Class Note for Structural Analysis 2 Fall Semester, 2013
Hae Sung Lee, Professor Dept. of Civil and Environmental Engineering Seoul National University Seoul, Korea
Contents Chapter 1 Slope Deflection Method
1
1.0 Comparison of Flexibility Method and Stiffness Method………………………… 2 1.1 Analysis of Fundamental System………………………………………………..... 5 1.2 Analysis of Beams………………………………………………………………… 8 1.3 Analysis of Frames………………………………………………………………... 17 Chapter 2 Iterative Solution Method & Moment Distribution Method 2.1 Solution Method for Linear Algebraic Equations………………………………… 2.2 Moment Distribution Method……………………………………………………... 2.3 Example - MDM for a 4-span Continuous Beam…………………………………. 2.4 Direct Solution Scheme by Partitioning…………………………………………... 2.5 Moment Distribution Method for Frames………………………………………… Chapter 3 Energy Principles
32 33 37 42 44 45 47
3.1 Spring-Force Systems……………………………………………………………... 48 3.2 Beam Problems……………………………………………………………………. 49 3.3 Truss problems……………………………………………………………………. 53 Chapter 4 Matrix Structural Analysis 4.1 Truss Problems…………………………………………………………………… 4.2 Beam Problems…………………………………………………………………… 4.3 Frame Problems…………………………………………………………………... Chapter 5 Buckling of Structures 5.0 Stability of Structures……………………………………………………………... 5.1 Governing Equation for a Beam with Axial Force………………………………... 5.2 Homogeneous Solutions…………………………………………………………... 5.3 Homogeneous and Particular solution…………………………………………….. 5.4 Energy Method……………………………………………………………………. 5.5 Approximation with the Homogeneous Beam Solutions…………………………. 5.6 Nonlinear Analysis of Truss……………………………………………………….
56 57 68 76 79 80 81 82 85 86 89 91
School of Civil, Urban & Geosystem Eng., SNU
1
Chapter 1 Slope Deflection Method
A
B
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
2
School of Civil, Urban & Geosystem Eng., SNU 2.0 Comparison of Flexibility Method and Stiffness Method Flexibility Method
Stiffness Method P
k1 z
P k2
k1
Remove redundancy (Equilibrium)
z
k2
Compatibility
P P
X δ1 = δ 2 = δ
z
z
Compatibility
Equilibrium
δ1 = δ 2
k1 X P− X = →X = P k1 + k 2 k1 k2
k1δ + k2δ = P → δ =
P k1 + k 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
3
School of Civil, Urban & Geosystem Eng., SNU Flexibility Method
L/2
L/2
P B
A
z
Stiffness Method
EI
EI
L
L z
Remove redundancy (Equilibrium)
B
A
C
P EI
EI
L
L
C
Compatibility
3PL 16
+
θB 1
+
θ BA = θ BC = θ B z
z
Compatibility δB0 =
2
L 1 PL PL 2L (1 + ) ×1 = , δ BB = 6 EI 2 4 16 EI 3EI
δ B 0 + δ BB M B = 0 → M B = −
δB0 3PL =− δ BB 32
Equilibrium
3EI 3PL f B B , M BC = 0 , M BA = M BC = θB L 16 ∑ M B =M BAf + M BCf + M BAB + M BCB = 0 →
f M BA =−
−
3PL 6 EI PL2 + θB = 0 → θB = 16 L 32 EI
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
4
School of Civil, Urban & Geosystem Eng., SNU Flexibility Method
Stiffness Method
1. Release all redundancies.
1.
Fix all Degrees of Freedom.
2. Calculate displacements induced by external loads at the released redundancies.
2.
Calculate fixed end forces induced by external loads at the fixed DOF.
3. Apply unit loads and calculate displacements at the released redundancies.
3.
Apply unit displacements and calculate member end forces at the DOFs.
4. Construct the flexibility equation by superposing the displacement based on the compatibility conditions.
4.
Construct the stiffness equation by superposing the member end forces based on the equilibrium equations.
5. Solve the flexibility equation.
5.
Solve the stiffness equation.
6. Calculate reactions and other quantities as needed.
6.
Calculate reactions and other quantities as needed.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
5
School of Civil, Urban & Geosystem Eng., SNU 2.1 Analysis of Fundamental System 2.1.1 End Rotation
A
B
θA z
Flexibility Method i) θ B = 0 MA
MB
L L MA + M B = −θ A 3EI 6 EI → M A = − 4 EI θ A , M B = 2 EI θ A L L L L MA + MB = 0 6 EI 3EI ii) θ A = 0 MA = −
2 EI 4 EI θB , M B = θB L L
z
Sign Convention for M :Counterclockwise “+”
z
θ A ≠ 0 , θB ≠ 0
4 EI 2 EI θA + θB L L 2 EI 4 EI MB = θA + θB L L MA =
2.1.2 Relative motion of joints
∆
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
6
School of Civil, Urban & Geosystem Eng., SNU z
Flexibility Method
∆
L L ∆ MA + MB = − 3EI 6 EI L L L ∆ MA + MB = 6 EI 3EI L
z
=−
6 EI ∆ 6 EI ∆ , MB = L L L L
or in the new sign convention : M A =
6 EI ∆ 6 EI ∆ , MB = L L L L
→M
A
Final Slope-Deflection Equation 4 EI 2 EI 6 EI θA + θB + L L L 2 EI 4 EI 6 EI MB = θA + θB + L L L
MA =
z
∆ L ∆ L
In Case an One End is Hinged 4 EI 2 EI 6 EI θA + θB + L L L 2 EI 4 EI 6 EI MB = θA + θB + L L L
MA =
EI ∆ 2 EI 3EI ∆ =0→ θ A = − θB − L L L L L ∆ 3EI 3EI ∆ = θB + L L L L
2.1.3 Fixed End Force z
Both Ends Fixed
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
7
School of Civil, Urban & Geosystem Eng., SNU z
One End Hinged
MB
MA +
MA/2
MA 3 MA 2 z
3 MA 2
Ex.: Uniform load case with a hinged left end
M Bf = −
3qL2 qL2 qL2 qL2 − =− =− , 12 24 24 8
M Af = 0
2.1.4 Joint Equilibrium
Ffixed Fjoint
Fmember
Joint i
− ∑ F fixed −∑ Fmember + ∑ F joint = 0
or
∑F
fixed
+ ∑ Fmember = ∑ F joint
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
8
School of Civil, Urban & Geosystem Eng., SNU 2.2 Analysis of Beams 2.2.1 A Fixed-fixed End Beam
P B
A
C EI
a z
DOF : θ B , ∆B
z
Analysis
b
i) All fixed : No fixed end forces ii) θ B ≠ 0 , ∆ B = 0
2 EI 4 EI 4 EI 2 EI 1 1 θ B , M BA = θ B , M 1BC = θ B , M CB = θB a a b b 6 EI 6 EI 1 1 1 = −VAB = 2 θ B , − VBC = VCB = 2 θB a b
M 1AB = 1 VBA
2 EI θB a
4 EI θB a
6 EI θB a2
6 EI θB a2
4 EI θB b
6 EI θB b2
2 EI θB b
6 EI θB b2
iii) θ B = 0 , ∆ B ≠ 0
6 EI 6 EI 6 EI 6 EI 2 2 2 = 2 ∆ B , M BC = − 2 ∆B ∆ B , M BA = − 2 ∆ B M CB 2 a b a b 12 EI 12 EI 2 2 2 = −VAB = 3 ∆ B , VBC = −VCB = 2 ∆B a b
2 M AB = 2 VBA
6 EI ∆B a2
12 EI ∆B a3
6 EI ∆B a2
12 EI ∆B a3
6 EI ∆B b2
12 EI ∆B b3
6 EI ∆B b2
12 EI ∆B b3
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
9
School of Civil, Urban & Geosystem Eng., SNU z
Construct the Stiffness Equation
1 1 1 1 1 1 2 2 = 0 → M BA + M BC + M BA + M BC = 0 → 4 EI ( + )θ B + 6 EI ( 2 − 2 )∆ B = 0 a b a b 1 1 1 1 6 EI ( 2 − 2 )θ B + 12 EI ( 3 + 3 )∆ B = P ∑VBi = P → VBA1 + VBC1 + VBA2 + VBC2 = P → a b a b 2 2 3 3 (b − a )a b ab θB = − P , ∆B = P 3 2 EIl 3EIl 3
∑M
i B
ab 2 2 EI 6 EI θB + 2 ∆ B = 2 P , a a l a 2b 2 EI 6 EI = θB − 2 ∆ B = − 2 P b b l
2 M AB = M 1AB + M AB = 1 2 M CB = M CB + M CB
2.2.2 Analysis of a Two-span Continuous Beam (Approach I)
q
qL B
A
2EI
EI L
z
DOF : θ B , θC
z
Analysis
C
L
i) Fix all DOFs and Calculate FEM. f = M AB
qL2 qL2 qL2 qL2 f f f , M BA =− , M BC = , M CB =− 8 12 12 8
ii) θ B ≠ 0 , θC = 0 M 1AB =
2 EI 4 EI 8 EI 4 EI 1 1 1 θ B , M BA = θ B , M BC = θ B , M CB = θB L L L L
iii) θ B = 0 , θC ≠ 0 2 M BC =
z
4 EI 8 EI 2 θC , M CB = θC L L
Construct the Stiffness Equation
∑M
i B
f f 1 1 2 = 0 → M BA + M BC + M BA + M BC + M BC =0→
qL2 EI EI + 12 θ B + 4 θ C = 0 L L 24
∑M
i C
f 1 2 = 0 → M CB + M CB + M CB =0→
−
qL2 EI EI + 4 θ B + 8 θC = 0 L L 8L
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
10
School of Civil, Urban & Geosystem Eng., SNU θB = − z
Member End Forces f M AB = M AB + M 1AB =
qL2 2 EI 3 + θ B = qL2 12 L 48
1 f M BA = M BA + M BA =−
qL2 4 EI 1 + θ B = − qL2 12 L 8
1 2 f M BC = M BC + M BC + M BC =
qL2 8 EI 4 EI 1 + θB + θC = qL2 8 L L 8
1 2 f M CB = M CB + M CB + M CB =−
z
qL3 qL3 , θC = 96 EI 48EI
qL2 4 EI 8 EI + θB + θC = 0 L L 8
Various Diagram
- Freebody Diagram
1 2 qL 16
1 2 qL 8
5 qL 8
9 qL 16
7 qL 16
3 qL 8
19 qL 16 - Moment Diagram
1 2 qL 16
17 2 qL 512
1 2 qL 8
3 2 qL 16
7 L 16
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
11
School of Civil, Urban & Geosystem Eng., SNU 2.2.3 Analysis of a Two-span Continuous Beam (Approach II)
q
qL B
A EI L z
DOF : θ B
z
Analysis
2EI
C
L
i) Fix all DOFs and Calculate FEM. f = M AB
qL2 qL2 qL2 1 qL2 3qL2 f f , M BA =− , M BC = + = 12 12 8 2 8 16
ii) θ B ≠ 0 M 1AB = z
Construct Stiffness Equation
∑M z
2 EI 4 EI 6 EI 1 1 θ B , M BA = θ B , M BC = θB L L L
B
1 1 f f = 0 → M BA + M BC + M BA + M BC =0
qL2 3qL2 EI EI qL2 + + 4 θB + 6 θB = 0 → θB = − 12 16 L L 96 EI
Member End Forces
qL2 2 EI 3 + θ B = qL2 12 L 48 2 qL 4 EI 1 1 f M BA = M BA + M BA =− + θ B = − qL2 12 L 8 2 3qL 6 EI 1 1 f M BC = M BC + M BC = + θ B = qL2 16 L 8 f M AB = M AB + M 1AB =
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
12
School of Civil, Urban & Geosystem Eng., SNU 2.2.4 Analysis of a Beam with an Internal Hinge (4 DOFs System) q
C
B
A EI l z
DOF : θ B , θCL , θCR , ∆C
z
Analysis
D EI
EI
l
l
i) All fixed M
f AB
ql 2 ql 2 f = , M BA = − 12 12
ii) θ B ≠ 0
M 1AB =
2 EI 4 EI 2 EI 6 EI 1 1 1 1 θ B , M BA = M BC = θ B , M CB = θ B , VCB = 2 θB l l l l
iii) θCL ≠ 0
2 M BC =
2 EI L 4 EI L 6 EI 2 2 θC , M CB = θC , VCB = 2 θCL l l l
iv) θCR ≠ 0
3 M CD =
4 EI R 2 EI R 6 EI 3 3 θC , M DC = θC , VCD = − 2 θCL l l l
v) ∆ C ≠ 0
4 4 M BC = M CB =
6 EI 6 EI 12 EI 4 4 4 4 ∆ C , M CD = M DC = − 2 ∆ C , VCD = VCB = 2 ∆C 2 l l l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
13
School of Civil, Urban & Geosystem Eng., SNU z
Construct Stiffness Equation ql 2 EI EI + 8 θ B + 2 θCL + l l 12 i EI EI 2 θ B + 4 θCL + ∑i M 2i = 0 → l l
∑M
i 1
=0→−
∑M
i 3
=0→
EI ∆C = 0 l2 EI 0 + 6 2 ∆C = 0 l EI EI 4 θCR − 6 2 ∆ C = 0 l l EI EI EI EI 6 2 θ B + 6 2 θCL − 6 2 θCR + 24 3 ∆ C = 0 l l l l
i
∑V
i 4
=0→
i
z
0
+6
Elimination of θCL and θCR
- 2nd and 3rd equation 2
EI L EI EI EI EI θC = −( θ B + 3 2 ∆ C ) , 2 θCR = 3 2 ∆ C l l l l l
