MDOF Dynamics
-
Upload
ronnie1478 -
Category
Documents
-
view
57 -
download
12
description
Transcript of MDOF Dynamics
Structural Dynamics of LinearElastic MDOF Systems
u1
u2
u3
MDOF Dynamics 1Revised 3/09/06
Structural Dynamics of ElasticMDOF Systems
• Equations of Motion for MDOF Systems• Uncoupling of Equations through use of
Natural Mode Shapes• Solution of Uncoupled Equations• Recombination of Computed Response• Modal Response History Analysis• Modal Response Spectrum Analysis• Equivalent Lateral Force Method
MDOF Dynamics 2
Relevance to ASCE 7-05
ASCE 7-05 Provides guidance for three specific analysis procedures:
• Equivalent Lateral Force Analysis (ELF)• Modal Superposition Analysis (MSA)• Response History Analysis (RHA)
ELF Usually Allowed ELF Not Allowed
Cs
See ASCE 7-05Table 12.6-1
MDOF Dynamics 3
3.5Ts TTs
ux
uyrz
Majority of Massis in Floors
Typical Nodal DOF
Motion isPredominantlyLateral
Planar Frame with 36 Degrees of Freedom
1 2 3 4
5 6 7 8
9 10 11 12
13 14 15 16
MDOF Dynamics 4
Planar Frame with 36 Static Degrees of Freedombut with only THREE Dynamic DOF
u1
u2
u3
1
2
3
uU = u
u
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
MDOF Dynamics 5
Development of Flexibility Matrix
d1,1
d2,1
d3,1
f1=1 kip
d1,1
d2,1
d3,1
MDOF Dynamics 6
Development of Flexibility Matrix
d1,2
d2,2
d3,2
f2=1 kipd1,1 d1,2
d2,1 d2,2
d3,1 d3,2
MDOF Dynamics 7
Development of Flexibility Matrix
d1,3
d2,3
d3,3
f3=1 kip
d1,1
d2,1
d3,1
d1,2
d2,2
d3,2
d1,3
d2,3
d3,3
MDOF Dynamics 8
Concept of Linear Combination of Shapes (Flexibility)
1,1 1,2 1,3
2,1 1 2,2 2 2,3 3
3,1 3,2 3,3
d d dU = d f + d f + d f
d d d
⎧ ⎫ ⎧ ⎫ ⎧ ⎫⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭ ⎩ ⎭
1,1 1,2 1,3 1
2,1 2,2 2,3 2
3,1 3,2 3,3 3
d d d fU d d d f
d d d f
⎡ ⎤⎧ ⎫⎢ ⎥⎪ ⎪= ⎨ ⎬⎢ ⎥
⎪ ⎪⎢ ⎥⎩ ⎭⎣ ⎦-1K = DD F = U K U = F
MDOF Dynamics 9
Static Condensation
DOF with mass
Massless DOF , ,
, ,
K K FUK K 0U
m m m n mm
n m n n n
⎡ ⎤ ⎧ ⎫⎧ ⎫=⎨ ⎬ ⎨ ⎬⎢ ⎥
⎩ ⎭ ⎩ ⎭⎣ ⎦
, ,K U K U Fm m m m n n m+ =1
, ,K U K U 0n m m n n n+ =2
MDOF Dynamics 10
Static Condensation(continued)
2 1, ,U K K U−= −n n n n m mRearrange
1 1, , , ,K U K K K U F−− =m m m m n n n n m m m
Plug into
1, , , ,K K K K U F−⎡ ⎤− =⎣ ⎦m m m n n n n m m mSimplify
MDOF Dynamics 11
Condensed Stiffness Matrix
1, , , ,K K K K K−= −m m m n n n n m
Idealized Structural Property Matricesm1
m3
m2k1
k2
k3
f1(t), u1(t)
f2(t), u2(t)
f3(t), u3(t)
1 11 1 2 2
2 2 3
k -k 0K -k k + k -k
0 -k k + k
⎡ ⎤⎢ ⎥= ⎢ ⎥⎢ ⎥⎣ ⎦
12
3
m 0 0M 0 m 0
0 0 m
⎡ ⎤⎢ ⎥= ⎢ ⎥⎢ ⎥⎣ ⎦
1
2
3
( )( ) ( )
( )
f tF t f t
f t
⎧ ⎫⎪ ⎪= ⎨ ⎬⎪ ⎪⎩ ⎭
1
2
3
( )( ) ( )
( )
u tU t = u t
u t
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
MDOF Dynamics 12
Note: Damping to be shown later
Coupled Equations of Motion for Undamped Forced Vibration
MU(t) KU(t) F(t)+ =
1 1 1
2 2 2
3 3 3
1 0 0 u ( ) 1 1 0 u ( ) f ( )0 2 0 u ( ) 1 1 2 2 u ( ) f ( )0 0 3 u ( ) 0 3 2 3 u ( ) f ( )
m t k k t tm t k k k k t t
m t k k k t t
−⎧ ⎫ ⎧ ⎫ ⎧ ⎫⎡ ⎤ ⎡ ⎤⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎢ ⎥ ⎢ ⎥+ − + − =⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎢ ⎥ ⎢ ⎥⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎢ ⎥ ⎢ ⎥− +⎣ ⎦ ⎣ ⎦⎩ ⎭ ⎩ ⎭ ⎩ ⎭
1 1 2 1
2 1 2 2 3 2
3 2 3 3 3
1u ( ) k1u ( ) k1u ( ) f ( )2u ( ) 1u ( ) 1u ( ) 2u ( ) 2u ( ) f ( )3u ( ) 2u ( ) 2u ( ) 3u ( ) f ( )
m t t t tm t k t k t k t k t tm t k t k t k t t
+ − =− + + − =
− + + =
DOF 1
DOF 2
DOF 3
MDOF Dynamics 13
We need to develop a way to solve theequations of motion.
