0% found this document useful (0 votes)
96 views10 pages

Science PDF

The document presents an exact dynamic stiffness theory for analyzing axial-torsional buckling in structural frames. It derives the exact stiffness matrix for interactive axial-torsional buckling of thin beams using a power series expansion method. This allows structural frames to be analyzed using stiffness or finite element methods. The theory considers second order effects of axial force and initial torque. It obtains the consistent natural boundary moments and forces to ensure symmetry of the dynamic stiffness matrix as required by the reciprocal theorem. Interaction diagrams of vibration frequency, axial force and torque are studied in detail.

Uploaded by

Nada Mady
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
96 views10 pages

Science PDF

The document presents an exact dynamic stiffness theory for analyzing axial-torsional buckling in structural frames. It derives the exact stiffness matrix for interactive axial-torsional buckling of thin beams using a power series expansion method. This allows structural frames to be analyzed using stiffness or finite element methods. The theory considers second order effects of axial force and initial torque. It obtains the consistent natural boundary moments and forces to ensure symmetry of the dynamic stiffness matrix as required by the reciprocal theorem. Interaction diagrams of vibration frequency, axial force and torque are studied in detail.

Uploaded by

Nada Mady
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

ARTICLE IN PRESS

Thin-Walled Structures 46 (2008) 110 www.elsevier.com/locate/tws

Exact dynamic stiffness for axial-torsional buckling of structural frames


A.Y.T. Leung
Building and Construction, City University of Hong Kong, Hong Kong Received 7 March 2007; received in revised form 22 June 2007; accepted 17 August 2007 Available online 27 September 2007

Abstract A dynamic axial-torsion buckling theory is proposed for analysis in structural mechanics. Second order effects of the axial force and torque are considered. The consistent natural boundary moments and forces are given to ensure the symmetry of the dynamic stiffness matrix in fullling the requirement of the reciprocal theorem. The exact dynamic stiffness matrix is obtained for the rst time using power series expansion. Generally distributed axial force can be analyzed without difculty. It is pointed out that non-uniform sections may not be effectively analyzed due to the convergent problem. The interaction diagrams due to vibration frequency, axial force and torque are studied in details. r 2007 Elsevier Ltd. All rights reserved.
Keywords: Beam column; Power series; Exact dynamic stiffness; Interactive axial-torsional buckling

1. Introduction In linear axial buckling theory of columns, the axial rigidity is assumed to be indenite so that the axial displacement is ideally zero always. Flexural displacements are bifurcated from zero to some undetermined nite values at buckling. The governing equations do not contain the axial displacement. Likewise, in linear torsional buckling theory, the torsional rigidity is assumed to be indenite so that the twist is ideally zero always and exural displacements are bifurcated from zero to some undetermined nite values at buckling. The governing equations do not contain the angular displacement of twist. The paper derives the exact stiffness matrix for interactive axial-torsional buckling for a doubly symmetric thin beam so that structural frames can be analyzed using the stiffness or the nite element methods. The power series method has been used extensively to exactly solving a set of governing ordinary linear differential equations with variable coefcients. Kim et al. [1] presented a paper on the exact stiffness matrices for exural-torsional/lateral buckling and vibration analysis of Timoshenko thin-walled beam based on the power series
Tel.: +852 2788 7600; fax: +852 2788 7612.

method. Eisenberger [2] used power series to obtain the exact dynamics stiffness for non-uniform members. Leung et al. [310] published a number of papers on the power series solutions for exact stiffness matrices. In the paper, the exact stiffness matrix including both the second order axial force and initial torque is presented for the rst time for the interactive axial-torsional buckling analysis of framed structures. Step beams and nonuniformly distributed axial forces are considered. Kim et al. [1] assumed that the applied end torque is shared equally in the x- and y-axis as initial shear stress. It is ne if the bending rigidities in the two axes are equal as proposed by Ziegler [14]. When the bending rigidities are different, the share should be in the ratio Ix:Iy rather than 1=2 : 1=2. The algorithm is very efcient. Global interaction diagrams can be constructed easily. Finally, if scaled properly, the three-dimensional (3D) curves can collapse in one single curve. It is advantageous in engineering design to have one curve for the buckling interaction among vibration, compression and torque.