- 1st equation ql 2 EI EI EI − + 8 θ B + 2 θ CL + 6 2 ∆ C = 12 l l l 2 ql EI EI EI EI ql 2 EI EI − + 8 θB − ( θB + 3 2 ∆C ) + 6 2 ∆C = 0 → − +7 θB + 3 2 ∆C = 0 12 12 l l l l l l - 4th equation
EI EI EI EI θ B + 6 2 θ CL − 6 2 θ CR + 24 3 ∆ C = 2 l l l l EI EI EI EI EI EI EI 6 2 θ B − 3( 2 θ B + 3 3 ∆ C ) − 3(3 3 ∆ C ) + 24 3 ∆ C = 0 → 3 2 θ B + 6 3 ∆ C = 0 l l l l l l l
6
2.2.5 Analysis of a Beam with an Internal Hinge (2 DOFs System) q
C
B
A EI l z
DOF : θ B , ∆C
z
Analysis
D EI
EI
l
l
i) All fixed f M AB =
ql 2 ql 2 f , M BA =− 12 12
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
14
School of Civil, Urban & Geosystem Eng., SNU ii) θ B ≠ 0
M 1AB =
2 EI 4 EI 3EI 3EI 1 1 1 1 θ B , M BA = θ B , M BC = θ B , VBC = −VCB = − 2 θB l l l l
iii) ∆ C ≠ 0
2 2 M BC = − M DC =
z
3EI 3EI 3EI 2 2 2 2 ∆ C , VBC = −VCB = − 2 ∆ C , VCD = −VDC = 2 ∆C 2 l l l
Construct the Stiffness Equation
∑M
=0→−
i 1
i
∑V
i 2
=0→
i
ql 2 EI EI + 7 θB + 3 2 ∆C = 0 12 l l EI EI 3 2 θB + 6 3 ∆C = 0 l l
θB =
ql 3 ql 4 , ∆C = − 66 EI 132 EI
2 EI R 3EI ∆ C 3 ∆C 3 ql 3 θC = − (− ) → θCR = =− l l l 2 l 264 EI 3 3 1 3∆ ql 3 ql ql 3 θCL = − θ B − =− + = 2 2L 132 EI 2 132 EI 264 EI
2.2.6 Beam with a Spring Support q
EI l z
C
B
A
D EI l
k
EI l
Analysis
i) All fixed f = M AB
ql 2 ql 2 f , M BA =− 12 12
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
15
School of Civil, Urban & Geosystem Eng., SNU ii) θ B ≠ 0
M 1AB =
2 EI 4 EI 3EI 3EI 1 1 1 1 θ B , M BA = θ B , M BC = θ B , − VBC = VCB = 2 θB l l l l
iii) ∆ C ≠ 0
k∆C 3EI ∆C l2 3EI 3EI 2 2 = − 2 ∆ C , VCD = −VDC = 2 ∆ C , VS2 = k∆ C l l
2 2 M BC = − M DC = 2 2 VBC = −VCB
z
Construct the Stiffness Equation
∑ M 1i = 0 → − i
∑V
i 2
=0→
i
θB = z
ql 2 EI EI + 7 θB + 3 2 ∆C = 0 12 l l EI EI 3 2 θ B + (6 3 + k ) ∆ C = 0 l l
1+ α ql 3 1 ql 4 , ∆C = − (1 + 14α / 11) 132 EI 1 + 14α / 11 66 EI
6 EI l3
α→0
θB = z
where k = α
ql 3 ql 4 , ∆C = − 66 EI 132 EI
α→∞
θB =
ql 3 , ∆C = 0 84 EI
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
16
School of Civil, Urban & Geosystem Eng., SNU 2.2.7 Support Settlement
l
l
B
A EI
C
EI δ
z
DOF : θ B
z
Analysis
i) All fixed f M BA =
6 EI δ 3EI δ f , M BC =− l l l l
ii) θ B ≠ 0 1 M BA =
z
4 EI 3EI 1 θ B , M BC = θB l l
Construct the Equilibrium Equation
∑M i
i 1
=0→6
3δ EI δ EI δ EI EI −3 + 4 θB + 3 θB = 0 → θ B = − l l l l l l 7l
2.2.8 Temperature Change A
B T1 T2
θA = z
α(T2 − T1 ) α(T2 − T1 ) l , θB = − l 2h 2h
Fixed End Moment
4 EI θA + L 2 EI MB = θA + L MA =
2 EI α(T2 − T1 ) EI θB = L h 4 EI α(T2 − T1 ) EI θB = − L h
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
17
School of Civil, Urban & Geosystem Eng., SNU 2.3 Analysis of Frames 2.3.1 A Portal Frame without Sidesway
P
l/2 B
C EI2
z
DOF : θ B , θC
z
Analysis
EI1
EI1
A
D
i) All fixed Pl Pl 0 0 = , M CB =− M BC 8 8 ii) θ B ≠ 0 M 1AB =
2 EI1 4 EI1 1 θ B , M BA = θB l l
1 M BC =
4 EI 2 2 EI 2 1 θ B , M CB = θB l l
iii) θC ≠ 0
z
2 M BC =
2 EI 2 4 EI 2 2 θC , M CB = θC l l
2 M CD =
4 EI1 2 EI1 2 θC , M DC = θC l l
Construct the Stiffness Equation
∑M
i B
=0→
4 EI1 4 EI 2 2 EI 2 Pl +( + )θ B + θC = 0 8 l l l
4 EI1 4 EI 2 Pl 2 EI 2 + θB + ( + ) θC = 0 8 l l l 1 Pl 2 − θ B = θC = 4 EI1 + 2 EI 2 8
∑M
i C
=0→−
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
18
School of Civil, Urban & Geosystem Eng., SNU z
Member End Forces
M AB =
2 EI1 2 EI1 Pl θB = − 4 EI1 + 2 EI 2 8 l
M BA =
4 EI1 4 EI1 Pl θB = − 4 EI1 + 2 EI 2 8 l
M BC =
2 EI 2 4 EI1 Pl 4 EI 2 Pl + θB + θC = 8 4 EI1 + 2 EI 2 8 l l
M CB = −
z
4 EI 2 4 EI1 Pl 2 EI 2 Pl + θB + θC = − 8 4 EI1 + 2 EI 2 8 l l
M CD =
4 EI1 4 EI1 Pl θC = 4 EI1 + 2 EI 2 8 l
M DC =
2 EI1 2 EI1 Pl θC = 4 EI1 + 2 EI 2 8 l
In case EI1 = EI 2
M AB = −
Pl Pl Pl Pl Pl , M BA = − , M BC = , M CB = − , M CD = 24 12 12 12 12
, M DC =
Pl 24
2.3.2 A Portal Frame without Sidesway – hinged suppoorts
P
l/2 B
C EI2
z
DOF : θ B , θC
z
Analysis
EI1
EI1
A
D
i) All fixed 0 = M BC
Pl Pl 0 , M CB =− 8 8
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU
19
ii) θ B ≠ 0 3EI1 θB l 4 EI 2 2 EI 2 1 = θ B , M CB = θB l l
1 M BA = 1 M BC
iii) θC ≠ 0 2 EI 2 4 EI 2 2 θC , M CB = θC l l 3EI1 = θC l
2 M BC = 2 M CD
z
Construct the Stiffness Equation
∑M
i B
=0→
Pl 3EI1 4 EI 2 2 EI 2 +( + )θ B + θC = 0 l l l 8
∑M
i C
=0→−
3EI1 4 EI 2 Pl 2 EI 2 + θB + ( + ) θC = 0 8 l l l
− θ B = θC = z
1 Pl 2 3EI1 + 2 EI 2 8
Member End Forces M AB = 0 3EI1 3EI1 Pl θB = − M BA = 3EI1 + 2 EI 2 8 l 2 EI 2 3EI1 Pl 4 EI 2 Pl + θB + θC = M BC = 8 3EI1 + 2 EI 2 8 l l 4 EI 2 3EI1 Pl 2 EI 2 Pl θB + θC = − M CB = − + 8 3EI1 + 2 EI 2 8 l l 3EI1 3EI1 Pl θC = M CD = 3EI1 + 2 EI 2 8 l M DC = 0
z
In case of EI1 = EI 2 M AB = 0 , M BA = −
3 3 3 3 Pl , M BC = Pl , M CB = − Pl , M CD = Pl , M DC = 0 40 40 40 40
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
20
School of Civil, Urban & Geosystem Eng., SNU 2.3.3 A Frame with an horizontal force
B
C
P
EI
A z
DOF : θ B , ∆
z
Analysis
i) All fixed : None fixed end moment ii) θ B ≠ 0 2 EI 4 EI 1 θ B , M BA = θB l l 3EI 6 EI 1 = θ B , VBA = 2 θB l l
M 1AB = 1 M BC
iii) ∆ ≠ 0 6 EI 6 EI 2 ∆ , M BA = 2 ∆ 2 l l 12 EI = 3 ∆ l
2 = M AB 2 VBA
z
Construct the stiffness equation
∑M ∑V
i
i B
=0→ (
4 EI 3EI 6 EI + )θ B + 2 ∆ = 0 l l l 6 EI 12 EI θB + 3 ∆ = P 2 l l
=P→
Pl 2 7 Pl 3 ,∆ = θB = − 8EI 48 EI z
Member end forces
M AB =
2 EI 6 EI 5 4 EI 6 EI 3 3EI 3 θ B + 2 ∆ = Pl , M BA = θ B + 2 ∆ = Pl , M BC = θ B = − Pl 8 8 8 l l l l l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
21
School of Civil, Urban & Geosystem Eng., SNU
2.3.4 A Portal Frame with an Unsymmetrical Load
a
P
B
A z
DOF : θ B , θC , ∆
z
Analysis
C
D
i) All fixed 0 = M BC
Pab 2 Pa 2b 0 , = − M CB l2 l2
ii) θ B ≠ 0 2 EI 4 EI 1 θ B , M BA = θB l l 4 EI 2 EI 6 EI 1 1 = θ B , M CB = θ B , VBA = 2 θB l l l
M 1AB = 1 M BC
iii) θC ≠ 0 2 EI 4 EI 2 θC , M CB = θC l l 4 EI 2 EI 6 EI 2 2 = θC , M DC = θC , VCD = 2 θC l l l
2 M BC = 2 M CD
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
22
School of Civil, Urban & Geosystem Eng., SNU iv) ∆ ≠ 0 6 EI 6 EI 3 3 ∆ , M CD = M DC = 2 ∆ , 2 l l 12 EI = 3 ∆ l
3 3 = M BA = M AB 3 3 = VCD VBA
z
Construct the Stiffness Equation
∑ M Bi = 0 →
Pab 2 8EI 2 EI 6 EI + θB + θc + 2 ∆ = 0 2 l l l l
∑ M Ci = 0 → −
Pa 2b 2 EI 8 EI 6 EI + θB + θc + 2 ∆ = 0 2 l l l l
∑V
i
6 EI 6 EI 24 EI θ B + 2 θc + 3 ∆ = 0 2 l l l
=0→
l ∆ = − ( θ B + θC ) 4 Pa 2b 13EI EI + θB + θc = 0 2 l 2l 2l Pab 2 EI 13EI − 2 + θB + θc = 0 2l 2l l 1 Pab (a + 13b) θB = − l 84 EI 1 Pab (13a + b) θC = l 84 EI 1 Pab ∆= (b − a) 28 EI
a=
l 4
2.3.5 A Portal Frame with a Bracing (Vertical Load)
a B
P C
EI = 0, EA ≠ 0
z
DOF : θ B , θC , ∆
z
Analysis
A
D
i) All fixed 0 = M BC
Pab 2 Pa 2b 0 , = − M CB l2 l2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
23
School of Civil, Urban & Geosystem Eng., SNU ii) θ B ≠ 0 2 EI 4 EI 1 θ B , M BA = θB l l 4 EI 2 EI 1 1 M BC = θ B , M CB = θB l l 6 EI 1 VBA = 2 θB l M 1AB =
iii) θC ≠ 0 2 EI 4 EI 2 θC , M CB = θC l l 4 EI 2 EI 2 2 M CD = θC , M DC = θC l l 6 EI 2 VCD = 2 θC l iv) ∆ ≠ 0 2 M BC =
6 EI M = M = 2 ∆, l 6 EI 3 3 M CD = M DC = 2 ∆, l 12 EI 3 3 VBA = VCD = 3 ∆ l 3 AB
3 BA
ABD = z
∆ ∆ 2
EA ∆ EA ∆ 1 EA ∆ (C) → VBD = = VBD = 2l 2 2 2l 2 2l 2
Construct the Stiffness Equation
Pab 2 8EI 2 EI 6 EI + θB + θc + 2 ∆ = 0 2 l l l l 2 Pa b 2 EI 8 EI 6 EI ∑ M Ci = 0 → − l 2 + l θB + l θc + l 2 ∆ = 0 6 EI 6 EI 24 EI θ B + 2 θc + 3 (1 + α)∆ = 0 ∑V i = 0 → 2 l l l EAl 2 1 l ∆=− (θ B + θC ) , α = 1+ α 4 48 2 EI
∑ M Bi = 0 →
z
Solution for b = 3a θB = −
40 + 52α Pl 2 16 + 28α Pl 2 3 Pl 3 , θC = − , ∆= 256(7 + 10α ) EI 256(7 + 10α) EI 128(7 + 10α) EI
For a w × h rectangular section and l = 20h , α = 50 2 . Pl 2 Pl 2 Pl 3 θ B = −0.0203 , θC = 0.0109 ∆ = 0.3282 × 10 −4 EI EI EI
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
24
School of Civil, Urban & Geosystem Eng., SNU z
Performance
with Bracing ( α = 50 2 )
Response
w/o bracing (α = 0)
Ratio(%)
θB (× Pl 2 / EI )
-0.0203
-0.0223
91.03
θC (× Pl 2 / EI )
0.0109
0.0089
122.47
∆ (× Pl 3 / EI )
0.3282×10-4
0.0033
0.99
ΜΑΒ (Pl)
-0.0404
-0.0248
162.90
ΜΒΑ (Pl)
-0.0810
-0.0694
116.71
ΜCD (Pl)
0.0438
0.0554
79.06
ΜDC (Pl)
0.0220
0.0376
58.51
ΜP (Pl)
0.1158
0.1216
95.23
ABD (P)
0.0788
Pmax (Pall)*
0.0720
0.0685
105.1
Pmax/vol.
0.0163
0.0228
71.5
-
-
*) Pall = σ all wh , M all = Pall h / 6 Unbalanced shear force in the columns =
6 EI (θ B + θC ) = 0.0564 P l2
The bracing carries 99 % of the unbalanced shear force between the two columns.
2.3.6 A Portal Frame with a Bracing (Horizontal Load)
B
P
C
EI = 0, EA ≠ 0
A z
DOF : θ B , θC , ∆
z
Analysis
D
i) All fixed: No fixed end forces ii)-iv) the same as the previous case
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
25
School of Civil, Urban & Geosystem Eng., SNU z
Construct the Stiffness Equation
8EI 2 EI 6 EI θB + θc + 2 ∆ = 0 l l l EI EI EI 2 8 6 ∑ M Ci = 0 → l θB + l θc + l 2 ∆ = 0 6 EI 6 EI 24 EI ∑V i = 0 → l 2 θB + l 2 θc + l 3 (1 + α)∆ = P
∑M
i B
=0→
5 5 θ B = θC , ∆ = − θ B l = − θC l 3 3 z
Solution 1 Pl 2 5 Pl 3 θ B = θC = − , ∆= (28 + 40α) EI 3(28 + 40α) EI
For α = 50 2 , θ B = θC = −0.3501 × 10− 3 z
Pl 2 Pl 3 , ∆ = 0.5835 × 10− 3 EI EI
Performance
Response
with Bracing ( α = 50 2 )
w/o bracing (α = 0)
Ratio(%)
θB (× Pl 2 / EI )
− 0.3501 × 10−3
− 0.3571 × 10−1
0.98
θC (× Pl 2 / EI )
− 0.3501 × 10−3
− 0.3571 × 10−1
0.98
∆ (× Pl 3 / EI )
0.5835 × 10−3
0.5952 × 10−1
0.98
ΜΑΒ (Pl)
0.2801 × 10−2
0.2857
0.98
ΜΒΑ (Pl)
0.2101 × 10−2
0.2143
0.98
ΜCD (Pl)
0.2101 × 10−2
0.2143
0.98
ΜDC (Pl)
0.2801 × 10−2
0.2857
0.98
ABD (P)
1.4004
-
-
Pmax(Pall)*
0.7141
0.0292
2448
Pmax/vol.