• This will be done by a transformation of coordinatesfrom Normal Coordinates (displacements at the nodes)To Modal Coordinates (amplitudes of the naturalMode shapes).
• Because of the Orthogonality Property of the natural modeshapes, the equations of motion become uncoupled,allowing them to be solved as SDOF equations.
• After solving, we can transform back to the normalcoordinates.
MDOF Dynamics 14
Solutions for System in Undamped Free Vibration(Natural Mode Shapes and Frequencies)
MU(t) KU(t) 0+ =
Assume U( ) sint tφ ω=
2K M 0φ ω φ− =Then has three (n) solutions:
2U( ) ω sint tφ ω= −
1,1
1 2,1 1
3,1
,φ
φ φ ωφ
⎧ ⎫⎪ ⎪= ⎨ ⎬⎪ ⎪⎩ ⎭
1,2
2 2,2 2
3,2
,φ
φ φ ωφ
⎧ ⎫⎪ ⎪= ⎨ ⎬⎪ ⎪⎩ ⎭
1,3
3 2,3 3
3,3
,φ
φ φ ωφ
⎧ ⎫⎪ ⎪= ⎨ ⎬⎪ ⎪⎩ ⎭
Natural Mode ShapeNatural Frequency
MDOF Dynamics 15
Solutions for System in Undamped Free Vibration
For a SINGLE Mode2K ω Mφ φ=
2K MΦ= ΦΩ For ALL Modes
[ ]Φ = φ φ φ1 2 3Where:
21
2 22
23
ω
ω
ω
⎡ ⎤⎢ ⎥
Ω = ⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
MDOF Dynamics 16
TM IΦ Φ=
1,i 1.0φ =orNote: Mode shape has arbitrary scale; usually
MODE 1 MODE 3MODE 2
1,1φ
2,1φ
3,1φ 3,2φ
2,2φ
1,2φ
3,3φ
2,3φ
1,3φ
Mode Shapes for Idealized 3-Story Frame
Node
Node
Node
MDOF Dynamics 17
Concept of Linear Combination of Mode Shapes(Transformation of Coordinates)
U Y=Φ
1,1 1,2 1,3
2,1 1 2,2 2 2,3 3
3,1 3,2 3,3
U y y yφ φ φφ φ φφ φ φ
⎧ ⎫ ⎧ ⎫ ⎧ ⎫⎪ ⎪ ⎪ ⎪ ⎪ ⎪= + +⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭ ⎩ ⎭
1,1 1,2 1,3 1
2,1 2,2 2,3 2
3,1 3,2 3,3 3
yU y
y
φ φ φφ φ φφ φ φ
⎡ ⎤⎧ ⎫⎪ ⎪⎢ ⎥= ⎨ ⎬⎢ ⎥⎪ ⎪⎢ ⎥⎩ ⎭⎣ ⎦
MDOF Dynamics 18
Mode Shape
Modal Coordinate = amplitude of mode shape
Orthogonality Conditions
[ ]Φ = φ φ φ1 2 3
*1
T *2
*3
m
M m
m
⎡ ⎤⎢ ⎥
Φ Φ= ⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
Generalized Mass*1
T *2
*3
k
K k
k
⎡ ⎤⎢ ⎥
Φ Φ= ⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
Generalized Stiffness
*1*2*3
f ( )F( ) f ( )
f ( )
⎧ ⎫⎪ ⎪Φ =⎨ ⎬⎪ ⎪⎩ ⎭
T
tt t
t
Generalized Force*1
T *2
*3
c
C c
c
⎡ ⎤⎢ ⎥
Φ Φ= ⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
Generalized Damping
MDOF Dynamics 19
Development of Uncoupled Equations of Motion
MU+CU+ KU F( )t=MDOF Equation of Motion:
U Y= ΦTransformation of Coordinates:
MFY+ CFY+ KFY F( )t=Substitution:T T T TM Y C Y K Y F(t)Φ Φ +Φ Φ +Φ Φ =ΦPremultiply by TΦ :
* * * *1 1 1 11 1 1
* * * *2 2 2 2 2 2 2
* * * *3 3 33 3 3 3
m c k f ( )y y ym y + c y + k y f ( )
y y ym c k f ( )
t
t
t
⎡ ⎤ ⎡ ⎤ ⎡ ⎤ ⎧ ⎫⎧ ⎫ ⎧ ⎫ ⎧ ⎫⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎪ ⎪⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪=⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎨ ⎬ ⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪
⎩ ⎭ ⎩ ⎭ ⎩ ⎭⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎪ ⎪⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎩ ⎭
Using Orthogonality Conditions: Uncoupled Equations of Motion are:
MDOF Dynamics 20
Development of Uncoupled Equations of Motion(Explicit Form)
* * * *1 1 1 1 1 1 1m y +c y + k y = f ( )tMODE 1
* * * *2 2 2 2 2 2 2m y +c y + k y = f ( )tMODE 2
* * * *3 3 3 3 3 3 3m y +c y + k y = f ( )tMODE 3
MDOF Dynamics 21
Simplify by Dividing Through by m* and defining*i
i *i i
c2m
ξω
=
Development of Uncoupled Equations of Motion(Explicit Form)
2 * *1 1 1 1 1 1 1 1y 2 y y f (t) / mξ ω ω+ + =MODE 1
2 * *2 2 2 2 2 2 2 2y 2 y y f (t) / mξ ω ω+ + =MODE 2
2 * *3 3 3 3 3 3 3 3y 2 y y f (t) / mξ ω ω+ + =MODE 3
MDOF Dynamics 22
ug r,1U
,1
,2
,3
,1
,2
,3
( ) u ( )F ( ) M ( ) u ( )
( ) u ( )
1.0 u ( )M 1.0 ( ) M u ( )
1.0 u ( )
⎧ ⎫+⎪ ⎪= + =⎨ ⎬⎪ ⎪+⎩ ⎭
⎧ ⎫⎧ ⎫⎪ ⎪ ⎪ ⎪+⎨ ⎬ ⎨ ⎬⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭
g r
I g r
g r
r
g r
r
u t tt u t t
u t t
tu t t
t
Move to RHS as EFFF ( ) = - M R ( )gt u t
Earthquake “Loading” for MDOF System
MDOF Dynamics 23
MDOF Dynamics 24
* TgF (t) MR u (t)= −Φ
1R= 1
1
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
m1=2
m2=3
m3=1u1
u2
u3 2 1M 3
1
+⎡ ⎤⎢ ⎥= ⎢ ⎥⎢ ⎥⎣ ⎦
1R= 1
0
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
u1
u2
u3
m1=2
m2=3
m3=1
2M 3
1
⎡ ⎤⎢ ⎥= ⎢ ⎥⎢ ⎥⎣ ⎦
Modal Earthquake Loading
( )u tg
( )u tg
Definition of Modal Participation Factor
For earthquakes: * Ti i gf (t) MRu (t)φ= −
Typical Modal Equation:
Modal Participation Factor pi
i
* T2 i i
i i i i i i g* *i
f (t) MRy 2 y y u (t)m m
φξω ω+ + = = −
MDOF Dynamics 25
Caution Regarding Modal Participation Factor
Ti iMφ φ
Ti
i *i
MRpm
φ=
It’s value is dependent on the (arbitrary) method used to scale the mode shapes.
MDOF Dynamics 26
Variation of First Mode Participation Factorwith First Mode Shape
p1=1.0 p1=1.4 p1=1.6
MDOF Dynamics 27
1.0 1.0 1.0
Concept of Effective Modal Mass
2 *i i im p m=For each Mode i
• The sum of the effective modal mass for all modes isequal to the total structural mass.
• The value of effective modal mass isindependent of mode shape scaling.
• Use enough modes in the analysis to providea total effective mass not less than 90% of thetotal structural mass
MDOF Dynamics 28
Variation of First Mode Effective Masswith First Mode Shape
1m / M1m / M1m / M
MDOF Dynamics 29
1.0
=0.71.0
=1.01.0
=0.9
Derivation of Effective Modal Mass (1)For each mode:
2i i i i i i i gy 2 y y p uξ ω ω+ + = −
SDOF system:
2i i i i i i gq 2 q q uξω ω+ + = −
Modal response history, qi(t) is obtained by first solving the SDOF system.