2. Governing equations We shall exclude initial stress due to bi-axial bending and shall apply the classical assumption of thin beams. The

E-mail address: Andrew.leung@cityu.edu.hk 0263-8231/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tws.2007.08.012

ARTICLE IN PRESS
2 A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110

y, v

x, u P L z P L l

In the presence of axial force P and inertia force o2rA due to harmonic vibration of frequency o, the total potential energy becomes 8 9T >u> > > > > > > > v > > > > > > > > > > > Z l > u0 > < = 1 2
0

Fig. 1. A beam element with second order axial force and torque.

2 6 6 6 6 6 6 6 6 6 6 6 6 6 4

> v0 > > > > > > > > > > 00 > >u > > > > > > > > > : 00 ; v o2 rA 0 0 0 0 0 o2 rA 0 0 0 0 0 0 P 0 0 I xL J 0 0 0 P
IyL J

0 0 0
I yL J

strain energy U for a beam element shown in Fig. 1 when subject to initial stress caused by initial torque L can be found in [14] when axial force was absent and when Ix Iy. Let the displacement at a point (x,y) away from the neutral axis be uT [ux uy uz] and the angle of twist be y. The displacement gradient is given by 9 8 0 8 09 > w xu00 yv00 > > u > > > > > = = < < qu 0 u0 yy 0 ; u;x u;z > > > qz > > > > 0 > ; ; : : v0 xy0 y 8 09 > v > > > = < and u;y y0 . > > > > ; : 0 where z is the neutral axis, a prime denotes differentiation with respect to z; u and v are vibration amplitudes of displacements along the x- and y-axis of the element of length l; EIy and EIx are the respective exural rigidities and J Ix+Iy. Consider the initial stresses due to the initial torque L only, t0 zy xL yL and t0 . zx J J

EI y 0

0 8 9 >u> > > > > > > > v > > > > > > > > > > 0> > > <u = > v0 > > > > > > > > > > > > u00 > > > > > > > > > : 00 ; v

7 0 7 7 7 I xL 7 J 7 7 7 0 7 7 7 0 7 7 5 EI x

dz,

in which z is the neutral axis, a prime denotes differentiation with respect to z; u and v are vibration amplitudes of displacements along the x- and y-axis of the element of length l; EIy and EIx are the respective exural rigidities and J Ix+Iy. If one partitions Eq. (1) according to [15], 8 9T 2 Z l > r > A00 < = 1 6A r0 U 4 10 2 0 > 00 > : ; A r 20 where frg & ' u A01 A11 A21 38 9 A02 > r > < = A12 7 r0 dz, 5 > > A22 : r00 ;

(2)

These satisfy the identity Z L xt0 yt0 dA. zy zx The strain energy due to initial stresses are given by, for example, Washizu [16], Z Z 1 t0 uT u;j dAdz Us ij ;i 2 where repeated index implies summation. Therefore, ! Z LI y 00 0 LI x 0 00 u v u v dz Us J J The share of L is Ix:Iy which becomes 1:1 only when Ix Iy. 22

, etc., then, after integration by parts v twice for equilibrium, Z dU


0 l

drT A22 D4 A12 A21 D3

A20 A11 A02 D2 A10 A01 D A00 r dz h i T drT S0 dr0 S1 0,


b

in which Dd/dz, subscript b denotes evaluation at the boundaries, Si, i 0,1 are the consistent boundary

ARTICLE IN PRESS
A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110 3

generalized forces given by S0 A10 A11 A20 D A12 A21 D2 A22 D3 r and S1 A20 A21 D A22 D2 r, and the governing equation is A22 D4 A12 A21 D3 A20 A11 A02 D2 A10 A01 D A00 r 0 Explicitly, the governing equations are EI y u0000 Lv000 Pu00 o2 rAu 0 and EI x v0000 Lu000 Pv00 o2 rAv 0,