0.1617
0.0097
1670
*) Governed by ABD for the structure with bracing, and by MDC for the structure without bracing. Pall = σ all wh , M all = Pall h / 6 The bracing carries about 99% of the external horizontal load.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
26
School of Civil, Urban & Geosystem Eng., SNU 2.3.7 A Portal Frame with a Spring
a
P
B
C k
A z z
D
DOF : θ B , θC , ∆ Analysis
k∆
iv) ∆ ≠ 0 6 EI 6 EI 3 3 ∆ , M CD = M DC = 2 ∆ , 2 l l 24 EI = 3 ∆ , VS3 = k∆ l
3 3 M AB = M BA = 3 3 VBA = VCD
z
Construct the Stiffness Equation
Pab 2 8EI 2 EI 6 EI ∑ M = 0 → l 2 + l θ B + l θc + l 2 ∆ = 0 Pa 2b 2 EI 8 EI 6 EI i M = 0 → − + θB + θc + 2 ∆ = 0 ∑ C 2 l l l l 6 EI 6 EI 24 EI θ B + 2 θc + 3 ∆ = − k∆ ∑V i = 0 → 2 l l l i B
6 EI 6 EI 24 EI θ B + 2 θc + ( 3 + k ) ∆ = 0 2 l l l z
Deformed Shapes (
∆S = 0.41 ) ∆
without a spring
with a spring ( k = 24
EI ) l3
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
27
School of Civil, Urban & Geosystem Eng., SNU 2.3.8 A Portal Frame Subject to Support Settlement
B
C
A
D δ
z
DOF : θ B , θC , ∆
z
Analysis
i) All fixed 6 EI δ l2 ii)-iv) the same as the previous problem 0 0 M BC = M CB =
z
Construct the Stiffness Equation 6 EI 8EI 2 EI 6 EI ∑ M Bi = 0 → l 2 δ + l θB + l θc + l 2 ∆ = 0 6 EI 2 EI 8EI 6 EI ∑ M Ci = 0 → l 2 δ + l θB + l θc + l 2 ∆ = 0 6 EI 6 EI 24 EI θ B + 2 θc + 3 ∆ = 0 ∑V i = 0 → 2 l l l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
28
School of Civil, Urban & Geosystem Eng., SNU 2.3.9 A Portal Frame with Unsymmetrical Supports
l/2 B
A z
DOF : θ B , θC , ∆
z
Analysis
P C
D
i) All fixed 0 M BC =
Pl Pl 0 , M CB =− 8 8
ii) θ B ≠ 0 3EI θB l 4 EI 2 EI 1 1 M BC = θ B , M CB = θB l l 3EI 1 VBA = 2 θB l 1 M BA =
iii) θC ≠ 0 2 EI 4 EI 2 θC , M CB = θC l l 4 EI 2 EI 2 2 M CD = θC , M DC = θC l l 6 EI 2 VCD = 2 θC l 2 M BC =
iv) ∆ ≠ 0 3 M BA =
3EI ∆, l2
6 EI ∆ , l2 3EI 12 EI 3 = 3 ∆ , VCD = 3 ∆ l l
3 3 M CD = M DC = 3 V BA
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
29
School of Civil, Urban & Geosystem Eng., SNU z
Construct the Stiffness Equation
Pl 7 EI 2 EI 3EI + θB + θc + 2 ∆ = 0 l l l 8 Pl EI EI EI 2 8 6 ∑ M Ci = 0 → − 8 + l θB + l θc + l 2 ∆ = 0 3EI 6 EI 15 EI θ B + 2 θc + 3 ∆ = 0 ∑V i = 0 → 2 l l l l ∆ = − (θ B + 2θC ) 5 2 1 Pl 1 9 Pl 2 1 Pl 3 θB = − , θC = , ∆=− 44 EI 44 8 EI 176 EI
∑M
z
i B
=0→
Load Location that Causes No Sidesway
P
a
C
B
A
D
l ∆ = − (θ B + 2θC ) = 0 → θ B = −2θC 5 - Stiffness equation
∑ M Bi = 0 →
Pab 2 7 EI 2 EI + θB + θc = 0 , 2 l l l Pab 2 12 EI − θC = 0 l2 l
,
∑ M Ci = 0 → − −
Pa 2b 2 EI 8EI + θB + θc = 0 2 l l l
Pa 2b 4 EI + θc = 0 l2 l
Pab 2 Pa 2b 3 = → b = 3a l2 l2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
30
School of Civil, Urban & Geosystem Eng., SNU 2.3.10 A Frame with a Skewed Member
P
B
C
A
z
DOF : θ B , ∆
z
Analysis
0 i) All fixed : M BC =
2 11 Pl Pl 3 0 + = Pl , VBC =− P 2 16 8 16 16
ii) θ B ≠ 0 M 1AB =
2 EI EI EI 3EI EI 1 1 1 θB = 2 θ B , M BA =2 2 θ B , M BC = θ B , VBA = 3 2 θB l l l l 2l
1 VBC =−
3 2 EI θB 2 l2
3 (2 2
EI 1 2 2 EI θB =3 θB l l 2 2 l2
EI EI EI 1 θB + 2 θB ) = 3 2 θB l l l 2l
iii) ∆ ≠ 0 2 2 M BA = M AB = 2 VBC =
3EI ∆ 3 2 EI EI 6 EI ∆ 3EI 2 2 = 2 ∆ M BC =− =− ∆ , VBA =3 2 3 ∆ , 2 l l l 2 l 2l 2l 2l
3 EI ∆ 2 l3
3 (3
2 EI 1 1 3 EI ∆ = ∆ 2 l 2 2 l3 2 l
EI EI EI 1 ∆ + 3 2 ∆) = 3 2 3 θB 2 l l l 2l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU z
31
Construct the stiffness equation
3 2 EI EI 3 ) 2 ∆ = − Pl θ B + (3 − 2 l l 16 3 3 EI EI 2 11 ∑V i = 0 → (3 − 2 2) l 2 θB + (3 2 + 2 ) l 3 ∆ = 2 16 P
∑M
i B
= 0 →(2 2 + 3)
EI EI θ B + 0.8787 2 ∆ = −0.1875 Pl l l EI EI 0.8787 2 θ B + 5.7426 3 ∆ = 0.4861P l l
5.8284
Pl 2 Pl 3 θ B = −0.0460 , ∆ = 0.0917 EI EI z
Results
- Deformed shape
- Moment diagram
- Shear force diagram
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU
32
Chapter 3 Iterative Solution Method & Moment Distribution Method
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
33
School of Civil, Urban & Geosystem Eng., SNU 3.1 Solution Method for Linear Algebraic Equations 3.1.1 Direct Method – Gauss Elimination a11 X 1 + a12 X 2 + L + a1i X i a 21 X 1 + a 22 X 2 + L + a 2i X M a i1 X 1 + ai 2 X 2 + L + aii X i M a n1 X 1 + a n 2 X 2 + L + a ni X i
+ L + a1n X n = b1 + L + a 2 n X n = b2 n → ∑ aij X j = bi for i = 1L n + L + a in X n = bi j =1 + L + a nn X n = bn
or in a matrix form
[ A]( X) = (b) By multiplying
ai1 to the first equation and subtracting the resulting equation from the i-th a11
equation for 2 ≤ i ≤ n , the first unknown X1 is eliminated from the second equation as follows. a11 X 1 + a12 X 2 + L + a1i X i ( 2) a 22 X 2 + L + a 2( 2i ) X M ( 2) ( 2) ai 2 X 2 + L + aii X i M ( 2) ( 2) an2 X 2 + L + an2 X i
where aij( 2) = aij −
ai1 a1 j a11
+ L + a1n X n = b1 + L + a 2( 2n) X n = b2( 2 ) + L + ain( 2) X n = bi( 2 ) ( 2) + L + a nn X n = bn( 2 )
. Again, the second unknown X2 is eliminated from the third equa-
ai(22) tion by multiplying ( 2) to the second equation and subtracting the resulting equation from a 22
the i-th equation for 3 ≤ i ≤ n .
The aforementioned procedures are repeated until the last
unknown remains in the last equation. a11 X 1 + a12 X 2 + L + a1i X i + L + a1n X n ( 2) a 22 X 2 + L + a 2( 2i ) X + L + a 2( 2n) X n M M (i ) (i ) aii X i + L + ain X n M (n) a nn X n
= b1 = b2( 2 ) M = bi( i ) M (n) = bn
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
34
School of Civil, Urban & Geosystem Eng., SNU
a
=a
( k −1) ij
( n −1) n −1, n −1
−
ai(,kk−−11) a k( k−−1,1j)
k ≤ i, j ≤ n , and a ij1 = aij . Once the system matrix is tria kk−−11,k −1 angularized, the solution of the given system is easily obtained by the back-substitution. bn( n ) (n) ( n) X n = (n) a nn X n = bn → a nn
where a
(k ) ij
X n −1 + a
( n −1) n −1, n
Xn = b
( n −1) n −1
→
X n −1 =
bn( n−1−1) − a n( n−−1,1n) X n a nn−−12,n −1 i +1
aii( i ) X i + L + a i(,in) X n = bi(i ) →
Xi =
bi( i ) − ∑ aik(i ) X k k =n i −1 ii
a
for 1 ≤ i ≤ n − 1
3.1.2 Iterative Method – Gauss-Jordan Method
A system of linear algebraic equations may be solved by iterative method. For this purpose, the given system is rearranged as follows. X1 =
b1 − (a12 X 2 + L + a1n X n ) a11
X2 =
b2 − (a 21 X 1 + a 23 X 3 + L + a 2 n X n ) a 22
Xi = Xn =
bi − (ai1 X 1 + L + ai ,i −1 X i −1 + ai ,i +1 X i −1 + L + ain X n ) aii bn − (a n1 X 1 + L + a n ,n −1 X n −1 ) a nn
Suppose we substitute an approximate solution ( X) k −1 into the right-hand side of the above equation, a new approximate solution ( X) k , which is not the same as ( X) k −1 , is obtained. This procedure is repeated until the solution converges. ( X i )k =
n 1 (bi − ∑ aij ( X j ) k −1 ) aii j =1 i≠ j
where the subscript k denotes the iterational count.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
35
School of Civil, Urban & Geosystem Eng., SNU 3.1.3 Iterative Method – Gauss-Siedal Method
When we calculate a new Xi value in the k-th iteration of Gauss-Jordan iteration, the values of X 1 ,L, X i −1 are already updated, and we can utilize the updated values to accelerate conver-
gence rate, which leads to the Gauss-Siedal Method. ( X1)k =
b1 − (a12 ( X 2 ) k −1 + L + a1n ( X n ) k −1 ) a11
( X 2 )k =
b2 − (a 21 ( X 1 ) k + a 23 X 3 ) k −1 + L + (a 2 n X n ) k −1 ) a 22
( X i )k = ( X n )k =
bi − (ai1 ( X 1 ) k + L + ai ,i −1 ( X i −1 ) k + ai ,i +1 ( X i +1 ) k −1 + L + ain ( X n ) k −1 ) aii bn − (a n1 ( X 1 ) k + L + a n ,n −1 ( X n −1 ) k ) a nn ( X i )k =
i −1 1 (bi − ∑ aij ( X j ) k − aii j =1 i >1
n
∑a
j =i +1 i
ij
( X j ) k −1 )
3.1.4 Example
30
35.6 B
A EI
4klf C
1.5EI
D EI
3@30=90 z
Stiffness Equation
3EI 6 EI 3EI + )θ B + θC = 0 l l l 3EI 6 EI 3EI − 133.3 + 450.0 + θB + ( + )θC = 0 l l l − 168.7 + 133.3 + (
For the simplicity of derivation,
EI EI θB → θB , θ C → θ C . The stiffness equation becomes l l − 35.4 + 9θ B + 3θ C = 0 316.7 + 3θ B + 9θ C = 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
36
School of Civil, Urban & Geosystem Eng., SNU z
Gauss-Jordan Iteration
1 (35.4 − 3(θ C ) k −1 ) 9 1 (θ C ) k = (316.7 − 3(θ B ) k −1 ) 9
(θ B ) k =
z
Gauss-Siedal Iteration
1 (35.4 − 3(θ C ) k −1 ) 9 1 (θ C ) k = (316.7 − 3(θ B ) k ) 9
(θ B ) k =
Gauss-Seidal
Gauss-Jordan
GAUSS-SIEDAL ITERATION ====================== ***** Iteration 1***** X(1) = 0.3933334E+01 X(2) = -0.3650000E+02 ERROR = 0.1000000E+01 ***** Iteration 2***** X(1) = 0.1610000E+02 X(2) = -0.4055556E+02 ERROR = 0.2939146E+00 ***** Iteration 3***** X(1) = 0.1745185E+02 X(2) = -0.4100617E+02 ERROR = 0.3197497E-01 ***** Iteration 4***** X(1) = 0.1760206E+02 X(2) = -0.4105624E+02 ERROR = 0.3544420E-02 ***** Iteration 5***** X(1) = 0.1761875E+02 X(2) = -0.4106181E+02 ERROR = 0.3937214E-03
GAUSS-Jordan ITERATION ====================== ***** Iteration 1***** X(1) = 0.3933334E+01 X(2) = -0.3518889E+02 ERROR = 0.1000000E+01 ***** Iteration 2***** X(1) = 0.1566296E+02 X(2) = -0.3650000E+02 ERROR = 0.2971564E+00 ***** Iteration 3***** X(1) = 0.1610000E+02 X(2) = -0.4040988E+02 ERROR = 0.9044394E-01 ***** Iteration 4***** X(1) = 0.1740329E+02 X(2) = -0.4055556E+02 ERROR = 0.2971564E-01 ***** Iteration 5***** X(1) = 0.1745185E+02 X(2) = -0.4098999E+02 ERROR = 0.9812154E-02
****** MOMENT ******
****** MOMENT ******
MBA MBC MCB MCD
=-115.84375 = 115.82707 =-326.81460 = 326.81458
MBA MBC MCB MCD
=-116.34444 = 115.04115 =-326.88437 = 327.03004
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
37
School of Civil, Urban & Geosystem Eng., SNU 3.2 Moment Distribution Method 30
35.6 B
A EI
4klf C
1.5EI
D EI
3@30=90
z
At the Joint B
- Moment distribution MB = (
1 = M BA
MB 3EI AB 4 EI BC 1 1 + + M BC → θB = )θ B = M BA 3EI AB 4 EI BC L AB LBC + L AB LBC 3EI AB L AB
3EI AB L AB θB = M = DBA M B 3EI AB 4 EI BC B + L AB LBC
4 EI BC 4 EI BC LBC 1 = θB = M BC M = DBC M B 3EI AB 4 EI BC B LBC + L AB LBC 2 EI BC 1 1 - Moment carry over to joint C: M CB = θ B = DBC M B 2 LBC z
At the Joint C
- Moment distribution MC = (
2 M CB =
2 = M CD
4 EI BC 3EI CD MC 2 2 + )θ C = M CB + M CD → θC = 4 EI BC 3EI CD LBC LCD + LBC LCD 4 EI BC LBC
4 EI BC LBC θC = M = DCB M C 4 EI BC 3EI CD C + LBC LCD
3EI CD LCD
3EI CD LCD θC = M = DCD M C 4 EI BC 3EI CD C + LBC LCD
2 - Moment carry over to joint B: M BC =
2 EI BC 1 θ C = DCB M C LBC 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
38
School of Civil, Urban & Geosystem Eng., SNU z
Stiffness Equation in terms of Moment at Joints
1 1 DCB M C = 0 M B = 35.4 − DCB M C 2 2 → 1 1 316.7 + DBC M B + M C = 0 M C = - 316.7 − DBC M B 2 2 − 35.4 + M B +
- Gauss-Siedal Approach 1 DCB ( M C ) k −1 2 1 ( M C ) k = − 316.7 − DBC ( M B ) k 2
( M B ) k = 35.4 −
- Gauss-Jordan Approach 1 DCB ( M C ) k −1 2 1 ( M C ) k = − 316.7 − DBC ( M B ) k −1 2
( M B ) k = 35.4 −
z
For the given structure
D BA = DCD = z
1 , 3
D BC = DCB =
2 3