MDOF Dynamics 30
Derivation of Effective Modal Mass (2)
y ( ) p q ( )i i it t=From Previous Slide
u ( ) y ( )i i it tφ=Recall
u ( ) p q ( )i i i it tφ=Substitute
MDOF Dynamics 31
Derivation of Effective Modal Mass (3)
Applied “static” forces required to produce ui(t):
( ) ( ) ( )i i i i iV t Ku t PK q tφ= =
2i i iK Mφ ω φ=Recall:
Substitute:
2( ) ( )i i i i iV t M P q tφ ω=
MDOF Dynamics 32
Derivation of Effective Modal Mass (4)T
i iV V R=Total shear in mode:
2 2( ) ( )T Ti i i i i iV M RP q t MRPφ ω φ= = ( )iq tω
“Acceleration” in modeDefine effective modal mass: T
i i iM MRPφ=
and2 ( )i i iV M q tω=
MDOF Dynamics 33
Derivation of Effective Modal Mass (5)
TT Ti
i i i i i iTi i i
MRM MRP M PM
φφ φ φφ φ⎡ ⎤
= = ⎢ ⎥⎣ ⎦
*2iii mPM =
MDOF Dynamics 34
Development of a Modal Damping Matrix
In previous development, we have assumed:
Φ ΦTCc
cc
=
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
1
2
3
*
*
*
• Rayleigh “Proportional Damping”• Wilson “Discrete Modal Damping”
Two Methods Described Herein:
MDOF Dynamics 35
Rayleigh Proportional Damping
MASSPROPORTIONALDAMPER
STIFFNESSPROPORTIONALDAMPER
C M K= +α β
MDOF Dynamics 36
Rayleigh Proportional Damping
C M K= +α β
MDOF Dynamics 37
For Modal Equations to Be Uncoupled:
2ω ξ φ φn n nT
nC=
Using Orthogonality Conditions:
2 2ω ξ α βωn n n= +
ξω
α ω βnn
n= +1
2 2
Φ ΦT M I=Assumes
Rayleigh Proportional Damping
Select Damping value in two modes, ξm and ξn
Compute Coefficients α and β:
⎭⎬⎫
⎩⎨⎧⎥⎦
⎤⎢⎣
⎡−
−−
=⎭⎬⎫
⎩⎨⎧
n
m
mn
mn
mn
nm
ξξ
ωωωω
ωωωω
βα
/1/12 22
C M K= +α βForm Damping Matrix
MDOF Dynamics 38
Rayleigh Proportional Damping (Example)
5% Critical in Modes 1 and 3
MDOF Dynamics 39
0.00
0.05
0.10
0.15
0 20 40 60Frequency, Radians/Sec.
Mod
al D
ampi
ng R
atio
MASSSTIFFNESSTOTAL
TYPEα = 0.41487β = 0.00324
Mode ω1 4.942 14.63 25.94 39.25 52.8
Structural Frequencies
Rayleigh Proportional Damping (Example)5% Damping in Modes 1 & 2, 1 & 3, 1 & 4, or 1 & 5
MDOF Dynamics 40
Modes α β1 & 2 .36892 0.005131 & 3 .41487 0.003241 & 4 .43871 0.002271 & 5 .45174 0.00173
0.00
0.05
0.10
0.15
0 20 40 60Fequency, Radians/sec
Mod
al D
ampi
ng R
atio
1,21,31,41,5
MODES
Proportionality Factors (5% each indicated mode)
Wilson Damping
Directly Specify Modal Damping Values ξi*
Φ ΦTCc
cc
mm
m=
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥=
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
1
2
3
1 1 1
2 2 2
3 3 3
22
2
*
*
*
* *
* *
* *
ωξω ξ
ω ξ
MDOF Dynamics 41
FORMATION OF EXPLICIT DAMPING MATRIXFROM “WILSON” MODAL DAMPING
(NOT Usually Required)
cC
nn
nn
T =
⎥⎥⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢⎢⎢
⎣
⎡
•••=ΦΦ
−−
ωξωξ
ωξωξ
22
22
11
22
11
( )Φ ΦT c C− − =
1 1
MDOF Dynamics 42
C M Mi i iT
ii
n
=⎡
⎣⎢
⎤
⎦⎥
=∑2
1
ξω φ φ
5 5 5
0
10
0
2
4
6
8
10
12
4.94 14.57 25.9 39.2 52.8
Frequency, Radians per second
Mod
al D
ampi
ng R
atio
Wilson Damping (Example)5% Damping in Modes 1 and 2, 310% in Mode 5, Zero in Mode 4
MDOF Dynamics 43
Wilson Damping (Example)5% Damping in all Modes
MDOF Dynamics 44
5 5 5 5 5
0
2
4
6
8
10
12
4.94 14.57 25.9 39.2 52.8
Frequency, Radians per second
Mod
al D
ampi
ng R
atio
Solution of MDOF Equations of Motion
• Explicit (Step by Step) Integration of Coupled Equations
• Explicit Integration of FULL SET of Uncoupled Equations
• Explicit Integration of PARTIAL SET of Uncoupled
Equations (Approximate)
• Modal Response Spectrum Analysis (Approximate)
MDOF Dynamics 45
Computed Response for Piecewise Linear Loading
Time, t
Forc
e, V
(t)
t1 t3t2t0
Vt2
Vt1 1δ
Time, τ
δ = −−
V Vt tt t2 2
2 1
MDOF Dynamics 46
EXAMPLE of MDOF Response of Structure Responding to 1940 El Centro Earthquake
MDOF Dynamics 47
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
Assume WilsonDamping with 5%critical in each mode.
N-S Component of 1940 El Centro earthquakeMaximum acceleration = 0.35 g
Example # 1
10 ft
10 ft
10 ft
Example # 1 (Continued)
Form Property Matrices:
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
M =⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
−10
152 0
..
.kip s / in2
K =−
− −−
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
60 60 060 180 1200 120 300
kip / in
MDOF Dynamics 48
Example # 1 (Continued)
K MΦ ΦΩ= 2
Solve Eigenvalue Problem:
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
Ω2 2
210966
212 4=⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
−
..
