Scribd
Upload
122 views

CE 636 Assignment 2

This document analyzes the response of a single degree of freedom mass-spring-damper system when subjected to an external force. The response is derived using two methods: traditional differential equations and Duhamel's Integral. The results of the two methods are plotted and compared, showing differences in the transient response but convergence in the steady state response as expected.

Uploaded by

Rodrigo Romero
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
122 views

CE 636 Assignment 2

This document analyzes the response of a single degree of freedom mass-spring-damper system when subjected to an external force. The response is derived using two methods: traditional differential equations and Duhamel's Integral. The results of the two methods are plotted and compared, showing differences in the transient response but convergence in the steady state response as expected.

Uploaded by

Rodrigo Romero
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

California State University, Northridge 1

Forced Response of Single Degree of Freedom Mass-


Damper-Spring System, Differential Equation v.
Duhamel Integral.
Rodrigo Romero

Civil Engineering 636 Structural Dynamics, CSUN, Prepared for Dr. Tadeh Zirakian for 2.5 percent on 2/25/19

Abstract- In this project, a single degree of freedom (SDOF)


mass-damper-spring system is evaluated and solved using two
methods. Traditional Differential Equations (DE) methods
and Duhamel Integral are the methods used to solve the
response of the system. The results of these methods are
compared and evaluated. As expected Duhamel’s method does
not reflect the actual transient response of the system, Figure 1.
however, it does reflect the steady state response. SDOF Mass-Spring-Damper System and FBD. Esfandiari [2].

Index Terms- Forced Response. Single Degree of Freedom. (1) ∑ 𝐹 = 𝑚𝑎 = 𝑚𝑥̈


Mass, Spring, Damper System. Differential Equation. Duhamel
Integral.
(2) 𝑚𝑥̈ + 𝑏𝑥̇ + 𝑘𝑥 = 𝐹(𝑡)
I. INTRODUCTION The parameters of the systems are given in the Table 1 below.

T his project analyzes the response, of a Forced Response SDOF


mass-spring-damper system, obtained via classical
Differential Equations method and Duhamel’s Integral. The
Table 1: Parameters of the System.
Given Value Unit
Mass m 1 kg
response of the system is derived using the procedure outline in
Spring Const. K 1 N/m
Zirakian [1]. The procedure is as follows: 1) Mathematical
Damper Const. b 1 N*s/m
modeling, 2) Derivation of governing equations, 3) Solution of the
Initial Position x0 1 m
governing equation, 4) Interpretation of the results. It is the third
Initial Velocity v 1 m/s
step that is critical in this paper, we analyze the difference in the
response when obtained in the aforementioned techniques. Both Force Applied F(t) 𝟏𝟎𝒆𝟐.𝟓𝒕 N
Derived from Given: Value Unit
responses are plotted in MATLAB and analyzed. Since Duhamel’s
Damping Ratio Zeta 0.5
Integral method does not take in to consideration the initial
Natural Fz. wn 1 rad/s
conditions we expect the two responses to vary in the transient
Damped Fz. wd 0.866 rad/s
response, however, as the response approaches the steady-state
response we expect these two responses to converge. Using Table 1 and Equation 2, we get the full EOM of the
The Project is divided into the following sections: system, shown in Equation 3.
I. Introduction (3) 𝑥̈ + 𝑥̇ + 𝑥 = 10𝑒2.5𝑡
II. System and Parameters
III. Derivation of Response III. DERIVATION OF RESPONSE
IV. Results
In this section we will show the equations and methods used to
V. Discussion
obtain the response of the system using (A) Classical Differential
Equations and (B) Duhamel’s Integral:
II. SYSTEM AND PARAMETERS
The system that will be analyzed is shown in Figure 1 below. A. Classical Differential Equations
Along with the system, Figure 1 also shows the Free Body In this approach, we need to solve two equations, the transient
Diagram (FBD) that is used to derive the Equation of Motion (homogenous) and the steady-state (particulate) responses. We use
(EOM) of the system. Newton’s Second law of motion, equation the method of undetermined coefficients to find the constants in
1, is used to derive the EOM. The EOM of the system is shown in the steady state equation, then use both equations along with the
equation 2. ICs to solve for the unknown constants in the transient response.
The response of the system:
(4) 𝑥(𝑡) = 𝑥ℎ (𝑡) + 𝑥𝑝 (𝑡)
The steady state response is assumed as:
(5) 𝑥𝑝 (𝑡) = 𝐴𝑒 2.5𝑡
California State University, Northridge 2

We then differentiate equation 5 twice and plug the results in to (11) 𝑥(𝑡) = 1.02564𝑒2.5𝑡 − 1.0256𝑒−.5𝑡 cos(.866𝑡) − 3.552𝑒−.5𝑡 sin(.866𝑡)
equation 3. We set the coefficients of like terms equal to each other
and obtain the steady-state response as: IV. RESULTS
2.5𝑡 The response derived using classical differential equations
(6) 𝑥𝑝 (𝑡) = 1.02564𝑒
method is shown in equation 8. The response derived using
For the transient response, we use the general solution from Duhamel’s Integral is shown in equation 11. The responses were
Zirakian [1]: plotted using MATLAB, the plot is shown in Figure 2 below.
(7) 𝑥ℎ (𝑡) = 𝜒𝑜 ∗ 𝑒 −𝜉𝜔𝑛 𝑡 ∗ sin(𝜔𝑑 𝑡 + Φ𝑜 )
V. CONCLUSION
Plugging in the results from equations 6 and 7, into equation 4 and
differentiating, we can use the ICs to solve for the unknowns. This After derivation of the system’s response, we find that the
yield: responses found in two methods, Classical Differential Equations
and Duhamel’s Integral, yield two slightly different responses. As
(8) 𝑥(𝑡) = −2.79813 ∗ 𝑒 −.5𝑡 ∗ sin(. 866𝑡 + 3.13243) + 1.02564𝑒2.5𝑡
expected Duhamel’s method does not reflect the actual transient
B. Duhamel’s Integral response of the system, however, it does reflect the steady state
From Zirakian [1], we use the general equation for Duhamel’s response. Unfortunately, because the system is unstable, the
Integral: convergence is not seen completely in figure 2 below. Had we
𝑡
shown the convergence, by extending the limits of the axis, then
(9) 𝑚 the transient response difference would be less noticeable. An
𝑥ℎ (𝑡) = ∫ ∗ 𝐹(𝜏) ∗ 𝑒 −𝜉𝜔𝑛 (𝑡−𝜏) ∗ sin( 𝜔𝑑 (𝑡 − 𝜏) ) 𝑑𝜏
0 𝜔𝑑 unstable system was expected as the forcing function is a positive
Where: exponential function which is expected to grow without bound as
(10) 𝐹(𝜏) = 10𝑒 2.5𝜏 time approaches infinity.
Inputting the forcing function, equation 10, into equation 9, we can REFERENCES
now differentiate to obtain the response of the system. The
[1] Zirakian, Tadeh, “Vibration Analysis Lecture Notes),” 2019.
integration was performed by hand, by doing several iterations of [2] Esfandiari, Ramin S, and Bei Lu. Modeling and Analysis of Dynamic
integration by parts. Due to the extensive hand calculations, these Systems, Second Edition. Hoboken: CRC Press, 2014. Internet resource..
are omitted from this report; available upon request.
After integration, the response is found to be:

Figure 2.
Response of System via D.E. and Duhamel Integral.

You might also like