Incremental form for the Gauss-Siedal Method
- For k = 1 ( M B ) 0 = ( M C ) 0 = 0 because we assume all degrees of freedom are fixed for step 0. 1 DCB ( M c ) 0 = 35.4 → (∆M B )1 = 35.4 2 1 12 ( M C )1 = −316.7 − DBC ( M B )1 = −316.7 − 35.4 → (∆M C )1 = −328.5 2 23 ( M B )1 = 35.4 −
- For k > 1 1 1 1 DCB ( M c ) k −1 = 35.4 − DCB ( M c ) k − 2 − DCB (∆M c ) k −1 2 2 2 1 1 = ( M B ) k −1 − DCB (∆M c ) k −1 → (∆M B ) k = − DCB (∆M c ) k −1 2 2 1 1 1 ( M C ) k = −316.7 − DBC ( M B ) k = −316.7 − DBC ( M B ) k −1 − DBC (∆M B ) k 2 2 2 1 1 = ( M C ) k −1 − DBC (∆M B ) k → (∆M C ) k = − DBC (∆M B ) k 2 2 ( M B ) k = 35.4 −
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU
39
- Iteration 1 (M B ) 0 = 0 , (M C ) 0 = 0 ( M B ) 1 = 35.4 −
1 DCB ( M C ) 0 = 35.4 → (∆M B ) 1 = 35.4 2
f + DBA (∆M B )1 = −168.7 + 0.33 × 35.4 = −168.7 + 11.8 → (∆M BA )1 = 11.8 ( M BA )1 = M BA f ( M BC )1 = M BC + DBC (∆M B )1 = 133.3 + 0.67 × 35.4 = 133.3 + 23.6 → (∆M BC )1 = 23.6
( M C )1 = −316.7 − f ( M CB )1 = M CB +
1 12 DBC (∆M B )1 = −316.7 − 35.4 = −328.5 → (∆M c )1 = −328.5 2 23 1 DBC (∆M B )1 + DCB (∆M C )1 = −133.3 + 11.8 − 219.0 → 2 (∆M CB )1 = 11.8 − 219.0
f ( M CD )1 = M CD + DCD (∆M C )1 = 450 − 0.33 × 328.5 = 450 − 109.5 → (∆M CD )1 = −109.5
- Iteration 2
(∆M BA ) 2 = DBA (∆M B ) 2 = 36.5 1 1 (∆M B ) 2 = − DCB (∆M c )1 = 109.5 → (∆M BC ) 2 = DCB (∆M c )1 + DBC (∆M B ) 2 2 2 = −109.5 + 73.0 (∆M ) = 1 D (∆M ) + D (∆M ) CB 2 BC B 2 CB C 2 1 2 (∆M C ) 2 = − DBC (∆M B ) 2 = −36.5 → = 36.5 − 24.3 2 (∆M CD ) 2 = DCD (∆M C ) 2 = −12.2
- Iteration 3
(∆M BA ) 3 = D BA ( ∆M B ) 3 = 4.1 1 1 (∆M B ) 3 = − DCB (∆M c ) 2 = 12.2 → (∆M BC ) 3 = DCB (∆M c ) 2 + D BC (∆M B ) 3 2 2 = −12.2 + 8.2 ( ∆M ) = 1 D (∆M ) + D (∆M ) = 4.1 − 2.7 1 CB 3 BC B 3 CB C 3 (∆M C ) 3 = − D BC (∆M B ) 3 = −4.1 → 2 2 ( ∆M CD ) 3 = DCD (∆M C ) 3 = −1.4
- Final Moments f M BA = M BA + ∑ (∆M BA ) k = −168.7 + 11.8 + 36.5 + 4.1 = −116.3 k
M BC = M
f BC
M CB = M
f CB
M CD = M
f CD
+ ∑ (∆M BC ) k = 133.3 + 23.6 + (−109.5 + 73.0) + (−12.2 + 8.2) = 116.4 k
+ ∑ (∆M CB ) k = −133.3 + (11.8 − 219.0) + (36.5 − 24.3) + (4.1 − 2.7) = −326.9 k
+ ∑ (∆M CD ) k = 450.0 − 109.5 − 12.2 − 1.4 = 326.9 k
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
40
School of Civil, Urban & Geosystem Eng., SNU 0.33
0.66
0.67
-168.7
133.3
-133.3
11.8
23.6
11.8
-109.5
-219.0
73.0
36.5
-12.2
-24.3
8.2
4.1
-1.4
-2.7
-1.4
0.5
0.9
-326.9
326.9
-115.8
115.9
36.5
4.1
z
0.33 450.0
-109.5
-12.2
Incremental form for the Gauss-Jordan Method
- For k = 1 ( M B ) 0 = ( M C ) 0 = 0 because we assume all degrees of freedom are fixed for step 0. 1 DCB ( M c ) 0 = 35.4 → (∆M B )1 = 35.4 2 1 ( M C )1 = −316.7 − DBC ( M B ) 0 = −316.7 → (∆M C )1 = −316.7 2 ( M B )1 = 35.4 −
- For k > 1 1 1 1 DCB ( M c ) k −1 = 35.4 − DCB ( M c ) k − 2 − DCB (∆M c ) k −1 2 2 2 1 1 = ( M B ) k −1 − DCB (∆M c ) k −1 → (∆M B ) k = − DCB (∆M c ) k −1 2 2 1 1 1 ( M C ) k = −316.7 − DBC ( M B ) k −1 = −316.7 − DBC ( M B ) k − 2 − DBC (∆M B ) k −1 2 2 2 1 1 = ( M C ) k −1 − DBC (∆M B ) k −1 → (∆M C ) k = − DBC (∆M B ) k −1 2 2 ( M B ) k = 35.4 −
- Iteration 1 (M B ) 0 = 0 , (M C ) 0 = 0 ( M B ) 1 = 35.4 −
1 DCB ( M C ) 0 = 35.4 → (∆M B ) 1 = 35.4 2
f + D BA ( M B ) 1 = −168.7 + 0.33 × 35.4 = −168.7 + 11.8 → (∆M BA ) 1 = 11.8 ( M BA ) 1 = M BA f ( M BC ) 1 = M BC + D BC ( M B ) 1 = 133.3 + 0.67 × 35.4 = 133.3 + 23.6 → (∆M BC ) 1 = 23.6
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
41
School of Civil, Urban & Geosystem Eng., SNU ( M C )1 = −316.7 −
1 DBC ( M B ) 0 = −316.7 = −316.7 → (∆M c )1 = −316.7 2
f ( M CB )1 = M CB + DCB ( M C )1 = −133.3 − 0.67 × 316.7 = −133.3 − 212.2 →
(∆M CB )1 = −212.2 f ( M CD )1 = M CD + DCD ( M C )1 = 450 − 0.33 × 316.7 = 450 − 104.5 → (∆M CD )1 = −104.5
- Iteration 2
(∆M BA ) 2 = DBA (∆M B ) 2 = 35.0 1 1 (∆M B ) 2 = − DCB (∆M c )1 = 106.1 → (∆M BC ) 2 = DCB (∆M c )1 + DBC (∆M B ) 2 2 2 = −106.1 + 71.1 (∆M ) = 1 D (∆M ) + D (∆M ) CB 2 BC B 1 CB C 2 1 2 (∆M C ) 2 = − DBC (∆M B )1 = −11.8 → = 11.8 − 7.9 2 (∆M CD ) 2 = DCD (∆M C ) 2 = −3.9
- Iteration 3
(∆M BA ) 3 = DBA (∆M B ) 3 = 1.3 1 1 (∆M B ) 3 = − DCB (∆M C ) 2 = 4.0 → (∆M BC ) 3 = DCB (∆M C ) 2 + DBC (∆M B ) 3 2 2 = − 4 .0 + 2 .7 (∆M ) = 1 D (∆M ) + D (∆M ) CB 3 BC B 3 CB C 3 1 2 (∆M C ) 3 = − DBC (∆M B ) 2 = −35.6 → = 35.6 − 23.9 2 (∆M CD ) 3 = DCD (∆M C ) 3 = −11.7
0.33
0.66
0.67
-168.7 11.8
35.0 1.3 4.0 0.2 0.5 -115.9
133.3 23.6 -106.1 71.1 -4.0 2. 7 -12.0 8.0 -0.5 0.3 -1.4 0.9 115.9
0.33 -133.3 -212.2 11.8 -7. 9 35.6 -23.9 1.4 -0.9 4.0 -2.7 0.2 0.1 -328.0
450.0 -104.5
-3.9 -11.7 -0.5 -1.3 0.1 328.2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
42
School of Civil, Urban & Geosystem Eng., SNU 3.3 Example - MDM for a 4-span Continuous Beam 100 A
DCD = 4 z
EI L
(4
1.5EI, L
1.5EI L
EI 1.5 EI 1.5 EI +4 ) = 0.4 , D BC = 4 L L L
(4
E
D
C
B EI, L
D BA = 4
50 1.5EI, L
(4
1.5EI 1.5 EI 1.5 EI +4 ) = 0.5 , D DC = 4 L L L
EI, L
EI 1.5 EI 1.5EI +4 ) = 0.6 , DCB = 4 L L L (4
(4
1.5EI 1.5EI +4 ) = 0.5 , L L
1.5EI EI EI + 3 ) = 0.67 , DDE = 3 L L L
(4
1.5 EI EI + 3 ) = 0.33 L L
Gauss-Siedal Approach 0.4 0.0 -25.0
-10.4
-1.1 36.5
0.0 -50.0
-20.8
-2.1 -72.9
0.6 125 -75 40.6
11.3 -31.1 3.9 1.3 -3.1 72.9
0.5 -125 -37.5 81.3
22.5 -15.6 7.8 2.6 -63.9
0.67
0.5 0.0
0.0
81.3 -45.0 22.5
40.6 -90.0 11.3
7.8 -5.1 2.6
3.9 -10.2 1.3 -0.9 -44.0
64.1
0.33 93.8
-44.3
-5.0 -0.4 44.1
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
43
School of Civil, Urban & Geosystem Eng., SNU z
Gauss-Jordan Approach 0.4 0.0
0.0 -50
-25.0 -12.5 -6.3 -6.9 -3.5 -2.0 -1.0 -1.1 -0.6 -36.4
-0.3 -72.8
0.6 125 -75 31.3 -18.8 17.3 -10.4 5.0 -3.0 2.8 -1.7 0.8 -0.5 72.8
0.5 -125 62.5 -37.5 34.5 -9.4 10.0 -5.2 5.5 -1.5 1.6 -0.9 1.0 -64.4
0.67
0.5 0.0 62.5 -31.4 34.5 -10.5 10.0 -5.8 5.5 -1.7 1.6 -1.0 1.0 64.7
0.0 -62.8 31.3 -21.0 17.3 -11.6 5.0 -3.4 2.8 -1.9 0.8 -0.5 -44.0
0.33 93.8 -31.0
-10.3 -5.7 -1.6 -0.9 -0.3 44.0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
44
School of Civil, Urban & Geosystem Eng., SNU 3.4 Direct Solution Scheme by Partitioning 20 L/3
B
C
A
z
D
Slope deflection (Stiffness) Equation
8EI L 2 EI L 6 EI L
2 EI L 8EI L 6 EI L
6 EI L θ B 88.8 [K θθ ] (K θ∆ ) (Θ) (P) 6 EI θC + − 44.4 = 0 → ∆ + 0 = 0 ( ) K K L ∆ θ ∆∆ 24 EI ∆ 0.0 L
[K θθ ](Θ) = −(P) − (K θ∆ )∆ → (Θ) = −[K θθ ] −1 (P) − [K θθ ] −1 (K θ∆ )∆ = (Θ) P + (Θ) ∆ (K ∆θ )(Θ) P + (K ∆θ )(Θ) ∆ + K ∆∆ ∆ = 0 z
Direct Solution Procedure by Partitioning -
Assume ∆ = 0 and calculate (Θ)P. ∆ Assume an arbitrary ∆ = and calculate ( Θ ) ∆ . α ∆ ( Θ) By linearity, ( Θ ) ∆ = α Calculate α by the second equation by ∆ and ( Θ ) ∆ . (K ∆θ )(Θ) P α=− K ∆∆ ∆ + (K ∆θ )(Θ ) ∆
-
Obtain (Θ) by
(Θ ) = (Θ ) P + α ( Θ ) ∆ .
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
45
School of Civil, Urban & Geosystem Eng., SNU 3.5 Moment Distribution Method for Frames z
z
Solution Procedure -
Assume there is no sideway and do the MDM.
-
Perform the MDM again for an assumed sidesway.
-
Adjust the Moment obtained by the second MDM to satisfy the second equation.
-
Add the adjusted moment to the moment by the first MDM.
First MDM – with no Sidesway L/3
20
B 0.5
0.5 C
0.5
-44.4
-44.4
-22.2
16.6
33.3
8.3
-4.2
1.1
2.1
-0.6
-0.6
-0.3
-53.3
53.3
0.2
0.2
-35.5
35.5
-44.4
-8.3
0.5
88.8
33.3
2.1
A
D
MAB = -53.3/2 = -26.7, MDC = -35.5/2 = 17.8, VAB = 2.67, VDC = -1.78, VP = 0.89 z
Second MDM – with an Arbitrary Sidesway B
C 10.0
10.0 -5.0
-2.5
-1.9
-3.8
1.0
0.5
-0.2
-0.3
-0.3
0.1
0.1
-5.9
5.9
6.1
-6.0
-5.0
1.0
A
-3.8
D
MAB = 10 – 3.9/2.0 = 8.1, MDC = 10 – 4.1/2.0 = 7.9, VAB =-0.47, VDC = -0.46, V ∆ = -0.93
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
46
School of Civil, Urban & Geosystem Eng., SNU z
Shear Equilibrium Condition (the 2nd equation)
Sidesway Only (2nd MDM)
No Sidesway (1st MDM)
0.93
0.89 53.3
35.5 2.67
26.7
5.9
6.1 0.47
1.78
17.8
8.1
0.46
7.9
α = -0.89/(-0.93) = 0.97
Total Moment = 1st Moment + α 2nd Moment
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
47
School of Civil, Urban & Geosystem Eng., SNU
Chapter 4 Energy Principles Principle of Minimum Potential Energy and Principle of Virtual Work
Mg
h
P
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
48
School of Civil, Urban & Geosystem Eng., SNU
Read Chapter 11 (pp.420~ 428) of Elementary Structural Analysis 4th Edition by C .H. Norris et al very carefully.
In this note an overbarred variable denotes a virtual quantity.
The vir-
tual displacement field should satisfy the displacement boundary conditions of supports if specified. For beam problems, displacement boundary conditions include boundary conditions for rotational angle. Variables with superscript e denote the exact solution that satisfies the equilibrium equation(s).
4.1 Spring-Force Systems
∆ P
z
Total Potential energy The energy required to return a mechanical system to a reference status ∆ 0 1 Π int = − ∫ k (∆ − u )du = ∫ k (∆ − u )du = k∆2 , Π ext = − P∆ 2 ∆ 0 Π total = Π int + Π ext =
z
1 2 k∆ − P∆ 2
Equilibrium Equation k∆e=P
z
Principle of Minimum Potential Energy for an arbitrary displacement ∆ = ∆e + ∆ .
1 k (∆e + ∆ ) 2 − P(∆e + ∆ ) 2 1 1 = k (∆e ) 2 + k∆e ∆ + k ( ∆ ) 2 − P(∆e + ∆ ) 2 2 1 1 = k (∆e ) 2 − P∆e + k ( ∆ ) 2 + ∆ (k∆e − P) 2 2 1 1 = k (∆e ) 2 − P∆e + k ( ∆ ) 2 2 2 1 e e = Π total + k ( ∆ ) 2 ≥ Π total 2
Π total =
In the above equation, the equality sign holds if and only if ∆ = 0 .
Therefore the total
potential energy of the spring-force system becomes minimum when displacement of spring satisfies the equilibrium equation.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
49
School of Civil, Urban & Geosystem Eng., SNU 4.2 Beam Problems q
w
any type of support z
w
Potential Energy of a Beam l
Π int =
l
l
l
l
1 d 2w 2 EI ( ) dx − ∫ qwdx 2 ∫0 dx 2 0
Equilibrium Equation EI
z
any type of support
d 2w 2 1 M2 1 = dx EI ( ) dx , Π ext = − ∫ qwdx 2 ∫0 EI 2 ∫0 dx 2 0
Π total = Π int + Π ext = z
we
d 2M e d 4 we = − q or EI =q dx 2 dx 4
Principle of Minimum Potential Energy for a virtual displacement w = w e + w . 1 d 2 (we + w ) d 2 (we + w ) ( )dx − ∫ ( w e + w )qdx EI 2 2 ∫ 20 dx dx 0 l
Πh =
l
l
=
l
1 d 2 we d 2 we ( )dx − ∫ w e qdx + EI 2 ∫0 dx 2 dx 2 0 l
l
l
1 d 2w d 2w d 2w d 2 we + ( )dx ∫ ( 2 EI )dx − ∫ w qdx EI 2 2 ∫0 dx 2 dx 2 dx dx 0 0 = Πe +
l
l
l
l
l
1 d 2w d 2w d 2w 1 d 2 we + − − ( EI ) dx ( EI ) ( EI )dx − ∫ w qdx ∫0 2 ∫0 dx 2 dx 2 EI dx 2 dx 2 0 l
1 d 2w d 2w MM e = Π + ∫ ( 2 EI )dx + ∫ dx − ∫ w qdx 2 0 dx EI dx 2 0 0 e
l
= Πe +
1 d 2w d 2w ( EI )dx ≥ Π e for all virtual w 2 2 ∫0 dx 2 dx
Since the equation in the box represents the total virtual work in a beam, the total potential energy
of a beam becomes minimum for all virtual displacement fields when the principle of virtual work holds. In the above equation, the equality sign holds if and only if w = 0 .