.sec
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
−−−=Φ
47.2676.0300.057.2601.0644.0
000.1000.1000.1
MDOF Dynamics 49
Normalization of Modes using Φ ΦT M I=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
−−−=Φ
514.0431.0223.0534.0384.0478.0
208.0638.0749.0
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
−−−
47.2676.0300.057.2601.0644.0
000.1000.1000.1vs
MDOF Dynamics 50
MODE 1 MODE 2 MODE 3
T=1.37 sec T=0.639 sec T=0.431 sec
Example # 1 (Continued)Mode Shapes and Periods of Vibration
ω=4.58 rad/sec ω=9.83 rad/sec ω=14.57 rad/sec
MDOF Dynamics 51
Example # 1 (Continued)
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
M MT*
..
.sec /= =
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
−Φ Φ1801
2 4552310
2kip in
ωn =⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
458983
1457
.
..
/ secrad Tn =⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
13706390431
...
sec
Compute Generalized Mass:
MDOF Dynamics 52
Example # 1 (Continued)
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
Compute Generalized Loading:
)()(* tvMRtVg
TΦ−=
*
2.5661.254 ( )2.080
n gV v t⎧ ⎫⎪ ⎪=− −⎨ ⎬⎪ ⎪⎩ ⎭
MDOF Dynamics 53
Example # 1 (Continued)
Write Uncoupled (Modal) Equations of Motion:
( ) /* *y y y V t m1 1 1 1 12
1 1 12+ + =ξω ω
k3=180 k/in
u1(t)
u2(t)
u3(t)
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5 k-s2/in
m3=2.0 k-s2/in
( ) /* *y y y V t m2 2 2 2 22
2 2 22+ + =ξ ω ω
( ) /* *y y y V t m3 3 3 3 32
3 3 32+ + =ξ ω ω
)(425.10.21458.0 111 tvyyy g−=++
)(511.06.96983.0 222 tvyyy g=++
)(090.04.212457.1 333 tvyyy g−=++
MDOF Dynamics 54
Modal Participation Factors
1.425
0.511
0.090
−
1.911
0.799
0.435
−ModeModeMode
123
φ φiT
iM =10.φi, .1 10=Modal Scaling
MDOF Dynamics 55
Modal Participation Factors
1.000 0.7441.425 0.644 1.911 0.480
0.300 0.223
⎧ ⎫ ⎧ ⎫⎪ ⎪ ⎪ ⎪=⎨ ⎬ ⎨ ⎬⎪ ⎪ ⎪ ⎪⎩ ⎭ ⎩ ⎭
MDOF Dynamics 56
usingφ11 1, = usingφ φ1 1 1T M =
Effective Modal Mass
MDOF Dynamics 57
2 *n n nM P m=
Mode 1Mode 2Mode 3
3 66 81 810 64 14 950 20 5 100%
.
.
.4.50 100%
%nM Accum%
Example # 1 (Continued)Solving Modal Equation via NONLIN:
For MODE 1:( ) /* *y y y V t m1 1 1 1 1
21 1 12+ + =ξω ω
)(425.10.21458.000.1 111 tvyyy g−=++
M = 1.00 kip-sec2/in
C = 0.458 kip-sec/in
K1 = 21.0 kips/inch
Scale Ground Acceleration by Factor 1.425
MDOF Dynamics 58
MDOF Dynamics 59
MODE 1
Example # 1 (Continued)
MODE 2
MODE 3
-2.00
-1.00
0.00
1.00
2.00
0 1 2 3 4 5 6 7 8 9 10 11 12
MODAL Displacement Response Histories (From NONLIN)
-6.00
-3.00
0.00
3.00
6.00
0 1 2 3 4 5 6 7 8 9 10 11 12
-0.20
-0.10
0.00
0.10
0.20
0 1 2 3 4 5 6 7 8 9 10 11 12
Time, Seconds
T1=1.37 sec
T2=0.64
T3=0.43
Maxima
Example # 1 (Continued)
-5
-4
-3
-2
-1
0
1
2
3
4
5
0 2 4 6 8 10 12
Time, Seconds
Mod
al D
ispl
acem
ent,
Inch
es
MODE 1MODE 2MODE 3
MODAL Response Histories:
MDOF Dynamics 60
MDOF Dynamics 61
-6-4-20246
0 1 2 3 4 5 6 7 8 9 10 11 12
-6-4-20246
0 1 2 3 4 5 6 7 8 9 10 11 12
Time, Seconds
u1(t)
u2(t)
u3(t)
=0.300 x MODE 1 - 0.676 x MODE 2 + 2.47 x MODE 3
Example #1 (Continued)Compute Story Displacement Response-Histories:
-6-4-20246
0 1 2 3 4 5 6 7 8 9 10 11 12
u t y t( ) ( )= Φ
MDOF Dynamics 62
u1(t)
u2(t)
u3(t)
Example #1 (Continued)Compute Story Shear Response-Histories:
-400-200
0200400
0 1 2 3 4 5 6 7 8 9 10 11 12
-400-200
0200400
0 1 2 3 4 5 6 7 8 9 10 11 12
-400-200
0200400
0 1 2 3 4 5 6 7 8 9 10 11 12
Time, Seconds
=k2[u2(t)-u3(t)]
Example #1 (Continued)
Displacements and Forces at time of Maximum Displacements (t = 6.04 seconds)
134.8
196.9
220.4
1348
3317
5521
0
Story Shear (k) Story OTM (ft-k)
134.8 k
62.1 k
23.5 k
5.11”
2.86”
1.22”
MDOF Dynamics 63
Example # 1(Continued)Displacements and Forces at time of Maximum Shear
(t = 3.18 seconds)
38.2 k
182.7 k
124.7 k
3.91”
3.28”
1.92”
38.2
162.9
345.6
382
2111
5567
0
Story Shear (k) Story OTM (ft-k)
MDOF Dynamics 64
Modal Response Response Spectrum Method• Instead of solving the time history problem for each
mode, use a response spectrum to compute theMAXIMUM response in each mode.