body forces, the initial torque L must be constant along the element. 4 4. Exact solution by power series To reduce the number of variable parameter, nondimensionalization is advantageous. 8 9T >u> > > > > > > > v > > > > > > > > > > > Z l > u0 > < = 1 U 2 0 > v0 > > > > > > > > > > 00 > >u > > > > > > > > > : 00 ; v 3 2 o2 rA 0 0 0 0 0 7 6 6 0 o2 rA 0 0 0 0 7 7 6 7 6 I xL 7 6 0 0 P 0 0 J 7 6 7 6 6 7 IyL 6 0 0 0 P 0 7 J 7 6 7 6 IyL 6 0 0 0 EI y 0 7 7 6 J 5 4 I xL 0 0 EI x 0 0 J 8 9 >u> > > > > > > > v > > > > > > > > > > 0> > > <u = , > v0 > > > > > > > > > > 00 > >u > > > > > > > > > : 00 ; v 3 8 9T 2 0 0 0 0 > u > w 0 > > > > 6 7 > > > v > 6 0 w > > 0 0 0 0 7 > > 6 7 > > > > 7 > > 6 Z 1 > u0 > 6 0 < = 0 P 0 0 eL 7 EI y 7 6 3 7 6 > v0 > 6 0 L 2l 0 > > 6 0 0 P 0 7 > > 7 > > 6 > > 7 > 00 > 6 >u > 0 0 L 1 0 7 > > 4 0 > > 5 > > > > : 00 ; 0 0 0 0 eL e v 8 9 u> > > > > > > > > v > > > > > > > > > > 0> > > <u = d , z 9 > v0 > > > > > > > > > > 00 > >u > > > > > > > > > : 00 ; v where w o2rAl4, z z=l; P Pl 2 =EI y ; L Ll=EJ; and e EI x =EI y and a prime denotes differentiation with respect to z. We shall drop the over-bars for convenience in the subsequent study. Therefore, the

and the boundary moments and forces are M u EI y u00 I yL 0 v; J F u EI y u000 Pu0 Lv00 , (7a) I xL 0 u; J

M v EI x v00

F v EI x v000 Pv0 Lu00 . (7b)

The boundary moments of Bolotin [11] M u EI y u00 Lv0 and M v EI x v00 Lu0 are quite different and those of Chen and Atsuta [12] did not even contain initial torque. The natural boundary conditions (7a and b) are important for forming the exact element stiffness matrix that must be symmetrical due to the reciprocal theorem or the conservation of energy. Most authors considered exact solutions of simply supported beam columns where the variationally consistent boundary conditions are not needed. In deriving these equations, we assumed the absence of applied surface and body forces, that is, the column is loaded only at the boundaries. Ziegler [14] considered P and L separately for rotating shafts. The present investigation can be used to study the dynamic stability of rotating shaft under the inuence of axial compression where combine effect of P and L is important. 3. Non-uniform distribution of axial force If the axial force P is a function of z due to, e.g., self weigh, then the governing equations become EI y u0000 Lv000 Pu0 0 o2 rAu 0 and EI x v0000 Lu000 Pv0 0 o2 rAv 0, 8 with the same associated boundary generalized forces (7a and b). Extension to non-uniform cross sections is logical. However, it will be shown later that the power series expansion is not readily applicable for non-uniform sections due to convergence problem. In the absence of

ARTICLE IN PRESS
4 A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110

governing equations are u0000 1 eLv000 Pu00 wu 0 and ev0000 1 eLu000 Pv00 wv 0 and the boundary moments and forces are M u u00 Lv0 ; M v ev eLu ;
00 0

10

by the nodal displacements. To get the shape functions from the displacement functions (15), the following displacement boundary conditions are required for the nodal displacement vector d, dT u0 v0 u0 0 v0 0 u1 v1 u0 1 v0 1 , 2 v0 3 2 v0 3

F u u000 Pu0 1 eLv00 , F v ev Pv 1 eLu ,


000 0 00

(11a)

(11b) or in vector form ( ) ( 00 ) Mu u Lv0 M and Mv ev00 eLu0 ( ) ( 000 ) u Pu0 1 eLv00 Fu V . ev000 Pv0 1 eLu00 Fv Express the solutions in power series uz
n X k0

6 0 7 6 0 7 6 v 0 7 6 v 0 7 1 7r Cr for C 6 7 6 d6 6 v1 7 or r C d and v1 7 5 5 4 4 v0 1 v0 1 (16) ( uz vz ) vxr vxC1 d Nzd, (17)