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
50
School of Civil, Urban & Geosystem Eng., SNU z
Principle of Virtual Work If a beam is in equilibrium, the principle of the virtual work holds for the beam,. l
∫ w ( EI 0
d 4 we − q )dx = 0 for all virtual displacement w dx 4
l
l
l
l
d 2w d 2 we dw d 2 we d 3 we − + − EI dx w qdx EI w EI ∫0 dx 2 dx 2 ∫0 dx dx 2 0 dx 3 l
l
l
=0 0
l
d 2w d 2 we MM e EI dx w qdx − = ∫0 dx 2 dx 2 ∫0 ∫0 EI dx − ∫0 wqdx = 0
In case that there is no support settlement, the boundary terms in above equation vanishes identically since either virtual displacement including virtual rotational angle or corresponding forces (moment and shear) vanish at supports. The principle of virtual work x in a beam by applying an unit load at ~ x yields the displacement of an arbitrary point ~ and by using the reciprocal theorem. l
l
l
l
e
MM ~ ~ ∫0 w qdx = ∫0 wq dx = ∫0 wδ( x − x )dx = w( x ) = ∫0 EI dx z
Approximation using the principle of minimum potential energy -
Approximation of displacement field n
w = ∑ ai g i i =1
-
Total potential energy by the assumed displacement field l
l
l
l
n n n 1 d 2w d 2w 1 Π = ∫ ( 2 EI 2 )dx − ∫ wqdx = ∫ (∑ ai g i′′)EI (∑ a j g ′j′ )dx − ∫ ∑ ai g i qdx 2 0 dx 2 0 i =1 dx j =1 0 0 i =1 h
-
The first-order Necessary Condition n n n ∂Π h ∂ 1 ( ∫ (∑ ai g i′′)EI (∑ ai g i′′)dx − ∫ ∑ ai g i qdx) = ∂a k ∂a k 2 0 i =1 i =1 0 i =1 l
l
l
l
l
n n 1 = [ ∫ g k′′ EI (∑ a j g ′j′ )dx + ∫ (∑ ai g i′′)EIg k′′dx] − ∫ g k qdx) 2 0 i =1 0 i =1 0 l
l
n
i =1 0
0
i =1
n
or
Ka = f
= ∑ ∫ g k′′ EIg i′′dxai − ∫ g k qdx = ∑ K ki ai − f k = 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
51
School of Civil, Urban & Geosystem Eng., SNU z
Example P
i) with one unknown
w = ax( x − l ) = a ( x 2 − lx) → w′′ = 2a l
Π total =
l
l
l l2 1 l2 1 1 2 2 2 ′ ′ EI w dx P x wdx EI a dx aP EI a l aP ( ) ( ) ( 2 ) 4 − δ − = + = + ∫0 2 ∫0 2 2 ∫0 4 2 4
∂Π total l2 Pl Pl = 0 → 4aEIl + P = 0 → a = − →w=− ( x 2 − xl ) ∂a 4 16 EI 16 EI l Pl 3 l Pl 3 Pl 3 w( ) = , we ( ) = = 0.0208 , Error = 2 48 EI 2 64 EI EI
l l w e ( ) − w( ) 2 2 = 0.25 l we ( ) 2
ii) with two unknowns
w = ax( x − l ) + bx( x 2 − l 2 ) → w′′ = 2a + 6bx l
Π total = =
l
l
1 1 3l 3 l l2 2 2 ′ ′ − δ − = + − − − ) ( ) ( ) ( 2 6 ) ( EI w dx P x wdx EI a bx dx P a b ∫0 2 ∫0 2 2 ∫0 4 8 1 3l 3 l2 l3 l2 ) EI (4a 2 l + 24ab + 36b 2 ) + P(a + b 2 2 3 4 8
∂Π total l2 = 0 → EI (4la + 6l 2 b) = − P ∂a 4 ∂Π total 3l 3 = 0 → EI (6l 2 a + 12l 3 b) = − P ∂b 8
→a=−
1 Pl , b = 0 ??? 16 EI
iii) with three unknowns
w = ax( x − l ) + bx( x 2 − l 2 ) + cx( x 3 − l 3 ) → w′′ = 2a + 6bx + 12cx 2 l
Π total =
l
1 l EI ( w′′) 2 dx − ∫ Pδ( x − ) wdx ∫ 20 2 0 l
l2 1 3l 3 7l 4 2 2 = ∫ EI (2a + 6bx + 12cx ) dx − P(−a − b −c ) 20 4 8 16 l3 l5 l2 l3 l4 1 EI (4a 2 l + 36b 2 + 144c 2 + 24ab + 48ac + 144bc ) + 2 3 5 2 3 4 2 3 4 3l 7l l P(a + b +c ) 4 8 16
=
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
52
School of Civil, Urban & Geosystem Eng., SNU ∂Π total l2 2 3 = 0 → EI (4la + 6l b + 8l c) = − P 4 ∂a ∂Π total 3l 3 = 0 → EI (6l 2 a + 12l 3 b + 18l 4 c) = − P 8 ∂b ∂Π total 144 5 7l 4 l c) = − P = 0 → EI (8l 3 a + 18l 4 b + 5 16 ∂c
w=
1 Pl a = 64 EI 5 P → b = − 32 EI 5 P c = 64 EIl
5 P 5 P 1 Pl x( x − l ) − x( x 2 − l 2 ) + x( x 3 − l 3 ) 64 EI 32 EI 64 EIl
1 1 5 3 5 7 21 Pl 3 l Pl 3 Pl 3 (− + − )= = 0.0205 , Error = 0.0144 w( ) = 2 64 4 32 8 64 16 1024 EI EI EI iv) with one sin function w = a sin
l π π 1 1 = ∫ EI ( w′′) 2 dx − ∫ Pδ( x − ) wdx = ∫ EI (a( ) 2 sin x) 2 dx − aP l l 20 2 20 0 l
Π total
π π π x =→ w′′ = a( ) 2 sin x l l l
l
l
π π π l 1 1 EIa 2 ( ) 4 ∫ sin 2 xdx + aP = EIa 2 ( ) 4 + aP l 0 l l 2 2 2 l
=
∂Π total π l π 2 Pl 3 2 Pl 3 = 0 → EIa( ) 4 − P = 0 → a = 4 →w= 4 sin x ∂a l l 2 π EI π EI l 1 Pl 3 l Pl 3 , w( ) = , we ( ) = 2 48.7045 EI 2 48EI
Error = 0.0145
v) with two sin function w = a sin
π 3π 3π 3π π π x + b sin x =→ w′′ = a( ) 2 sin x + b( ) 2 sin x l l l l l l
l
Π total
l
1 l = ∫ EI ( w′′) 2 dx − ∫ Pδ( x − ) wdx 20 2 0 1 3π 3π 2 π π EI (a( ) 2 sin x + b( ) 2 sin x) dx − aP + bP ∫ 20 l l l l l
=
1 3π π π π 3π π xdx + = EIa 2 ( ) 4 ∫ sin 2 xdx +EIab( ) 2 ( ) 2 ∫ sin sin 2 l 0 l l l 0 l l l
l
1 3π 3π EIb 2 ( ) 4 ∫ sin 2 xdx − aP + bP 2 l 0 l l
=
1 π l 1 3π l EIa 2 ( ) 4 + EIb 2 ( ) 4 − aP + bP 2 l 2 2 l 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
53
School of Civil, Urban & Geosystem Eng., SNU ∂Π total π l 2 Pl 3 = 0 → EIa( ) 4 − P = 0 → a = 4 ∂a l 2 π EI ∂Π total 3π l 2 Pl 3 = 0 → EIb( ) 4 + P = 0 → b = − ∂b l 2 (3π) 4 EI π 2 Pl 3 1 3π (sin x − sin x) 4 l 81 l π EI l Pl 3 Pl 3 l Pl 3 w( ) = 0.0205(1 + 0.0123) = 0.0208 , w e ( ) = 0.0208 , Error ≅0 2 2 EI EI EI w=
4.3 Truss problems − F ji − Fmi ( i )
X
i
Xi
Y
i
− F1i
Yi z
Potential Energy Π int =
njn 1 nmb ( Fi ) 2 li Π = − , ( X i u i + Yi vi ) ∑ ∑ ext 2 i =1 EA i i =1
Π total = Π int + Π ext =
1 nmb ( Fi ) 2 li njn − ∑ ( X i u i + Yi vi ) ∑ 2 i =1 EA i i =1
where nmb and njn denotes the total number of members and the total numbers of joints in a truss. z
Equilibrium Equations m(i )
m(i )
− ∑ H + X = 0 , − ∑ V ji + Y i = 0 j =1
i j
i
j =1
for
i = 1, L , njn
where m(i), H ij and V ji are the number of member connected to joint i, the horizontal component and the vertical component of the bar force of j-th member connected to joint i, respectively.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
54
School of Civil, Urban & Geosystem Eng., SNU z
Principle of Minimum Potential Energy Π total
1 nmb ( Fi e + Fi ) 2 l i njn = ∑ − ∑ ( X i (u i +u i ) + Yi (vi + vi )) 2 i =1 EAi i =1 =
1 nmb ( Fi e ) 2 li njn 1 nmb ( Fi ) 2 li nmb Fi e Fi li njn ( ) X u Y v − + + +∑ − ∑ ( X i u i + Yi vi ) ∑ ∑ ∑ i i i i 2 i =1 EAi 2 i =1 EAi EAi i =1 i =1 i =1
=
nmb ( Fi ) 2 li 1 nmb ( Fi e ) 2 li njn ( ) X u Y v − + + ∑ ∑ ∑ i i i i 2 i =1 EAi EAi i =1 i =1
1 nmb ( Fi ) 2 l i =Π + ∑ ≥ Πe 2 i =1 EAi e
for all virtual displacement fields
where Fi e and Fi are the bar force of i-th member induced by the real displacement of joints and virtual displacement induced by the virtual displacement of joints. Since the equation in the box represents the total virtual work in a truss, the total potential energy of a truss becomes
minimum for all virtual displacement fields when the principle of virtual work holds. In the above equation, the equality sign holds if and only if the virtual displacements at all joints are zero. z
Virtual Work Expression If a truss is in equilibrium, the principle of the virtual work holds for the truss,. njn
m(i )
m(i )
i =1
j =1
j =1
∑ ((− ∑ H ij + X i )u i + (-∑V ji + Y i )v i ) = 0
(v 2 − v1 ) sin θ i (v 2 − v1 )
(u 2 − u1 ) cos θ i
(u 2 − u1 )
θi
θi
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
55
School of Civil, Urban & Geosystem Eng., SNU njn
m(i )
m(i )
i =1
j =1
j =1
∑ ((− ∑ F ji cos θ j + X i )u i + (-∑ F ji sin θ j + Yi )vi ) = 0 m(i )
njn
∑ (u ∑ F i
i =1
j =1
i j
m(i )
njn
j =1
i =1
cos θ j + δv i ∑ F ji sin θ j ) = ∑ ( X i u i + Yi vi )
nmb
njn
i =1
i =1
∑ ( Fi e cos θ i (ui2 − u i1 ) + Fi e sin θi (vi2 − vi1 )) = ∑ ( X i ui + Yi vi ) nmb
nmb
nmb
i =1
i =1
∑ Fi e (cos θ i (ui2 − u i1 ) + sin θ i (vi2 − vi1 )) = ∑ Fi e ∆li = ∑ Fi e i =1
nmb
∑ i =1
nmb Fi l i F eF l =∑ i i i ( EAi ) i =1 ( EAi )
e
n Fi Fi l i = ∑ ( X i u i + Yi vi ) ( EAi ) i =1
The principle of virtual work yields the displacement of a joint k in a truss by applying an unit load at a joint k in an arbitrary direction and by using the reciprocal theorem. Fi e Fi l i X k u k + Yk v k = X u k cos α = u k cos α = ∑ i =1 ( EAi ) nmb
Since α represnts the angle between the applied unit load and the displacement vector, u k cos α are the displacement of the joint k in the direction of the applied unit load.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU
56
Chapter 5 Matrix Structural Analysis
Mr. Force & Ms. Displacement Matchmaker: Stiffness Matrix
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
57
School of Civil, Urban & Geosystem Eng., SNU 5.1 Truss Problems 5.1.1 Member Stiffness Matrix R f yR δ y
f yL δ Ly
f xL δ Lx
z
f xR δ Rx
Force – Displacement relation at Member ends EA R (δ x − δ Lx ) L EA R (δ x − δ Lx ) f xR = L L R fy = fy = 0
f xL = −
z
Member Stiffness Matrix in Local Coordinate System
e
f xL 1 L fy EA 0 = f xR L − 1 R fy 0
0 − 1 0 δ Lx 0 0 0 δ Ly 0 1 0 δ Rx 0 0 0 δ Ry
e
(f )e = [k ]e (δ )e z
Transformation Matrix
Vy
vy
vx
②
Vx
φ
① V x = cos φv x − sin φv y V x → V y = sin φv x + cos φv y V y
cos φ − sin φ v x = v sin cos φ φ y
(V ) = [ γ ]T ( v) → ( v) = [ γ ](V )
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
58
School of Civil, Urban & Geosystem Eng., SNU z
Member End Force e
Fx1 0 cos φ − sin φ 0 1 sin φ cos φ 0 0 Fy F2 = 0 0 cos φ − sin φ x F2 φ φ 0 0 sin cos y z
e
f xL f yL → (F)e = [Γ]T (f )e f xR f yR
Member End Displacement e
e
δ Lx 0 0 ∆1x cos φ sin φ L − sin φ cos φ 0 0 ∆1y δy 2 → (δ)e = [Γ](∆)e δR = 0 φ φ 0 cos sin ∆ x x 2 δR − φ φ 0 0 sin cos ∆ y y z
Member Stiffness Matrix in Global Coordinate (F)e = [Γ]T (f )e = [Γ]T [k ]e (δ ) e = [Γ]T [k ]e [Γ](∆ ) e (F ) e = [K ] e ( ∆ ) e
− cos 2 φ − sin φ cos φ cos 2 φ sin θ cos φ − sin 2 φ [K 11 ]e [K 12 ]e sin 2 φ − sin φ cos φ EA sin θ cos φ e [K ] = = L − cos 2 φ − sin θ cos φ cos 2 φ sin φ cos φ [K 21 ]e [K 22 ]e − sin 2 φ sin φ cos φ sin 2 φ − sin φ cos φ
5.1.2 Global Stiffness Equation
− (F1 ) m (3) z
Nodal Equilibrium
− (F 2 ) m ( 2 )
n-th joint
− (F 2 ) m ( 4 )
− (F1 ) m (1)
− (F1 ) m ( 5) Pm
m-th joint
Pm
i-th member (P) m = (F 1 ) m (1) + (F 2 ) m ( 2) + (F 1 ) m ( 3) + (F 2 ) m ( 4) + (F 1 ) m (5) =
nm ( m )
∑ (F
1 or 2 m ( k )
)
k =1
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
59
School of Civil, Urban & Geosystem Eng., SNU
Pn
(F 2 ) i
n-th joint 2
1 i-th member
Pm m-th joint
0 0 nm (1) 1 or 2 1( k ) ∑ (F ) M M (P) k =1 1 i M ( ) F I M p p nm ( m ) 1 or 2 ( ) m k m ( P ) = ( P ) = ∑ (F ) = ∑ M = ∑ M k =1 i =1 (F 2 )i i =1 0 M M nm ( q ) q M 1 or 2 q ( k ) (P) M ( ) F ∑ 0 k =1 0 1
(F1 )i
0 M m-th row 0 1 i p (F ) M 2 i = ∑ [E]i (F)i = [E](F ) (F ) i =1 I n-th row M 0
(F ) 1 M [E] = [[E]1 L [E]i L [E] p ] , (F) = (F) i M p (F )
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
60
School of Civil, Urban & Geosystem Eng., SNU z
Compatability Condition un
n
(∆ 2 )i
2
m
um
1
(∆ 1 )i
i-th member ( ∆ 1 )i I 0 u m (∆ 1 )i = u m , ( ∆ 2 )i = u n → ( ∆ )i = 2 i = (∆ ) 0 I u n u1 M u m 1 i (∆ ) 0 L I L 0 L 0 i M = [C]i (u) (∆ ) = = ( ∆ 2 ) i 0 L 0 L I L 0 u n M n-th column u q
