfactory.
The transient and steady-state regimes are identified respectively below with the black and
green rectangles.
Figure 8: Transient and steady-state regimes
Following this, instead of applying a sinusoidal signal at the first floor, we will subject the
building to various types of seismic events and evaluate its response more comprehensively.
To accomplish this, we will utilize the code below. This code initially loads data for the selected
seismic event (in this case, Christchurch_2011) and then calculates key parameters like story
drifts, peak acceleration, RMS acceleration, and peak mechanical energy based on defined
output variables using parameters C and D (output and feedthrough matrix). The variables C
and D are configured in the code to obtain both displacement and velocity simultaneously.
9
10
Subsequently, to obtain the acceleration response, we substitute C and D with the pre-defined
C111 and D111, and then employ the following code to extract the desired values.
11
The analysis was conducted for three types of seismic events, and we obtain the following
results.
➢ Christchurchc_2011 earthquake
Figure 9: Ground movement, story displacements, story drifts
Figure 10: Ground movement, story accelerations
12
Figure 11: Kinetic energy, elastic energy, mechanical energy, dissipative energy, total
energy
➢ Kobe earthquake
Figure 12: Ground movement, story displacements, story drifts
13
Figure 13: Ground movement, story accelerations
Figure 14: Kinetic energy, elastic energy, mechanical energy, dissipative energy, total
energy
14
➢ El centro earthquake
Figure 15: Ground movement, story displacements, story drifts
Figure 16: Ground movement, story accelerations
15
Figure 17: Kinetic energy, elastic energy, mechanical energy, dissipative energy, total
energy
RMS acceleration Peak story Peak Mechanical
[m/s²] acceleration [m/s²] energy [J]
Christchurchc_2011 7.2396 × 10−4 0.0082 1.8470 × 10−4
earthquake
Kobe earthquake 0.0013 0.0059 4.3238 × 10−5
El centro earthquake 2.4048 × 10−4 0.0016 5.4172 × 10−6
Table 1: Performance indicators of the building under various earthquakes
The RMS acceleration here is a statistical measure of the building's acceleration amplitude
during the seismic event. The peak story acceleration represents the acceleration experienced
by the building during the seismic event. Mechanical energy is the sum of kinetic and elastic
energy in the structural system; kinetic energy represents the energy associated with mass
movement within the structure during the seismic event, while elastic energy is the energy
stored in elastically deformed structural elements during seismic motion.
Among the three earthquakes studied, the Christchurchc_2011 earthquake has the highest
amplitude (approximately 22m/s²). This is evident from the peak story acceleration, which in
this case is 0.0082m/s², the highest among the three earthquakes. The "El Centro earthquake"
has the lowest amplitude among the three earthquakes, which is also reflected in the total
16
accumulated mechanical energy, indicating that the building does not accumulate much
energy due to minimal movement.
Regarding the building's performance, focusing on the Christchurch earthquake, for an
excitation of 22m/s², the structure's maximum acceleration is 0.0082m/s², demonstrating a
high capacity of the building to withstand significant seismic forces.
17