12

uk1 zk and vz

n X k0

vk1 zk ,

(13)

where Nx vxC1 is the shape function matrix in the nite element sense. The exact stiffness matrix K is given by 9 8 > V0 > > > > > > > < M0 = Kd, (18) F > V1 > > > > > > ; : M1 > It can be shown that the stiffness matrix thus obtained is

Substitute in Eq. (10) and compare similar terms,

u5k and v5k

w k1 kp3 u1k 1 k2 p2 u2k 1 k2 kp1 u3k 1 eL1 k2 k3 kv4k , 1 k2 k3 k4 k w k1 kp3 v1k 1 k2 p2 v2k 1 k2 kp1 v3k 1 eL1 k2 k3 ku4k , e1 k2 k3 k4 k

(14)

It is obvious that uk,vk, k44, can be expressed in terms of u1 ; u2 ; u3 ; u4 ; v1 ; v2 ; v3 ; v4 . Back substituting into Eq. (10), it can be seen that the residue decays rapidly. The slowest term converges about (4+k) times the previous one. When k is about 10, each increment increases the number of signicant digit by one. Fifty terms should achieve more than machine precision in all cases. Therefore, the power series displacement solutions can be written as ( ) h i uz v1 x v2 x v8 x vz T u1 u4 v1 v4 vzr 15

symmetrical to fulll the requirement of the reciprocal theorem. 5. The power series failed for non-uniform sections To study the convergence of the power series method for non-uniform beam, let us consider a much simpler problem of harmonically vibrating beam whose governing equation is ev00 00 wv 0,
4

(19)

where w o2 rAl =EI, o is the frequency, rA is mass per unit length, EI is the nominal exural rigidity and e 1 fz3 to study the effect of non-uniformity due to linear varying depth. The power series solution is

vk5

wvk1 f 3 k1 k2 2 kvk2 3f 2 1 k2 2 k2 vk3 3f 1 k2 k2 3 kvk4 1 k2 k3 k4 k

(20)

The row vector v(z) contains the displacement shape functions and r the integration constants to be determined

That the numerator and denominator equally contain k to the power four poses a serious convergent problem.

ARTICLE IN PRESS
A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110 5

Back-substitute into Eq. (19) using 10 terms, the output residue is excessively large and convergence cannot be guaranteed. 6. Numerical examples Consider the interaction diagram between the buckling axial load P and the frequency w of a uniform cantilever column with rigidity ratio e 0.5 shown in Fig. 2 as the rst example. The Euler exural buckling load P p2/4 p [13] on the major axis is p accurately shown as P 1:5708 while on the minor axis P 1:5708 0:707 1:1106. In the subsequent examples, we use L; P1=2 ; w1=4 as the primary parameters so that the spread of the curves is more uniform. The buckling torque is dened as the maximum torque in the member at the rst elastic exural buckling. Fig. 3 shows the interaction diagram between the buckling torque and natural frequency. The torque decreases the natural frequency and at the natural frequency becomes zero indicating the static buckling torque L 1.178, c.f. Ziegler L 2.221/1.5 1.481. Therefore, when using the equal share assumption, the buckling load is 1.481/1.178

L=0 1.6 1.4 1.2 P1/2 1 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0.8 1 w1/4
Fig. 2. Interaction buckling diagram for rigidity ratio e 0.5.

1.2

1.4

1.6

1.8

1.257% or 25.7% higher. The second buckling torque is at L 1.779. Fig. 4 shows the interaction diagram between the buckling torque and compression. The torque decreases the buckling compression and at L 1.178, the buckling compression becomes zero indicating the pure buckling torque. The interaction diagram for dynamic axial-torsional buckling is given in Fig. 5 where the axial compression is taken as a parameter. The rst buckling occurs at p p P 2p 1.1106. Therefore, the lowest branch is disp 2 appeared when P is slightly over thep value. The second branch will also be disappeared when P is slightly over p 2 1.5708. The interaction diagram for dynamic axial-torsional buckling is given in Fig. 6 where the axial torque is taken as a parameter. The rst buckling occurs at L 1.178 and the second at L 1.779. Therefore, the lowest branch is disappeared when L is slightly over the rst value and the second branch will also be disappeared when L is slightly over the second value. The inuence of the axial torque to the interaction diagram is shown in Fig. 7 where the torque L is taken as a parameter. Obviously, the axial tension increases the natural frequency but the torque decreases it. There is a qualitative change between L 1.1 and 1.2, a range covers the buckling torque L 1.178. Fig. 7 shows the tensionfrequency relation just beyond the rst buckling torque. A second qualitative change between is shown between L 1.7 and 1.8, a range covers the second buckling torque L 1.779. The inuence of axial tension on buckling torque is shown in Fig. 8. It is obvious that the axial tension increases the buckling torque. The inuence of axial tension on the interaction diagram of frequency and torque is shown in Fig. 9 taking axial tension as a parameter. It is seen that axial tension increases both the natural frequency and buckling torque.