m-th column
(∆ )1 [C]1 u1 (∆ ) = M = M M = [C](u) (∆ ) p [C] p u q Compatibility Matrix z
Contragradient (P)T ⋅ (u) = (F)T ⋅ (∆) → (P)T ⋅ (u) = (F)T [C](u) → ((P)T − (F)T [C])(u) = 0 for all possible (u) → (P)T = (F)T [C] (P) = [C]T (F) = [E](F) → [C]T = [E]
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
61
School of Civil, Urban & Geosystem Eng., SNU z
Unassembled Member Stiffness Equation (F )1 [K ]1 M M (F ) i = 0 M M p (F) 0
z
L
0
L
L
M
L
L [K ] i L L
M
L
L
0
L
0 ( ∆ )1 M i → (F) = [K ](∆) 0 (∆) M M [K ] p (∆) p
Global Stiffness Equation (P) = [C]T (F) = [C]T [K ](∆) = [C]T [K ][C](u) (P) = [K ](u)
z
where
[K ] = [C]T [K ][C]
Direct Stiffness Method
[
[K ] = [C]T [K ][C] = [C]1 L[C]i L[C] p T
T
T
]
[K ]1 M 0 M 0
L L
0 M
L [K ] i L M L 0
1 0 [C] M i L 0 [C] L M M p L [K ] p [C]
L L
= [C]1 [K ]1 [C]1 + L + [C]i [K ]i [C]i + L + [C] p [K ] p [C] p T
T
0 M I iT i i [C] [K ] [C] = M M M 0 0 M I = M M M 0
0 M 0 i [K ] M 11 i [K ] I 21 M 0
T
[K 12 ]i 0 L I L 0 L 0 [K 22 ]i 0 L 0 L I L 0
0 M 0 i 0 L [K 11 ] M 0 L [K 21 ]i I M 0
m-th row
0 0 0 [ K ] i 11 L [K 12 ]i L 0 0 = 0 L [K 22 ]i L 0 0 [K 21 ]i n-th row 0 0 m-th column
0 0 0 0 0
0 [K 12 ] 0 0 0 i [K 22 ] 0 0 0 n-th column 0
i
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
62
School of Civil, Urban & Geosystem Eng., SNU 5.1.3 Example
P4, u4
P3, u3 2 L
2
1
P2, u2 1 P1, u1
P6, u6
3
P5, u5
3 2L
z
Member Stiffness Matrix − cos 2 φ − sin φ cos φ cos 2 φ sin φ cos φ − sin 2 φ [K 11 ]e [K 12 ]e sin 2 φ − sin φ cos φ EA sin φ cos φ = [K ] e = L − cos 2 φ − sin φ cos φ cos 2 φ sin φ cos φ [K 21 ]e [K 22 ]e − sin 2 φ sin φ cos φ sin 2 φ − sin φ cos φ
- Member 1: φ = 45o 1 2 1 EA 2 [K ]1 = 2 L − 1 2 1 − 2
1 2 1 2 1 − 2 1 − 2
−
1 2 1 − 2 1 2 1 2
1 − 2 1 − 2 1 2 1 2
1 2 1 EA − 2 [K ] 2 = 2 L − 1 2 1 2
1 2 1 2 1 2 1 − 2
1 2 1 2 1 2 1 − 2
1 2 1 − 2 1 − 2 1 2
- Member 2: φ =-45 o −
−
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
63
School of Civil, Urban & Geosystem Eng., SNU - Member 3: φ = 0o 1 EA 0 [K ] 3 = 2 L − 1 0 z
Equilibrium Equation
0 − 1 0 0 0 0 0 1 0 0 0 0
P4 P3
P2
P6 P5
P1
P1 = ( Fx1 )1 + ( Fx2 ) 3 P2 = ( F ) + ( F ) 1 1 y
2 3 y
P3 = ( Fx2 )1 + ( Fx1 ) 2 P4 = ( Fy2 )1 + ( Fy1 ) 2 P5 = ( Fx2 ) 2 + ( Fx1 ) 3 P6 = ( Fy2 ) 2 + ( Fy1 ) 3
P1 1 P2 0 P 0 3= P4 0 P5 0 P 0 6
0 1 0 0 0 0
0 0 1 0 0 0
[E]1
0 0 0 1 0 0
0 0 1 0 0 0
0 0 0 1 0 0
0 0 0 0 1 0
[E]2
0 0 0 0 0 1
0 0 0 0 1 0
0 0 0 0 0 1
1 0 0 0 0 0
[E]3
F 1 1 x Fy1 2 Fx 2 0 Fy 2 1 Fx1 0 Fy1 0 Fx2 0 Fy2 0 F 1 3 x F1 y2 Fx F 2 y
(P) = [[E1 ] [E 2 ] [E 2 ]](F) = [E](F)
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
64
School of Civil, Urban & Geosystem Eng., SNU z
Compatibility Condition u4 u3
u2
u6
u5
u1 ∆1 1 x 1 ∆1y 2 0 ∆ x 0 ∆2 y 2 0 ∆1x 0 1 ∆ y 0 ∆2 = 0 2x ∆ y 0 1 3 ∆ x 0 ∆1 0 2y ∆ x 1 ∆2 0 y
0 1 0 0 0 0 0 0 0 0 0 1
0 0 1 0 1 0 0 0 0 0 0 0
0 0 0 1 0 1 0 0 0 0 0 0
0 0 0 0 0 0 1 0 1 0 0 0
0 [C]1 0 0 0 u1 u1 0 u 2 1 u2 [C] 0 u 3 2 u 3 = [C] 0 u 4 3 u 4 [C] u 1 u 5 5 u 0 u6 6 1 [C]2 0 [C]3 0
(∆ ) = [C](u) [E1 ] = [C1 ]T , [E 2 ] = [C 2 ]T , [E 3 ] = [C 3 ]T → [E] = [C]T z
Global Stiffness Matrix
[K ] = [C]T [K ][C] = [C]1 [K ]1 [C]1 + [C] 2 [K ] 2 [C] 2 + L + [C]3 [K ]3 [C]3 T
T
T
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
65
School of Civil, Urban & Geosystem Eng., SNU 1 0 EA 0 [C1 ]T [K 1 ][C1 ] = 2 L 0 0 0 1 0 EA 0 = 2 L 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
0 1 0 0 0 0
0 1 0 12 0 2 1 − 1 0 2 1 0 − 2
1 2 1 2 1 − 2 1 − 2
0 0 1 EA [C 2 ]T [K 2 ][C 2 ] = 2 L 0 0 0 1 0 EA 0 = 2 L 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
0 0 0 0 0 1 0 0 0 0
0 0 EA 0 [C 3 ]T [K 3 ][C 3 ] = 2 L 0 1 0
0 0 1 0 0 0
0 0 0 1 0 0
0 0 0 0 1 0
1 2 1 0 − 2 1 0 − 2 1 0 2
0
0 0 0 0 0 1
1 0 0 0 0 0
0 1 0 12 0 2 1 − 1 0 2 1 0 − 2 1 2 1 − 2 1 2 1 2 −
1 2 1 − 2 1 2 1 2 −
0 1 0 12 0 − 2 0 − 1 0 2 1 1 2 1 2 1 2 1 2 1 − 2
−
0 1 1 0 0 0 − 1 0 0 0
1 − 2 1 1 0 − 0 1 2 1 0 0 2 0 0 1 2 1 2 0 0 1 2 0 0 EA − 1 = 2 L 2 0 0 1 − 2 0 0 0 0
1 2 1 2 1 − 2 1 − 2
1 2 0 1 − 0 2 1 0 − 2 0 1 2 0 0 1 2 0 1 − 2 = EA 0 1 2L − 2 0 1 2 0
1 2 1 2 1 2 1 − 2
−
1 2 1 2 1 2 1 − 2
−
1 2 1 − 2 1 2 1 2 −
1 2 1 2 1 2 1 − 2
0 0 1 0
0 0 0 1
0 0 0 0
0 0 0 0
1 1 1 − − 2 2 2 1 1 1 − − 2 2 2 1 1 1 − 2 2 2 1 1 1 − 2 2 2 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
−
0 − 1 0 0 0 0 0 0 0 1 0 1 0 0 0 0
0 0 0 1
0 0 0 0
0 0 0 0
0 0 0 0 0 0
0 0 0 0
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
0 0 0 0 0 0 0 0 1 1 1 1 − − 2 2 2 2 1 1 1 1 − − 2 2 2 2 1 1 1 1 − − 2 2 2 2 1 1 1 1 − − 2 2 2 2
1 0 0 0
0 1 0 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
66
School of Civil, Urban & Geosystem Eng., SNU 0 0 EA 0 = 2 L 0 1 0
0 0 0 0 0 1
1 0 0 0 0 0
0 1 − 1 0 0 0 1 0 0 0
1 1 2 2 1 1 2 2 1 EA 1 − − [K ] = 2 2L 2 1 1 − − 2 2 0 0 0 0 1 0 0 0 0 0 EA 0 0 0 + 2L 0 0 0 − 1 0 0 0 0 0 1 1 2 2 + 2 1 2 2 1 − EA 2 2 = L − 1 2 2 1 − 2 0
0 0 0 0
0 0 0 0
1 0 1 0 0 0 EA 0 = − 1 0 2 L 0 − 1 0 0 0
0 0 0 0
0 0 0 0 0 0
0 − 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 − − 2 2 2 2 EA 1 1 1 1 0 0 + 0 0 − − 2L 2 2 2 2 1 1 1 1 0 0 − 0 0 − 2 2 2 2 0 0 1 1 1 1 0 0 − − 0 0 0 2 2 2 2 0 0 0 0 0 0
1 1 − 2 2 1 1 − − 2 2 1 1 2 2 1 1 2 2 0 0 −
0 0 −1 0 0 0
0
0
0
0
1
0
0 1
2 2 1 2 2 1 − 2 2 1 − 2 2 0 0
− − 1 2 2 1 2 2 −
1
−
2 2 1
−
2 2 + −
1 2 2 1
2 2 1
2 2 1
2 2
1
−
2 2 1
2 2 1 1 − 2 2 2 2 1 1 + 2 2 2 2 1 2 2 1 − 2 2
1 2
0 −
1 2 2 1
2 2 1 1 + 2 2 2 1 − 2 2
0 1 2 2 1 − 2 2 1 − 2 2 1 2 2 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
67
School of Civil, Urban & Geosystem Eng., SNU z
Stiffness Equation
1 + 2 2 2 1 Unknown P1 2 2 1 P Unknown 2 − Known P3 EA 2 2 = L − 1 Known P4 Known P5 2 2 −1 Unknown P6 2 0 z
2 2 1 2 2 1 − 2 2 1 − 2 2 0 0
− −
1 2 2 1 2 2 1
− −
−
1 2 2 1
2 1
2 2 1+ 2
2 2 1 − 2 2
2 2 1 − 2 2
1
2 2
1 2
0
2 2
1
0
2 2 1
−
2 2 1 0
2
−
1
0 u1 1 u 2 2 2 u 3 1 u 4 − 2 2 u 5 1 u − 6 2 2 1 2 2 0
Known Known Unknown Unknown Unknown Known
Application of Support Conditions (Boundary Conditions) 1 + 2 2 2 1 P1 2 2 1 P2 − P EA 3 2 2 = P4 L − 1 2 2 P5 P −1 6 2 0
z
1
1 2 2 1 2 2 1 − 2 2 1 − 2 2 0 0
− −
1 2 2 1 2 2 1 2 0
−
1 2 2 1
2 2
− −
1
−
2 2 1
0
2 2 0 1
1 2
−
1 2 2 1
2 1
2 2 1+ 2
2 2 1 − 2 2
2 2 1 − 2 2
0 u1 1 u 2 2 2 u 3 1 u 4 − 2 2 u 5 1 u − 6 2 2 1 2 2 0
Final Stiffness Equation
1 P3 2 EA P4 = 0 L P 5 − 1 2 2
0 1 2 1 2 2
1 1 − 2 2 u 3 u3 2 L 1 u4 → u4 = 0 EA 2 2 u u 5 1 + 2 5 − 1 2 2 2 2
0 1 2 1 2 2
1 − 2 2 1 2 2 1+ 2 2 2
−1
P3 P4 P 5
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
68
School of Civil, Urban & Geosystem Eng., SNU 5.2 Beam Problems 5.2.1 Member Stiffness Matrix
R f yR δ y
f yL δ Ly
m R θR
m L θL
z
Force-Displacement Relation at Member Ends
mL =
4 EI e L 2 EI e R 6 EI e L 6 EI e R θ + θ + 2 δy − 2 δy Le Le Le Le
mR =
2 EI e L 4 EI e R 6 EI e L 6 EI e R θ + θ + 2 δy − 2 δy Le Le Le Le
M eL + M eR 6 EI e L 6 EI e R 12 EI e L 12 EI e R f = = 2 θ + 2 θ + 3 δy − 3 δy Le Le Le Le Le L y
6 EI 6 EI 12 EI 12 EI M eL + M eR = − 2 e θ L − 2 e θ R − 3 e δ Ly + 3 e δ Ry f =− Le Le Le Le Le R y
z
Transformation Matrix is not required
f →F, m→M z
,
δ→∆
,θ → Θ
Member Stiffness Matrix e
Fy1 12 L2 e 1 6 M EI e Le = L 12 e − Fy2 2 Le 6 2 Le M
e
6 Le
12 − 2 Le 6 4 − Le 6 12 − Le L2e 6 2 − Le
1 6 ∆ y Le 1 2 Θ or (F)e = [K ]e (∆ )e 6 − ∆2 Le y 4 Θ 2y
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
69
School of Civil, Urban & Geosystem Eng., SNU 5.1.2 Global Stiffness Matrix z
Nodal Equilibrium
Vi Mi
i-th Member
(i-1)-th Member
→ P i = (F) iR−1 + (F) iL 2 1 i M = ( M ) i −1 + ( M ) i
V i = ( Fy2 ) i −1 + ( Fy1 ) i
P 1 (F )1L R L 2 P (F ) 1 + ( F ) 2 M M P i −1 (F ) R + (F) L i −1 i−2 i R L P = (F ) i −1 + (F) i = [E]1 L [E]i L [E] p i +1 R L P (F ) i + (F) i +1 M M p R L P (F ) p −1 + (F) p p +1 R P ( ) F p
[
i-th row (i+1)-th row
z
Compatibility
0 0 M M (F ) L I iR = ( F ) i 0 M M 0 0
]
0 M 0 (F )iL = [E]i (F)i I (F)iR M 0 u 2 i +1
u2i −1 u2i
(i-1)-th Member
(F ) 1 M (F) = [E](F) i M (F ) p
u2i + 2
i-th Member
(∆1y )i = u2i −1 (Θ1 )i = u2i (∆2y )i = u2i +1
→
(∆ ) iL I 0 u i = (∆ ) iL = u i , (∆ ) iR = u i +1 → (∆ ) i = i +1 R 0 I ( ) ∆ u i
(Θ 2 ) i = u 2 i + 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
70
School of Civil, Urban & Geosystem Eng., SNU i-th column
u1 M (∆ ) iL 0 L I 0 L 0 u i = (∆ ) i = i +1 = [C]i (u) R I L L 0 0 0 ∆ ( ) u i M (i+1)-th column p +1 u (∆ )1 [C]1 u1 (∆ ) = M = M M = [C](u) (∆ ) [C] u p +1 p p z
Unassembled Member Stiffness Equation
(F)1 [K ]1 M M (F ) i = 0 M M (F ) p 0 z
L
0
L M L [K ] i L M L 0
L L L L L
0 ( ∆ )1 M 0 (∆) i → M M [K ] p (∆) p
(F) = [K ](∆)
Global Stiffness Equation (P) = [C]T (F) = [C]T [K ](∆) = [C]T [K ][C](u) (P) = [K ](u)
z
where
[K ] = [C]T [K ][C]
Direct Stiffness Method 0 0 0 M M M 0 [ K L 11 ]i [C]Ti [K ]i [C]i = 0 L [K 21 ]i M M M 0 0 0
0
M [K 12 ]i [K 22 ]i M 0
i-th column
0 M M L 0 L 0 M M 0 0 0
i-th row i+1-th row
i+1-th column
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
71
School of Civil, Urban & Geosystem Eng., SNU 5.2.3 Example
V 1 , w1 M 1 , θ1
z
V 3 , w3
V 2 , w2
V 4 , w4 M 3 , θ3
M 2 , θ2
M 4 , θ4
Equilibrium Equation V 1 = V1L
, M 1 = M 1L
V 2 = V1R + V2L , M 2 = M 1R + M 2L V 3 = V2R + V3L , M 3 = M 2R + M 3L V 4 = V4R
V 1 M 1 2 V M 2 (P) = 3 V 3 M 4 V M 4
1 0 0 0 = 0 0 0 0
0 1 0 0 0 0 0 0
0 0 1 0 0 0 0 0 [E]1
z
0 0 0 1 0 0 0 0
, M 4 = M 4R
0 0 1 0 0 0 0 0
0 0 0 1 0 0 0 0
0 0 0 0 1 0 0 0
[E]2
0 0 0 0 0 1 0 0
0 0 0 0 1 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 0 0 1 0 [E]3
V1L M 1L R 0V1 R 0 M 1 L 0 V2 0 M 2L = [E](F) 0V2R 0 M R 2 0 V L 3 1 L M3 R V3 M R 3
Compatability Condition
w1L = w1 , θ1L = θ1 , w1R = w 2 , θ1R = θ 2 w2L = w 2 , θ 2L = θ 2 , w2R = w 3 , θ 2R = θ 3 w3L = w 3 , θ 3L = θ 3 , w3R = w 4 , θ 3R = θ 4
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
72
School of Civil, Urban & Geosystem Eng., SNU w1L 1 θ1L R 0 w1 0 R θ1 0 w2L 0 L θ2 0 (∆) = R = w2 0 θ R 0 2 w3L 0 L 0 θ3 R 0 w3 0 θR 3 z
0 1 0 0 0 0 0 0 0 0 0 0
0 0 0 1 0 1 0 0 0 0 0 0
0 0 0 0 0 0 1 0 1 0 0 0
0 0 0 0 0 0 0 1 0 1 0 0
0 0 0 0 0 0 0 0 0 0 1 0
0 0 1 0 w 1 0 θ 2 0 w 0 θ 2 0 w 3 0 θ 3 0 w 4 0 4 θ 0 1
[C]1 w1 θ1 2 w [C]1 θ 2 = [C] 2 w 3 [ ] C 3 3 θ 4 w θ4 [C]3 [C]
= [C](u)
2
Unassembled Member Stiffness Matrix 0 (F)1 [K ]1 (F ) = (F ) 2 = 0 [K ] 2 (F ) 0 0 3
z
0 0 1 0 1 0 0 0 0 0 0 0
0 (∆) 1 0 (∆) 2 = [K ](∆) [K ]3 (∆) 3
Global Stiffness Equation (P) = [E](F) = [E][K ](u) = [C]T [K ][C](u) = [K ](u) 0 0 [C]1 [K ]1 (P) = [C]1T , [C]T2 , [C]T3 0 [K ] 2 0 [C] 2 (u) 0 0 [K ]3 [C]3 = [K ](u) = ( [C]1T [K ]1 [C]1 + [C]T2 [K ] 2 [C] 2 + [C]T3 [K ]3 [C] 3 )(u)
[
]
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
73
School of Civil, Urban & Geosystem Eng., SNU -
Member 1 1 0 0 EI 1 0 L1 0 0 0 0
-
0 1 0 0 0 0 0 0
12 L2 1 6 L1 12 EI 1 − L2 1 L1 6 L1 0 0 0 0 Member 2 0 0 1 EI 2 0 L2 0 0 0 0
0 0 0 1 0 0 0 0
0 0 0 0 EI 2 L2 0 0 0 0
0 0 0 0 0 0 0 0
12 0 2 0 L 0 6 1 L1 0 12 − 0 L2 1 0 6 0 L 1 6 12 − 2 L1 L1 6 4 − L1 6 12 − L1 L12 6 2 − L1 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0
0 12 2 0 L2 0 6 0 L2 0 12 − 1 L22 0 6 0 L 2 0 0 0 0 12 6 L22 L2 6 4 L2 12 6 − 2 − L2 L2 6 2 L2 0 0 0 0
0 0 0 0 1 0 0 0
6 L1
12 L12 6 4 − L1 6 12 − L1 L12 6 2 − L1 6 L1 2 6 L1 4 0 0 0 0 6 L2
−
0 1 0 0
0 0 1 0
0 0 0 1
0 0 0 0
0 0 0 0
0 0 0 0
0 0 = 0 0
0 0 0 0
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
0 0 0 0
0 0 = 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12 L22 6 4 − L2 6 12 − L2 L22 6 2 − L2 0 0 12 − 2 L2 6 − L2 12 L22 6 − L2 0 0
6 L1 2 1 0 6 0 − L1 0 4
−
0 0 6 L2 2 −
6 L2 4 0 0
6 L2 2 0 0 6 0 − L2 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
74
School of Civil, Urban & Geosystem Eng., SNU -
z
Member 3 0 0 0 EI 3 0 L3 1 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 0 0 1 0
0 12 2 0 L3 0 6 0 L3 0 12 − 0 L23 0 6 1 L 3
0 0 0 0 0 EI 3 L3 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 12 L23 6 L3 12 − 2 L3 6 L3
6 L3
12 L23 6 4 − L3 6 12 − L3 L23 6 2 − L3 0 0 0 0 6 L3 4 −
6 L3 2
−
0 0 0 0 12 − 2 L3 6 − L3 12 L23 6 − L3
6 L3 2 0 0 6 0 − L3 0 4 0 0 0 0 6 L3 2 6 − L3 4
0 0 0 0
0 0 0 0
0 0 0 0
1 0 0 0
0 1 0 0
0 0 1 0
0 0 = 0 1
Global Stiffness Matrix 12 I 6I1 12 I 1 6 I 1 0 0 0 0 − 31 2 2 L3 L1 L1 L1 1 6 4 6 2 I I I I1 1 1 1 0 0 0 0 − 2 L12 L1 L1 L1 12 I 6 I 12 I 1 12 I 2 6 I 1 6 I 2 12 I 2 6I 2 − 3 1 − 21 0 0 + − + − L1 L13 L32 L12 L22 L32 L22 L1 6I1 2 I1 6I1 6I 2 4 I1 4 I 2 6I 2 2I 2 0 0 − + + − L12 L1 L12 L22 L2 L2 L12 L2 E 12 I 2 6I 2 12 I 2 12 I 3 6 I 2 6 I 3 12 I 3 6 I 3 0 0 − 3 − 2 + 3 − 2 + 2 − 3 L2 L2 L32 L3 L2 L3 L3 L23 6I 3 2I 3 6I 2 2I 2 6I 2 6I 3 4I 2 4I 3 0 0 − 2 + 2 + − 2 L22 L2 L2 L3 L2 L3 L3 L3 12 I 3 6I 3 12 I 3 6I 3 0 0 0 0 − 3 − 2 − 2 L3 L3 L33 L3 6I 3 2I 3 6I 3 4I 3 0 0 0 0 − 2 2 L L L3 L3 3 3
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
75
School of Civil, Urban & Geosystem Eng., SNU z
Application of support Conditions
6I1 12 I 1 L3 L12 1 4I1 6I1 2 L1 L1 V 1 12 I 6 I − 3 1 − 21 M 1 L1 L1 2 6I1 V 2I1 M 2 2 L1 = E L1 3 V 0 0 3 M 4 0 0 V M 4 0 0 0 0
z
12 I 1 6I1 0 0 0 0 3 2 L1 L1 6I1 2I1 − 2 0 0 0 0 w1 L1 L1 12 I 1 12 I 2 6I1 6I 2 12 I 2 6I 2 θ1 + − + − 0 0 3 3 2 2 3 2 L1 L2 L1 L2 L2 L2 2 w 6I1 6I 2 4I1 4I 2 6I 2 2I 2 − 2 + 2 + − 2 0 0 2 L1 L2 L2 L2 L1 L2 θ 6I 12 I 6 I 3 w 3 12 I 6I 12 I 2 12 I 3 6I − 32 − 22 + 3 − 22 + 23 − 3 3 3 L2 L2 L2 L3 L2 L3 L3 L23 θ 3 6I 3 2 I 3 4 6I 2 2I 2 6I 2 6I 3 4I 2 4I 3 − 2 + 2 + − 2 w L22 L2 L2 L3 L2 L3 L3 L3 4 12 I 3 6I 3 12 I 3 6 I 3 θ − 3 − 2 − 2 0 0 L3 L3 L33 L3 6I 3 2I 3 6I 3 4I 3 − 0 0 L23 L3 L23 L3 −
Final Stiffness Equation 4 I1 4I 2 L + L 2 2 M 2 2 I 2 M 3 L2 4 = E V 0 M 4 0
2I 2 L2 4I 2 4I 3 + L2 L3 6I − 23 L3 2I 3 L3
0 6I 3 L23 12 I 3 L33 6I − 23 L3 −
0 2 2 I 3 θ 3 L3 θ 6 I 4 − 23 w L3 θ 4 4 I 3 L3
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
76
School of Civil, Urban & Geosystem Eng., SNU 5.3 Frame Problems 5.3.1 Member Stiffness Matrix f yR , δ Ry
f yL , δ Ly
mL , θL
mR , θR
f xL , δ Lx
z
f xR , δ Rx
Force-Displacement Relation at Member Ends - Beam action
mL =
4 EI e L 2 EI e R 6 EI e L 6 EI e R θe + θe + 2 δ y − 2 δ y Le Le Le Le
mR =
2 EI e L 4 EI e R 6 EI e L 6 EI e R θe + θe + 2 δ y − 2 δ y Le Le Le Le
f yL =
M eL + M eR 6 EI e L 6 EI e R 12 EI e L 12 EI e R = 2 θe + 2 θe + 3 δ y − 3 δ y Le Le Le Le Le
f yR = −
6 EI 6 EI 12 EI 12 EI M eL + M eR = − 2 e θeL − 2 e θeR − 3 e δ Ly + 3 e δ Ry Le Le Le Le Le
- Truss action EA R (δ x − δ Lx ) L EA f xR = (δ Rx − δ Lx ) L L R Ve = Ve = 0 f xL = −
z
Member Stiffness Matrix f xL 0 Ae L 12 I e fy 0 L2e 6I e mL 0 E Le = 0 −A f R Le e Ie 12 x 0 − L2e R 6I e 0 fy Le mR
0 6I e Le
− Ae
4I e
0
0 6I e − Le
Ae
2I e
0
0
0
0 12 I e − 2 Le 6I − e Le 0 12 I e L2e 6I − e Le
δ Lx 0 L 6 I e δ y Le θ L 2I e e or (f )e = [k ]e (δ )e 0 R 6 I e δ x − Le R 4 I e δ y θR e
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
77
School of Civil, Urban & Geosystem Eng., SNU Transformation Matrix
z
Vy
vy
vx
②
Vx
θ
① ①②
Vx = cos θvx − sin θv y V cos θ − sin θ 0 v x x Vy = sin θvx + cos θv y → Vy = sin θ cos θ 0 v y M 0 0 1 m M =m (V ) = [ γ ]T ( v) → ( v) = [ γ ](V ) z
Member End Force e
Fx1 cos θ − sin θ 1 Fy sin θ cos θ 1 0 0 M2 = 0 Fx 0 F2 0 y 0 2 M 0 0 z
0 0 0 0 0 0 1 0 0 0 cos θ − sin θ 0 sin θ cos θ 0 0 0
e
0 f xL L 0 f y 0 m L e T e R → (F) = [Γ] (f ) 0 f x R 0 f y 1 m R
Member End Displacement e
e
0 0 0 ∆1x δ Lx cos θ sin θ 0 L ∆1 − sin θ cos θ 0 0 0 0 δy y L 0 0 1 0 0 0 θ1 e e θ Re = 2 → (δ) = [Γ](∆) ∆ δx 0 0 cos θ sin θ 0 x 0 δR 2 0 0 0 − sin θ cos θ 0 ∆ y y θR 0 0 0 0 1 θ1 0 e
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU z
78
Member Stiffness Matrix in Global Coordinate (F)e = [Γ]T (f )e = [Γ]T [k ]e (δ ) e = [Γ]T [k ]e [Γ](∆ ) e [K11 ]e [K12 ]e (F) = [K ] (∆ ) where [K ] = e e [K 21 ] [K 22 ] e
z
e
e
e
Nodal Equilibrium & Compatibility The same as the truss problems.
z
Global Stiffness Matrix (P) = [C]T (F) = [C]T [K ](∆) = [C]T [K ][C](u) (P) = [K ](u)
z
where
[K ] = [C]T [K ][C]
Direct Stiffness Method The same as the truss problems.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
79
School of Civil, Urban & Geosystem Eng., SNU
Chapter 6 Buckling of Structures
P
P
θ
L
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
80
School of Civil, Urban & Geosystem Eng., SNU 6.0 Stability of Structures
Q
Q
θ
L
z
Stable state
KθL × L > QθL → KL > Q The structure would return its original equilibrium position for a small perturbation in θ.
z
Critical state
KθL × L = QθL → KL = Q
z
Unstable state
KθL × L < QθL → KL < Q The structure would not return its original equilibrium position for a small perturbation in θ.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
81
School of Civil, Urban & Geosystem Eng., SNU 6.1 Governing Equation for a Beam with Axial Force q
M
M+dM
w
Q V+dV
V
dw dx dx Q
z
Equilibrium for vertical force (V + dV ) − V + qdx = 0 →
z
Equilibrium for moment ( M + dM ) − M − Vdx + qdx
z
dV = −q dx
dx dw dM dw −P =0→ −Q =V dx dx dx 2
Elimination of shear force
d 2M d 2w − Q = −q dx 2 dx 2 z
Strain-displacement relation ε=−
z
d 2w du y+ 2 dx dx
Stress-strain relation (Hooke law) σ = Eε = − E
z
Definition of Moment
M = ∫ σydA = ∫ EεydA = − ∫ ( E A
z
d 2w du y+E 2 dx dx
A
A
d 2w 2 du d 2w − ) = − y E y dA EI dx 2 dx dx 2
Beam Equation with Axial Force
EI
d 4w d 2w + Q =q dx 4 dx 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
82
School of Civil, Urban & Geosystem Eng., SNU 6.2 Homogeneous Solutions z
Characteristic Equation for P > 0 w = e λx λx
e (λ + β λ ) = 0 → λ = ±β i, 0 z
4
2 2
where β 2 =
Q EI
Homogeneous solution
w = Aeβix + Be −βix + Cx + D z
Exponential Function with Complex Variable
i 2 2 i3 3 i 4 4 i5 5 i6 6 x + x + x + x + x +L 2! 3! 4! 5! 6! 2 3 4 (−i ) 2 (−i ) 3 (−i ) 4 (−i )5 5 (−i )6 6 e − ix = 1 − ix + x + x + x + x + x +L 2! 3! 4! 5! 6! 1 1 1 eix + e − ix = 2(1 − x 2 + x 4 − x 6 + L) = 2 cos x 2! 4! 6! i i i eix − e − ix = 2(ix − x 3 + x 5 − x 7 + L) = 2i sin x 3! 5! 7!
eix = 1 + ix +
eix = cos x + i sin x , e − ix = cos x − i sin x z
Homogeneous solution
w = A(cos βx + i sin βx) + B(cos β x − i sin βx) + Cx + D = ( A + B) cos β x + i ( A − B) sin β x + Cx + D = A cos β x + B sin β x + Cx + D z
Characteristic Equation for P < 0 w = e λx λx
e (λ − β λ ) = 0 → λ = ±β, 0 z
4
2 2
where β 2 =
Q EI
Homogeneous solution for P < 0
w = Aeβx + Be −βx + Cx + D eβ x + e − β x eβ x − e − β x = ( A + B) + ( A − B) + Cx + D 2 2 = A cosh β x + B sinh β x + Cx + D
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
83
School of Civil, Urban & Geosystem Eng., SNU Simple Beam
z
Q
Q
− Boundary Condition w(0) = A + D = 0 , w′′(0) = − Aβ2 = 0 → A = 0 w( L) = B sin β L + CL = 0 , w′′( L) = − Bβ2 sin β L = 0 → B = C = 0
− Characteristic Equation A = B = C = D = 0 → w = 0 (trivial solution) or n 2 π 2 EI βL = nπ → Q = , n = 1,2,3L L2 nπ w = B sin βx = B sin x L z
Fixed-Fixed Beam
Q
Q
− Boundary Condition w(0) = A + D = 0 w′(0) = β B + C = 0 w( L) = A cos βL + B sin βL + CL + D = 0 w′( L) = − Aβ sin βL + Bβ cos βL + C = 0 − Characteristic Equation 1 0 0 β cos βL sin βL − β sin β L β cos β L
1 0 0 1 0 1 A 0 0 1 0 1 0 B 0 β )=0 → Det ( = cos βL sin βL L 1 L 1 C 0 1 0 D 0 − β sin βL β cos βL 1 0
1 0 0 1 0 1 1 0 β β 0 1 0 β sin βL L = sin βL L 1 − cos βL cos βL sin βL L 1 β cos β L 1 0 − β sin βL β cos βL 1 − β sin βL β cos βL 1 0
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU
84
− β − (−β cos β L) − (−β(cos βL + β L sin β L) + β) = β(2 cos β L − 2 + β L sin βL) = 0 2 cos β L − 2 + β L sin β L = 2(cos βL − 1) + βL sin βL = βL βL βL + 2β L sin cos = 2 2 2 βL βL βL βL βL βL βL βL sin ( cos − sin ) = 0 → sin = 0 or cos − sin =0 2 2 2 2 2 2 2 2 − 4 sin 2
− Eigenvalues Symmetric modes sin
βL βL 4n 2 π 2 EI =0 → = nπ → Q = , n = 1,2,3L 2 2 L2
w(0) = A + D = 0 , w′(0) = w′( L) = β B + C = 0, w( L) = A + CL + D = 0 A + D = 0 → A = − D → w = A(cos
2nπ x − 1) for A ≠ 0 L
Anti-symmetric modes βL βL βL βL βL 8.18π 2 EI cos − sin =0→ = tan →Q = 2 2 2 2 2 L2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
School of Civil, Urban & Geosystem Eng., SNU z
85
Cantilever Beam Q
Q
− Boundary Condition w(0) = A + D = 0 w′(0) = β B + C = 0 M ( L) = − EIw′′( L) = − EI (− Aβ 2 cos β L − Bβ 2 sin β L) = 0 V ( L) = − EI Q=
d 3w dw −P =0 3 dx dx
n 2 π 2 EI , n = 1,2,3L 4 L2
6.3.Homogeneous and Particular solution w = wh + wp = A cos β x + B sin β x + Cx + D + wp EI
z
d 4 ( wh + w p ) dx 4
+Q
d 2 ( wh + w p ) dx 2
d 4 wp d 2wp d 4 wh d 2 wh = EI +Q + EI +Q dx 4 dx 2 dx 4 dx 2 d 4wp d 2 wp = EI +Q =q dx 4 dx 2
Four Boundary Conditions for Simple Beams w(0) = A + D + w p (0) = 0 , M (0) = − EIw′′(0) = − EI (− Aβ 2 + w′p′ (0)) = 0 w( L) = A cos β L + B sin βL + CL + D + w p ( L) = 0 M ( L) = − EIw′′( L) = − EI (− Aβ 2 cos β L − Bβ 2 sin β L + w′p′ ( L)) = 0 1 0 0 1 A w p ( 0) B w′′ (0) − β2 0 0 0 + p = 0 → KX + F = 0 cos βL L 1 C wp ( L) sin βL 2 2 − β cos β L − β sin β L 0 0 D w′p′ ( L)
z
The homogenous solution is for the boundary conditions, while the particular solution is for the equilibrium.
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
86
School of Civil, Urban & Geosystem Eng., SNU 6.4.Energy Method z
Total Potential Energy l
l
l
dw dw 1 d 2w d 2w 1 Π = ∫ 2 EI 2 dx − Q ∫ dx − ∫ wqdx 2 0 dx 2 0 dx dx dx 0
∆
L−∆
L=
∫ ds = ∫ 0
L−∆
∫
L= L−∆+
0
z
L−∆
1 + ( w′) dx ≈
L−∆
2
0
1 ( w′) 2 dx → ∆ = 2
∫ 0
L−∆
∫ 0
Q
1 (1 + ( w′) 2 )dx 2 L
1 1 ( w′) 2 dx ≈ ∫ ( w′) 2 dx for ∆ << L 2 20
Principle of the Minimum Potential Energy 1 d 2 (w e + w ) d 2 (we + w ) 1 d (we + w ) d (we + w ) EI dx Q dx − ∫ ( w e + w )qdx − 2 2 ∫ 2 ∫0 2 dx dx dx dx 0 0 l
Πh =
l
l
l
l
l
1 d 2 we d 2 we d 2 w e dw e dw e d 2 w dw e dw e )dx − ∫ w qdx + ∫ ( )dx EI Q EI Q = ∫( − − 2 0 dx 2 dx dx dx dx dx 2 dx 2 dx 2 0 0 l
− ∫ w qdx + 0
l
1 d 2w d 2 w dw dw ( )dx EI Q − 2 ∫0 dx 2 dx dx dx 2
l
l
d 4 we d 2 we 1 d 2w d 2 w dw dw Q q ) dx ( EI Q )dx = Π + ∫ w ( EI + − + − 2 ∫0 dx 2 dx dx dx 4 dx 2 dx 2 0 e
l
= Πe + z
1 d 2w d 2 w dw dw ( EI − Q )dx 2 ∫0 dx 2 dx dx dx 2
The principle of minimum potential energy holds if and only if l
∫( 0
z
d 2w d 2 w dw dw EI − Q )dx > 0 for all possible w dx dx dx 2 dx 2
The principle of the minimum potential energy is not valid for the following cases. l
∫( 0
d 2w d 2 w dw dw EI − Q )dx ≤ 0 for some w dx dx dx 2 dx 2
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
87
School of Civil, Urban & Geosystem Eng., SNU z
The critical status of a structure is defined as l
∫( 0
z
d 2w d 2 w dw dw EI − Q )dx = 0 dx dx dx 2 dx 2
Approximation n
− Approximation of displacement: w = ∑ ai g i i =1
− Critical Status l
n
n
l
n
0
i =1
n
l
n
∫ (∑ ai g i′′)EI (∑ a j g ′j′ )dx − P ∫ (∑ ai g i′ )(∑ a j g ′j )dx = i =1
0
n
n
j =1
l
n
j =1
n
n
n
n
∑∑ ai ∫ g i′′EIg i′′dxa j − Q∑∑ ai ∫ g i′g ′j dxa j = ∑∑ ai K ij a j − Q∑∑ ai K ijG a j = i =1 j =1
i =1 j =1
0
i =1 j =1
0
i =1 j =1
(a ) (K − K )(a ) = 0 → Det (K − K ) = 0 T
z
G
T
G
Example - Simple Beam Q
− with a parabola: w = ax( x − l ) → g1′ = 2 x − l , g1′′ = 2 l
l
0
0
l
l
∫ g i′′EIg i′′dx = ∫ 2 EI 2dx = 4 EIl l
4 1 3 2 2 2 3 ∫0 g i′EIg i′dx = ∫0 (2 x − l ) dx = ∫0 (4 x − 4 xl + l )dx = ( 3 − 2 + 1)l = 3 l 1 12 EI Det (4 EIl − Q l 3 ) = 0 → Qcr = 2 3 l
− with one sine curve: w = a sin
(exact :
EI π 2 EI = 9.86 2 , error = 22%) 2 l l
πx πx π πx π → g1′ = cos , g1′′ = ( ) 2 sin l l l l l
π 4 π 4 l 2 πx ∫0 g i′′EIg i′′dx = EI ( l ) ∫0 sin l dx = EI ( l ) 2 l
l
π 2 π 2 l 2 πx ∫0 g i′EIg i′dx = ( l ) ∫0 cos l dx = ( l ) 2 l
l
π l π l π 2 EI Det ( EI ( ) 4 − Q( ) 2 ) = 0 → Q = 2 (exact ) l 2 l 2 l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
88
School of Civil, Urban & Geosystem Eng., SNU z
Example – Cantilever Beam Q
Q
− with one unknown: w = ax 2 → g1′ = 2 x , g1′′ = 2 l
l
0
0
∫ g i′′EIg i′′dx = ∫ 2 EI 2dx = 4EIl ,
l
l
0
0
2 ∫ g i′EIg i′dx = ∫ 4 x dx =
4 3EI Det (4 EIl − Q l 3 ) = 0 → Qcr = 2 3 l
4 3 l 3
π 2 EI EI (exact : = 2.46 2 , error = 22%) 2 4l l
− with two unknowns: w = ax 2 + bx 3 → g1′ = 2 x , g 2′ = 3 x 2 , g1′′ = 2 , g 2′′ = 6 x l
l
l
4 6 9 G G = ∫ 6 x 3dx = l 4 , K 22 = ∫ 9 x 4 dx = l 5 K11G = ∫ 4 x 2 dx = l 3 , K12G = K 21 3 4 5 0 0 0 l
l
l
K11 = ∫ 4dx = 4l , K12 = K 21 = ∫ 12 xdx = 6l , K 22 = ∫ 36 x 2 dx = 12l 3 2
0
0
4l 6l 2 1 40l 3 Det ( EI 2 −Q 4 3 30 45l 6l 12l
0
45l 4 4l − 40l 3 α 6l 2 − 45l 4 α 1 Q = 0,α = )=0→ 2 5 4 3 5 30 EI 6l − 45l α 12l − 54l α 54l
(4l − 40l 3 α)(12l 3 − 54l 5 α) − (6l 2 − 45l 4 α) 2 = 0 → 4l 4 − 52l 6 α + 45l 8 α 2 = 0 EI EI 26l 6 ± 26 2 l 12 − 180l 12 26 ± 22.27 0.0829 1.0727 α= = = or → Qcr = 2.487 2 or 32.181 2 8 2 2 2 l l l l 45l 45l
Qexact = 2.49
EI EI (error = 1.2%) or Qexact = 22.19 2 (error = 45%) 2 l l
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
89
School of Civil, Urban & Geosystem Eng., SNU 6.5 Approximation with the Homogeneous Beam Solutions z
Homogeneous Solution of Beam u2i −1
u 2 i +1 P
u2 i + 2
u2i i-th member
δ Ly θ Lz i L L R R w = N 1δ y + N 2 θ z + N 3 δ y + N 4 θ z = ( N 1 , N 2 , N 3 , N 4 ) R = ( N )(∆ i ) = (N)(∆ i ) δy R θz 2 3 2 3 2 x 2 x3 3x 2x 2x x 3x 2 x3 N1 = 1 − 2 + 3 , N 2 = x − + 2 , N3 = − 2 , N4 = − + 2 L L L L L L L L 6.5.1 Beam Analysis z
Total Potential Energy l l p p l d 2 wi dwi dwi 1 d 2 wi 1 Π=∑ ∫ EI dx − ∑ Q ∫ dx − ∑ ∫ wi qi dx 2 dx dx dx 2 i =1 2 i =1 0 i =1 2 0 dx 0 p
p
=∑ i =1
l l p p l d 2 wi dwi T dw i 1 d 2 wi T 1 − − ( ) ( ) ( wi ) T qi dx EI dx Q dx ∑ ∑ 2 ∫ ∫ 2 ∫0 dx 2 2 dx dx dx i =1 i =1 0 0 l
l
l
p p 1 1 d 2N T d 2N dN T dN T T = ∑ (∆ i ) ∫ ( 2 ) EI dx(∆ i ) − ∑ Q(∆ i ) ∫ ( ) dx(∆ i ) − ∑ (∆ i ) T ∫ (N) T qi dx 2 dx dx dx i =1 2 i =1 2 i =1 0 dx 0 0 p
=
p 1 p 0 T G ∆ − ∆ − ( ) ([ ] [ )( ) (∆ i ) T (f i ) K K ∑ i ∑ i i i 2 i =1 i =1
=
p 1 T p (u) ∑ [C i ]T ([K i0 ] − [K Gi ])[C i ](u) − ∑ (u) T [C i ]T (f i ) 2 i =1 i =1
p 1 T p (u) ∑ [C i ]T [K i ][C i ](u) − (u) T ∑ [C i ]T (f i ) 2 i =1 i =1 1 T = (u) [K ](u) − (u) T (P ) 2
=
12 L2 i 6 EI Li K i0 = i Li − 12 2 Li 6 Li
6 Li 4 −
6 Li 2
12 L2i 6 − Li 12 L2i 6 − Li
−
6 Li 2 , 6 − Li 4
6 5L i 1 K Gi = Q 10 6 − 5 Li 1 10
1 10 2 Li 15 1 10 L − i 30
6 5Li 1 10 6 5Li 1 − 10
−
1 10 Li − 30 1 − 10 2 Li 15
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
90
School of Civil, Urban & Geosystem Eng., SNU z
Principle of Minimum Potential Energy for Q < Qcr Π=
n n 1 T 1 n (u) [K ](u) − (u)T (P) = ∑ ui ∑ K ij u j − ∑ ui Pi 2 2 i=1 j =1 i =1
∂Π 1 n ∂u i = ∑ ∂u k 2 i =1 ∂u k
z
n
∑ K ij u j + j =1
n n ∂u j 1 n u K − u i Pi ∑ i ∑ ij ∂u ∑ 2 i =1 j =1 i =1 k
n
=
1 1 n 1 n 1 n K u + u K − P = K u + K ki u i − Pk ∑ kj j 2 ∑ ∑ kj j 2 ∑ i ik k 2 j =1 2 j =1 i =1 i =1
=
n 1 n 1 n K kj u j + ∑ K kj u j − Pk = ∑ K kj u j − Pk = 0 for k = 1, L , n → [K ](u) = (P) ∑ 2 j =1 2 j =1 j =1
Calculation of the Critical Load
Det ([K ]) = Det ([K 0 ] − [K G ]) = 0 z
Frame Members
f xL L fy mL E =( f R Le x R fy mR
Ae 0 0 Ae 0 0
0 12 I e L2e 6Ie Le 0 12 I e − 2 Le 6Ie Le
0 6Ie Le
Ae
4Ie
0
0 6Ie − Le
Ae
2Ie
0
0
0
0 12 I e − 2 Le 6Ie − Le 0 12 I e L2e 6I − e Le
0 0 6 0 5 Le 1 2 I e 0 10 − P 0 0 0 6 6Ie 0 − − 5Le Le 1 0 4Ie 10 0 6Ie Le
P = EA
0 1 10 2 Le 15 0 1 10 L − e 30
0 0 0 0 0 0
0 6 − 5 Le 1 10 0 6 5Le 1 − 10
δ Lx 0 δL 1 y 10 L L − e θe 30 ) 0 1 δR − x 10 2 Le R δ 15 y θR e
δ Lx − δ Rx Le
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
91
School of Civil, Urban & Geosystem Eng., SNU 6.6 Nonlinear Analysis of Truss R f yR δ y
f yL δ Ly
L f xL δ x
z
f xR δ Rx
Force – Displacement relation at Member ends EA R (δ x − δ Lx ) L EA R f xR = (δ x − δ Lx ) L δ Ry − δ Ly R L R fy = − fy = fx l f xL = −
z
Member Stiffness Matrix
e
1 f xL L fy EA 0 = f xR l − 1 f yR 0
0 − 1 0 0 0 0 δ Rx − δ Lx + 0 1 0 l 0 0 0
0 0 0 1 0 0 0 − 1
0 δ Lx 0 − 1 δ Ly 0 0 δ Rx 0 1 δ Ry 0
e
(f )e = ([k ]e0 + f xR [k ]eg )(δ )e = ([k ]e0 + p e [k ]eg )(δ )e z
Equilibrium Analysis (f )ie = ([k ]e0 + pie [k ]eg )(δ )ie (F)e = [Γ]T (f )e = [Γ]T [k ]e (δ )e = [Γ]T ([k ]0e + pie [k ]eg )[Γ](∆ )e = [K ]e (∆ )e
z
Successive substitution (F) ie−1 ≈ [Γ]T ([k ]e0 + pie−1 [k ]eg )[Γ](∆ ) e = [K ]ie−1 (∆ ) e
(P) = ([K ]0 + [K ( pie−1 )]G )(u)i
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr
92
School of Civil, Urban & Geosystem Eng., SNU z
Newton-Raphson Method
(f )ie = ([k ]e0 + pie [k ]eg )(δ )ie = ([k ]e0 + ( pie−1 + ∆pie )[k ]eg )(δ ie−1 + ∆δ ie ) = ([k ]0e + pie−1[k ]eg )δ ie−1 + ([k ]0e + pie−1[k ]eg )∆δ ie + ∆pie [k ]eg (δ ie−1 + ∆δ ie ) ≈ ([k ]e0 + pie−1[k ]eg )δ ie−1 + ([k ]e0 + pie−1[k ]eg )∆δ ie + ∆pie [k ]eg δ ie−1 = (f )ie−1 + ([k ]e0 + pie−1[k ]eg + [k ]eσ )∆δ ie ([k ]0e + pie−1[k ]eg + [k ]eσ )∆δ ie = (f )ie − (f )ie−1 = (∆f )ie
([K ]0 + [K ( pie−1 )]G + [K ]σ )∆u = (∆P)i
∆pie [k ]eg δ ie−1
0 0 0 1 EA = 0 l 0 0 − 1 L EA δ y = 2 l δR y
e
0 δ Lx 0 − 1 δ Ly ∆δ Rx − ∆δ Lx 0 0 δ Rx l R 0 1 δ y i −1 0
e
e
∆δ Lx L − δ Ry ∆δ y ( ) 1 0 1 0 − R 0 ∆δ x L ∆δ R − δy y i i −1 0
0 EA R L e 1 = 2 (δ y − δ y ) i −1 0 l − 1
e
0 0 ∆δ Lx 0 − 1 0 ∆δ Ly 0 0 0 ∆δ Rx 0 1 0 ∆δ Ry i 0
= [k ]eσ ∆δ ie
Structural Analysis Lab. Prof. Hae Sung Lee, http://strana.snu.ac.kr