• These maxima are generally NONCONCURRENT
• Combine the maximum modal responses using somestatistical technique, such as Square Root of the Sum ofthe Squares (SRSS), or Complete Quadratic Combination(CQC)
• The technique is APPROXIMATE
• It is the basis for the Equivalent Lateral Force (ELF) Method
MDOF Dynamics 65
Example #1 (Response Spectrum Method)
MDOF Dynamics 66
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Period, Seconds
Spec
tral
Dis
plac
emen
t, In
ches
1.20”
3.04”
3.47”
MODE 3T=0.431 Sec. MODE 2
T=0.639 Sec.
MODE 1T=1.37 Sec.
Displacement Response Spectrum1940 El Centro, 0.35g, 5% Damping
MODAL RESPONSE
)(425.10.21458.0 111 tvyyy g−=++
)(511.06.96983.0 222 tvyyy g=++
)(090.04.212457.1 333 tvyyy g−=++
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Period, Seconds
Spec
tral
Dis
plac
emen
t, In
ches
1.20”
3.04”
3.47”
y1 1425 347 4 94= =. * . . "y2 0511 304 155= =. * . . "y3 0090 120 0108= =. * . . "
Modal Equations of Motion Modal Maxima
Example #1 (Continued)
MDOF Dynamics 67
MDOF Dynamics 68
MODE 1
MODE 2
MODE 3
-2.00
-1.00
0.00
1.00
2.00
0 1 2 3 4 5 6 7 8 9 10 11 12
-6.00
-3.00
0.00
3.00
6.00
0 1 2 3 4 5 6 7 8 9 10 11 12
-0.20
-0.10
0.00
0.10
0.20
0 1 2 3 4 5 6 7 8 9 10 11 12
Time, Seconds
T=1.37
T=0.64
T=0.43
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Period, Seconds
Spec
tral
Dis
plac
emen
t, In
ches
1.20 x 0.090”
3.04 x 0.511”
3.47x 1.425”The Scaled ResponseSpectrum values givethe Same Modal Maximaas the Previous ResponseHistories
Example #1(Continued)
Example #1 (Continued)Computing NONCONCURRENT Story Displacements
10000 6440 300
4 9404 94031811482
.
.
..
.
..
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
=⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 1
10000 6010 676
155015500 9311048
...
....
−−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
= −−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 2
10002 570
2 4700108
01080 278
0 267
..
..
..
.−⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
= −⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 3
MDOF Dynamics 69
Example #1 (Continued)Modal Combination Techniques (For Displacement)
MDOF Dynamics 70
At time of Max. Displacement
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧=⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
++++++
80.239.460.6
267.0048.1482.1278.0931.0181.3108.0550.1940.4
Sum of Absolute Values
⎪⎭
⎪⎬⎫
⎪⎩
⎪⎨⎧
22.186.215.5
“Exact”
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧=⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
++++++
84.133.318.5
267.0048.1482.1278.0931.0181.3108.0550.1940.4
222
222
222
Square Root of the Sum of the Squares Envelope of story displacement
⎪⎭
⎪⎬⎫
⎪⎩
⎪⎨⎧
93.118.315.5
Response Spectrum Results Response History Results
Example #1 (Continued)Computing Interstory Drifts
4 940 31813181 1482
1482 0
175916991482
. .
. ..
.
.
.
−−−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪=⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 1
1550 09310931 1048
1048 0
248101171048
. ( . ). ( . )
.
.
..
− −− − −
− −
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪=−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 2
0108 02780278 02670267 0
03860545
0267
. ( . ). ..
..
.
− −− −
−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪= −⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 3
MDOF Dynamics 71
Example #1 (Continued)Computing Interstory Shears (Using Drift)
1759 601699 1201482 180
105520392668
. ( ). ( ). ( )
.
.
.
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪=⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 1
2481 600117 1201048 180
14891401886
. ( ). ( ). ( )
..
.−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪=
−
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 2
0386 600545 120
0267 180
232654
481
. ( ). ( )
. ( )
...
−⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪= −⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
MODE 3
MDOF Dynamics 72
106 149 232204 14 65 4267 189 481
220215331
2 2 2
2 2 2
2 2 2
+ ++ ++ +
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪=⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
...
Example #1 (Continued)Computing Interstory Shears: SRSS Combination
207203346
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
“Exact”135197220
⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
“Exact”38 2163346
.⎧
⎨⎪
⎩⎪
⎫
⎬⎪
⎭⎪
“Exact”
At time ofMax. Shear
At time of Max.Displacement
Envelope = Maximumper Story
MDOF Dynamics 73
Caution:
Do NOT compute story shears from the storydrifts derived from the SRSS of the storydisplacements.
Calculate the story Shears in each mode(using modal drifts) and then SRSSthe results.
MDOF Dynamics 74
Using Less than Full (Possible)Number of Natural Modes
-5
-4
-3
-2
-1
0
1
2
3
4
5
0 2 4 6 8 10 12
Time, Seconds
Mod
al D
ispl
acem
ent,
Inch
es
MODE 1MODE 2MODE 3
MODAL Response Histories:
MDOF Dynamics 75
Using Less than Full Number of Natural Modes
Time-History for MODE 1
y ty t y t y t y t y t y t y t y t y tny t y t y t y t y t y t y t y t y tny t y t y t y t y t y t
( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) (
=1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2
3 3 3 3 3 3
1 2 3 4 5 6 7 81 2 3 4 5 6 7 81 2 3 4 5 6 7 83 3 3) ( ) ( ) .... ( )y t y t y tn
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
[ ]u t y t( ) ( )= φ φ φ1 2 3
3 x nt 3 x 3 3 x nt
3 x nt
Transformation:
Time-History for DOF 1
u tu t u t u t u t u t u t u t u t u tnu t u t u t u t u t u t u t u t u tnu t u t u t u t u t u t
( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) (
=1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2
3 3 3 3 3 3
1 2 3 4 5 6 7 81 2 3 4 5 6 7 81 2 3 4 5 6 7 83 3 3) ( ) ( ) .... ( )u t u t u tn
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
MDOF Dynamics 76
Using Less than Full Number of Natural Modes
Time-History for MODE 1
y ty t y t y t y t y t y t y t y t y tny t y t y t y t y t y t y t y t y tn
( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )
=⎡
⎣⎢
⎤
⎦⎥
1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2
1 2 3 4 5 6 7 81 2 3 4 5 6 7 8
NOTE: Mode 3 NOT Analyzed
MDOF Dynamics 77
u tu t u t u t u t u t u t u t u t u tnu t u t u t u t u t u t u t u t u tnu t u t u t u t u t u t
( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) .... ( )( ) ( ) ( ) ( ) ( ) (
=1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2
3 3 3 3 3 3
1 2 3 4 5 6 7 81 2 3 4 5 6 7 81 2 3 4 5 6 7 83 3 3) ( ) ( ) .... ( )u t u t u tn
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
3 x nt 3 x 2 2 x nt
3 x nt
[ ]u t y t( ) ( )= φ φ1 2
Transformation:
Time-History for DOF 1
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧=⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
++++++
82.131.318.5
84.133.318.5
267.0048.1482.1278.0931.0181.3108.0550.1940.4
222
222
222
Square Root of the Sum of the Squares
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧=⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
++++++
53.2112.449.6
80.239.460.6
267.0048.1482.1278.0931.0181.3108.0550.1940.4
Sum of Absolute Values
Envelope of StoryDisplacement
Using Less than Full Number of Natural Modes(Modal Response Spectrum Technique)
⎪⎭
⎪⎬⎫
⎪⎩
⎪⎨⎧
93.118.315.5
MDOF Dynamics 78
2 modes3 modes
ASCE 7 allow an ApproximateModal Analysis Technique Called the
“EQUIVALENT LATERAL FORCE PROCEDURE”
• Empirical Period of Vibration• Smoothed Response Spectrum• Compute Total Base Shear V as if SDOF• Distribute V Along Height assuming
“Regular” Geometry• Compute Displacements and Member
Forces using Standard Procedures
MDOF Dynamics 79
EQUIVALENT LATERAL FORCE PROCEDURE
• Method is based on FIRST MODE response
• Higher modes can be included empirically• Has been calibrated to provide a
reasonable estimate of the envelopeof story shear, NOT to provide accurate estimates of story force
• May result in overestimate of overturningmoment.
MDOF Dynamics 80
EQUIVALENT LATERAL FORCE PROCEDURE• Assume first mode effective mass = Total Mass = M = W/g
• Use Response Spectrum to obtain Total Acceleration at T1
Acceleration, g
Sa1
Period, Sec.T1
WSg
WgSMgSV aaaB 111 )()( ===
MDOF Dynamics 81
MDOF Dynamics 82
dr
EQUIVALENT LATERAL FORCE PROCEDURE
Assume Linear First Mode Response
gWtd
hhtf x
rx
x21)()( ω=
hx
Wxfx
i
nstories
ii
rnstories
iiB Wh
hgtdtftV ∑∑
==
==1
21
1
)()()( ωh
Portion of base shear applied to story i
∑=
= nstories
iii
xx
B
x
Wh
WhtVtf
1
)()(
ELF EXAMPLE
k3=180 k/in
k1=60 k/in
k2=120 k/in
m1=1.0 k-s2/in
m2=1.5
m3=2.0
MDOF Dynamics 83
3h
RecallT1=1.37 seconds
ELF EXAMPLETotal Weight = M x g = (1.0 + 1.5 + 2.0) 386.4 = 1738 kips
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Period, Seconds
Spec
tral
Dis
plac
emen
t, In
ches
3.47”
1.37sec
Spectral Acceleration = ω2SD = (2π/1.37)2 x 3.47 = 72.7 in/sec2
= 0.188g
Base Shear = SaW = 0.188 x 1738 = 327 kips
MDOF Dynamics 84
ELF EXAMPLE (Story Forces)
3386(3 ) 0.381 0.375(327) 125
386(3 ) 579(2 ) 722( ) Bhf V kips
h h h= = = =
+ +
125 k
MDOF Dynamics 85
W=386 k
3h
W=579 k
W=722 k
125 k
77 k
327 k
125 k(220 k)
250 k(215 k)
Story Shear (k)
327 k(331 k)
ELF EXAMPLE (Story Displacements)Units=inches
5.983.891.82
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
ELFTime History(Envelope)
Modal ResponseSpectrum
⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
84.133.318.55.15
3.181.93
⎧ ⎫⎪ ⎪⎨ ⎬⎪ ⎪⎩ ⎭
MDOF Dynamics 86
ELF EXAMPLE (Summary)
• ELF procedure gives GOOD CORRELATIONwith base shear (327 kips ELF vs 331 kps ModalResponse Spectrum)
• ELF story force distribution is NOT as GOOD.ELF underestimates shears in upper stories.
• ELF gives reasonable correlation with displacements
MDOF Dynamics 87
ASCE 7-05 ELF APPROACH
• Uses Empirical Period of Vibration• Uses Smoothed Response Spectrum• Has Correction for Higher Modes• Has Correction for Overturning Moment• Has Limitations on Use
MDOF Dynamics 88
Approximate Periods of Vibrationxnta hCT =
Ct = 0.028, x = 0.8 for Steel Moment FramesCt = 0.016, x = 0.9 for Concrete Moment FramesCt = 0.030, x = 0.75 for Eccentrically Braced FramesCt = 0.020, x = 0.75 for All Other SystemsNote: Buildings ONLY!
NTa
1.0=For Moment Frames < 12 stories in height, minimumstory height of 10 feet. N= Number of Stories.
MDOF Dynamics 89
Adjustment Factor on Approximate Period
computedua TCTT ≤=
SD1 Cu> 0.40g 1.4
0.30g 1.40.20g 1.50.15g 1.6< 0.1g 1.7
Applicable ONLY if Tcomputed comes from a “properlysubstantiated analysis”
MDOF Dynamics 90
ASCE 7 Smoothed Design Acceleration Spectrum(for use with ELF)
“Long Period” Acceleration2
1
“Short Period” Acceleration
2
1
3 3
WCV S= ⎟⎠⎞
⎜⎝⎛
=
IR
SC DSS
⎟⎠⎞
⎜⎝⎛
=
IRT
SC DS
1
Spe
ctra
l Res
pons
e A
ccel
erat
ion,
Sa
SDS
SD1
Varies
T = 1.0T=0.2 Period, T
MDOF Dynamics 91
R is the Response Modification Factor which is afunction of system inelastic behavior. This is coveredin the topic on inelastic behavior. For now, we willUse R=1 which implies linear elastic behavior.
I is the Importance Factor. This depends on the Seismic Use Group. I=1.5 for Essential Facilities,1.25 for important high occupancy structures,and 1.0 for normal structures. This is coveredin the topic on the ASCE 7 Provisions. For now, we will use I=1.
MDOF Dynamics 92
EQUIVALENT LATERAL FORCE PROCEDUREHigher Mode Effects
+ =
Combined1st Mode 2nd Mode
MDOF Dynamics 93
Distribution of Forces along Height
VCF vxx =
∑=
= n
i
kii
kxx
vx
hw
hwC
1
MDOF Dynamics 94
k accounts for Higher Mode Effects
0.5 2.5
2.0
1.0
Period, Sec
k
k = 0.5T + 0.75(sloped portion only)
k=2k=1
MDOF Dynamics 95
ELF EXAMPLE (Story Forces)V=327 kips T=1.37 seconds k=0.5(1.37)+0.75=1.435
146 k
120 k
61 k
327 k
W=386 k (125 k)
(125 k)
(77 k)
MDOF Dynamics 96
146 k
266 k
Story Shear (k)
327 k
W=579 k
3hW=722 k
ASCE 7 ELF Limitations
• Applicable ONLY to “Regular” structures with Tless than 3.5 Ts Note:(Ts=SD1/SDS)
Adjacent story stiffness does not vary more than 30%
Adjacent story strength does not vary more than 20%
Adjacent story masses does not vary more than 50%
If violated, must use more advanced analysis (typicallyModal response spectrum analysis)
MDOF Dynamics 97
ASCE 7 ELF Other considerations that effect loading
• Orthogonal Loading Effects• Redundancy • Accidential Torsion• Torsional Amplification• P-Delta Effects
• Importance Factor• Ductility and Overstrength
MDOF Dynamics 98