2.5 2 1.5 L 1 0.5 0

P1/2=0

2.5 2 1.5 L 1 0.5 0

w=0

0.2

0.4

0.6

0.8

1 w1/4

1.2

1.4

1.6

1.8

0.2

0.4

0.6

0.8 P1/2

1.2

1.4

1.6

Fig. 3. The interaction diagram between the buckling torque and natural frequency.

Fig. 4. The interaction diagram between the buckling torque and compression.

ARTICLE IN PRESS
6
P1/2=1.1 2.5 2 1.5 L 1 0.5 0 0.2 0.4 0.6 0.8 1 w1/4 2.5 2 1.5 L 1 0.5 0 P1/2=1.55 1.2 1.4 1.6 1.8 2 0.2 0.4 0.6 0.8 1 w1/4 1.2 1.4 1.6 1.8 2

A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110


P1/2=1.15

2.5 2 1.5 L 1 0.5 0

0.2

0.4

0.6

0.8

1 w1/4

1.2

1.4

1.6

1.8

Fig. 5. The interaction diagram for dynamic axial-torsional buckling.

L=1.1 1.6 1.4 1.2 P1/2 1 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0.8 1 w1/4 L=1.7 1.6 1.4 1.2 P1/2 1 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0.8 1 w1/4 1.2 1.4 1.6 1.8 2 1.2 1.4 1.6 1.8 2 P1/2 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0.2 0.4 0.6 0.8

L=1.2

1 w1/4

1.2

1.4

1.6

1.8

Fig. 6. The interaction diagram for dynamic axial-torsional buckling.

ARTICLE IN PRESS
A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110 7

Fig. 10 shows a comprehensive 3D interaction diagram of natural frequency, axial compression and torque. Two groups of curves are observed. The small group is for the minor bending axis and the lager group is for the major bending axis. It is interesting to nd that when one uses w=w0 1=4 and P=P0 1=2 instead of w1=4 and P1=2 as axes, where w0 and P0 are the respective compression free natural frequency and static buckling force, Fig. 10

collapses to a single curve as shown in Fig. 11. That will be useful for design curve proposes. Finally, consider a 2-section step column of length l 2 with clamp-free boundary condition as shown in Fig. 12. The system is clamped at the thicker section having exural rigidities of 4EIy and 4EIx while the thinner section has exural rigidities of EIy and EIx, respectively and Ix 2Iy. The interactive buckling diagram for the rst two buckling

L=0 3 2.5 2 1/2 P 1.5 1 0.5 0


1/2

3 2.5 2 1.5 1 0.5 0

L=1.1

0.5

1.5 w
1/4

2.5

0.5

1.5 w
1/4

2.5

L=1.2 3 2.5 2 1.5 1 0.5 0 1/2


1/2

L=1.3 3 2.5 2 1.5 1 0.5 0

0.5

1.5 w
1/4

2.5

0.5

1.5 w
1/4

2.5

3 2.5 2 1.5 1 0.5 0

L=1.7

L=1.8 3 2.5 2 1/2 P 1.5 1 0.5 0

1/2

0.5

1.5 w
1/4

2.5

0.5

1.5 w
1/4

2.5

Fig. 7. The inuence of the axial torque.

ARTICLE IN PRESS
8 A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110

2.5

w=0

2 L 1.5 L 1

1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0

0.5

0.5

1 w 1/4 1.5

0.8 1 0.2 0.4 0.6 P1/2

1.2 1.4 1.6

Fig. 10. Three-dimensional interaction diagram.

0.5

1.5 P
1/2

2.5

Fig. 8. Inuence of axial tension on buckling torque.

