Response of SDOF Systems
to
Non-periodic Excitation
Response of SDOF Systems to Non-periodic Excitation
Ideal step input
Ideal step input
Initial conditions -> A and B
Given, initial conditions
Solution
Particular solution:
Complementary solution:
Total solution, Step Response:
Response of SDOF Systems to Non-periodic Excitation
Ideal step input
Under damped system:
Response =
Undamped system: Response =
Ideal step input
Undamped system:
Under damped system:
𝑃𝑜
𝑢 𝑡 = (1 − 𝑐𝑜𝑠𝜔𝑛 𝑡)
𝑘
𝑢 𝑡
= (1 − 𝑐𝑜𝑠𝜔𝑛𝑡)
𝑃𝑜
𝑘
Response ratio = 𝑅 𝑡 = (1 − 𝑐𝑜𝑠𝜔𝑛𝑡)
Dynamic load factor =
under damped system undamped system
𝑅 𝑡 = 1 − 𝑐𝑜𝑠𝜔𝑛𝑡
𝑅 𝑡 =
Rmax = 2
Thus, when a load is applied instantaneously to an undamped system,
a maximum displacement of twice the static displacement is attained.
This is the reason that a safety factor of 2 is frequently applied to the
design of structures that will be subjected to rapidly applied loads.
Rectangular Pulse Loading
Forced- free vibration after time td Undamped system:
Vibration
Case 1: Forced-Vibration Era (0 < t < td)
𝑃𝑜
Response: 𝑢1 𝑡 = (1 − 𝑐𝑜𝑠𝜔𝑛𝑡)
𝑘
Case 2: Residual-Vibration Era (td < t)
Free vibration after time td
𝑢ሶ 1 𝑡𝑑
Response: 𝑢2 𝑡 = 𝑢1 𝑡𝑑 𝑐𝑜𝑠𝜔𝑛 𝑡 − 𝑡𝑑 + sin𝜔𝑛 𝑡 − 𝑡𝑑
𝜔𝑛
Ramp Loading
Given,
Initial conditions
Ramp input function.
Particular solution Using the initial conditions
total solution
The Solution for t>tr , can be determined as follows
as for step input
SHORT-DURATION IMPULSE
large force acting for a very short time
The equation of motion and initial conditions
Eq. 1
Force p(t) of duration td << Tn having an impulse
SHORT-DURATION IMPULSE
Eq. 1
Integrating Eq. 1 with respect to time and applying initial conditions:
Eq.2
uavg is the (small) average displacement in the time interval 0 < t < td
For td -> 0, As (td << Tn)
The second term in Eq.2 can be ignored, leaving
Thus, an impulse has the effect of giving the mass an initial velocity =
leaving it with an initial displacement of
SHORT-DURATION IMPULSE
Accordingly, it becomes free vibration problem with initial condition,
Undamped system
The impulse response =
Unit impulse response function, When, I = 1,
Viscous-damped SDOF system
The impulse response
Unit impulse response function
Response of SDOF Systems
to
General Dynamic Excitation
RESPONSE OF TO GENERAL DYNAMIC EXCITATION
CONVOLUTION INTEGRAL METHOD/ Duhamel Integral Method
The convolution integral is
The total response at time t will be the sum
based on the Principle of Superposition of the responses due to all incremental
Valid for linear systems impulses at times τ prior to time t
Convolution integral
General Dynamic Loading Duhamel integral expression
The response of the system
to an impulse dI = p(τ) dτ
Incremental response
CONVOLUTION INTEGRAL METHOD
Response for general loading,
The unit impulse response function
undamped system viscous-damped systems
Response for general loading with initial condition,
undamped system under damped system
Note: In General, Convolution integral form,
Problems
Rigid beam
step pulse, having Po = 55 kips
Determine
the maximum displacement,
the maximum base shear, and
the maximum lateral resistance of the two columns.
Given:
W12 × 53 : I = 425, S = 70.6, Z = 77.9, A572 Grade 50
The maximum elastic resistance is less
than the demand, the structure will be
pushed into the inelastic range and will
have a permanent displacement when the
load is removed.
lateral force, Vdemand
lateral force capacity
Now let the moment capacity of the
columns increase to the full plastic
moment capacity:
lateral force capacity (62.3 kips)< Demand (110 kips)
Problems
Consider the frame is subjected to a short duration pulse, as shown in Figure
Problem
The single-bay frame is subjected to an impulse load having the shape of a half
sine wave and a peak load of 500 kips.
The properties of the columns : I = 1000 in4 and S = 143 in3.
Estimate the maximum stress in the columns.
0.4 sec
Example:
(a) Use the Duhamel integral to determine the response of an undamped SDOF system to
a “blast” loading specified by the triangular pulse shown in Fig.
(b) Obtain expressions that are valid for t < td and for t > td. The system is initially at rest
Triangular pulse
excitation force
Using integration by parts, we get
Simplifying this expression, we get the following expression for the response ratio for
(b) For td < t, the Duhamel integral expression for the response is
From Example, we can appreciate that although the response of an SDOF system to an arbitrary input
may, in principle, be obtained by use of the Duhamel integral, the work involved in evaluating the
integrals may be tedious. Also, it may be necessary to obtain the response to an input that is known
graphically but not in analytical form: for example, a plot of ground acceleration versus time. In cases
such as these, a numerical procedure is needed. Numerical procedures for computing dynamic response
are discussed in Lecture 6.
Response of SDOF Systems
to
General Dynamic Loading
Numerical Method
RESPONSE TO GENERAL DYNAMIC LOADING:
STEP BY STEP METHODS
At time step, i
At time step, i+1
Notation for time-stepping methods
RESPONSE TO GENERAL DYNAMIC LOADING:
STEP BY STEP METHODS
INTEGRATION METHODS
Constant average acceleration
INTEGRATION METHODS
linearly varying acceleration
Newmark Beta Methods
Constant average acceleration Method Linear acceleration Method
and and
These equations are in Implicit form and requires iterations
Constant average acceleration Method
Newmark Beta Methods
Conversion to Explicit Formulation
Constant average acceleration Method
Writing the equations of dynamic equilibrium at time
t1
Newmark Beta Methods
Constant average acceleration Method
The displacement at the end of the time step, v1
is calculated.
The velocity at that time is calculated
The Acceleration at that time is calculated
Newmark Beta Methods
Linear acceleration Method
The displacement at the end of the time step, v1
is calculated.
The velocity at that time is calculated
The Acceleration at that time is calculated
An SDF system : m = 0.2533 kip-sec2/in., k = 10 kips/in., Tn = 1 sec (ωn = 6.283
rad/sec), and ζ = 0.05.
Determine the response u(t) of this system to p(t) as shown in Figure
(use the constant average acceleration method) using Δt = 0.1 sec.
m = 0.2533
k = 10
c = 0.1592
u0 = 0
ů0 = 0
p0 = 0
Solution
Initial calculations:
Δt = 0.1 sec
Solution
Solve by linear acceleration method using Δ t = 0.1 sec.