2.5 2 1.5 L

P1/2=0

2.5 2 1.5 L 1 0.5 0

P1/2=0.5

1 0.5 0 0.5 1 1.5 w1/4 2.5 2 1.5 L 1 0.5 0 2 2.5 3

0.5

1.5 w1/4

2.5

P1/2=1.5

0.5

1.5 w1/4

2.5

Fig. 9. Inuence of axial tension on the interaction diagram of frequency and torque.

ARTICLE IN PRESS
A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110 9

1 0.8 1

modes of the system is shown in Fig. 13 and 3D, interaction diagram is shown in Fig. 14. Similar phenomena as the single section column can be observed. 7. Conclusion The interactive axial-torsion buckling has been proposed and analyzed in structural mechanics. The consistent

(P

0.6

/P 0

) 1/2

0.4 0.2 0 0 0.1 0.2 0.3

0.4

0.5

0.6

0.7

0.8

0.9

w (w/

0)

1/4

1.5 1

Fig. 11. Collapse of all curves in Fig. 10 into one design curve.

y, v P L x, u z P l L

L 0.5 0 0 0.5 1 1.5 0 0.5


w1/4

1.5

Fig. 12. A 2-section column.

Fig. 14. Three-dimensional interaction diagram for the 2-section column.

Fig. 13. Interaction diagram taking L as a parameter.

1/ 2

ARTICLE IN PRESS
10 A.Y.T. Leung / Thin-Walled Structures 46 (2008) 110 [5] Leung AYT, Zhou WE. Dynamic stiffness analysis of nonuniform Timoshenko beams. J Sound Vibration 1995;181:44756. [6] Leung AYT, Zhou WE. Dynamic stiffness analysis of axially loaded nonuniform Timoshenko columns. Comput Struct 1995;56:57788. [7] Leung AYT, Zhou WE. Dynamic stiffness of laminated composite plates. Thin-walled Struct 1996;25:10934. [8] Leung AYT, Zhou WE, Lim CW, Yuen RKK. U Lee. Dynamic stiffness for piecewise non-linear Timoshenko columns by power seriesPart 1: Conservative axial force. Int J Num Meth Eng 2001;51(5):50529. [9] Leung AYT, Zhou WE, Lim CW, Yuen RKK. U Lee. Dynamic stiffness for piecewise non-linear Timoshenko columns by power seriesPart 2: Follower force. Int J Num Meth Eng 2001;51(5): 53152. [10] Zhu Bin, Leung AYT. Dynamic stiffness for thin-walled structures by power series. J Zhejiang Univ Sci A 2006;7(8):13517. [11] Bolotin VV. Nonconservative problems of the theory of elastic stability. Pergamon; 1963. p125. [12] Chen WF, Atsuta T. Theory of beam-columns, space behaviour and design, vol. 2. New York: McGraw-Hill; 1977. [13] Timoshenko SP, Gere JM. Theory of elastic stability. New York: McGraw-Hill; 1961. [14] Ziegler H. Principles of structural stability. 2nd ed. Basel: Birkhauser Verlag; 1977. [15] Leung AYT. The dynamic stiffness and substructures. London: Springer; 1993. [16] Washizu K. Variational methods in elasticity and plasticity. Pergamon; 1956.

natural boundary moments and forces are presented. These are important to ensure the symmetry of the stiffness matrix. The exact stiffness matrix is obtained for the rst time using power series expansion. Generally distributed axial force can be analyzed without difculty. It is pointed out that non-uniform sections may not be effectively analyzed due to the convergent problem. Acknowledgement The research is fully supported by the Research Grant Council of Hong Kong grant number 115706. References
[1] Kim NI, Fu CC, Kim MY. Stiffness matrices for exural-torsional/ lateral buckling and vibration analysis of thin-walled beam. J Sound Vibration 2007;299:73956. [2] Eisenberger M. Exact static and dynamic stiffness matrices for general variable cross-section members, AIAA. Journal 1990;28(6): 11059. [3] Leung AYT, Zhou WE. Dynamic stiffness of cylindrical shells. ISOPE 1993;4:4828. [4] Leung AYT, Zhou WE. Dynamic stiffness analysis of curved thinwalled beams. J Shock Vibration 1993;1:7788.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy