This action might not be possible to undo. Are you sure you want to continue?

Specialized Methods

5.1 INTRODUCTION

The ﬁnite strip method, pioneered in 1968 by Y. K. Cheung (1968a,b), is an efﬁcient tool for analyzing structures with regular geometric platform and simple boundary conditions. Basically, the ﬁnite strip method reduces a two-dimensional problem to a one-dimensional problem. In some cases computational savings by a factor of 10 or more are possible compared to the ﬁnite element method (Cheung and Tham, 1998). Originally, the ﬁnite strip method was designed for rectangular plate problems [similar to Levy’s solution; Timoshenko and WoinowskyKrieger (1971)]. Later, the ﬁnite strip method was extended to treat curved plates (Cheung, 1969b), skewed (quadrilateral) plates, folded plates, and box girders. Formulated as an eigenvalue problem, the ﬁnite strip method can be applied to vibration and stability problems of plates and shells with relative ease. Finite prism and ﬁnite layer methods were also introduced by Cheung and Tham (1998). These methods reduce three-dimensional problems to two- and one-dimensional problems, respectively, by choosing an appropriate choice of displacement functions. In addition, composite structures, such as sandwich panels with cold-formed facings, can be analyzed efﬁciently by coupling the ﬁnite strip method with the ﬁnite prism or ﬁnite layer methods (Cheung et al., 1982b; Chong, 1986; Chong et al., 1982a,b; Tham et al., 1982). The ﬁnite strip method has been modiﬁed through the use of spline functions to analyze plates of arbitrary shape (Cheung et al., 1982a;

153

154

SPECIALIZED METHODS

Chong and Chen, 1986; Li et al., 1986; Tham et al., 1986; Yang and Chong, 1982, 1984). Complicated boundary conditions can be accommodated. In general, ﬁnite strip methods based on spline functions are more involved than the ﬁnite strip method based on trigonometric series. However, they are still more efﬁcient than ﬁnite element methods, and they require less input and less computational effort.

5.2 FINITE STRIP METHOD

The ﬁnite strip method (FSM), ﬁrst proposed by Y. K. Cheung (1968a,b), is approaching a state of maturity as a structural analysis technique (Puckett et al., 1987; and Wiseman et al., 1987). Two comprehensive books on the method and its applications are available (Cheung and Tham, 1998; Loo and Cusens, 1978), as are papers on advances in the ﬁeld (Cheung, 1981; and others). The paper by Wiseman et al., (1987), summarizes developments in the methods of ﬁnite strips, ﬁnite layers, and ﬁnite prisms, and includes a review of 114 references. To examine the ﬁnite strip method, consider a rectangular plate with x and y axes in the plane of the plate and axis z in the thickness direction (Fig. 5.1). Let the corresponding displacement components be denoted by (u, v, w). Then similar to Levy’s solution (Timoshenko and Woinowsky-Krieger, 1971), for a typical strip (Fig. 5.1), the w displacement component is (Cheung and Tham, 1998).

y,v

1

L

2

t

x,u z,w b

Figure 5.1 Rectangular bending strips.

3) Four boundary conditions are needed to determine the coefﬁcients C1 to C4. Y(0) Y (0) Y(a) Y (a) 0 (5. they satisfy the relations a 0 Ym Yn dy 0 for m n (5. for both ends simply supported. that is.7) . r (5.6) Also. For example. . (5. they are orthogonal (Meirovitch. it can be shown that (Cheung and Tham. 3.2) is Y(y) C1 sin y a C2 cos y a C3 sinh y a C4 cosh y a (5. . Since the functions Ym are mode shapes. m 1. 1986). and geometric properties. . The functions Ym can be taken as basic functions (mode shapes) of the beam vibration equation d4Y dx 4 4 a4 Y 0 (5. 3. a 0 Y m Y n dy 0 for m n (5.5.4) Equations (5. . . 1998).2) is a parameter related to where a is the beam (strip) length and frequency.1) in which the functions ƒm(x) are polynomials and the functions Ym(y) are trigonometric terms that satisfy the end conditions in the y direction.2 FINITE STRIP METHOD 155 r w(x. 2. . . material.4) yield the mode shape functions Ym(y) sin m y a m 1. 2. The general solution of Eq. y) m 1 ƒm(x)Ym(y) (5. r.5) where m m .3) and (5. .

we have { 1} [C1] [C2] w1 1 { 2} 3x 2 3 w2 2 [(1 [(3x 2 2x ) x(1 x(x 2 2x x)] x 2)] (5. (5. (5.11) is equivalent to a one-dimensional beam element.10). thus minimizing computational storage and computational time. x/b. with nodal displacements wi and ﬁrst derivatives (nodal slopes) i w i .1) can be expressed as ƒm(x) [[C1] [C2]] { 1} { 2} (5.8) m where subscripts 1 and 2 denote sides 1 and 2 of the plate (strip). by Eq. equivalent to shape functions in one-dimensional ﬁnite elements.10) 2x 3) By Eqs. Similarly to the ﬁnite element method.156 SPECIALIZED METHODS The orthogonal properties of Ym result in structural matrices with very narrow bandwidths. For higher-order functions. we obtain ƒm(x) [(1 3x 2 w1 1 2x 3) x(1 2x x 2) (3x 2 2x 3) x(x 2 x)] w2 2 (5.9) Equation (5. Taking Let b be the width of a strip in the plate and let x the functions ƒm(x) as linear functions of x.11) Equation (5. we have ƒm(x) [(1 x) x] w1 w2 (1 x)w1 xw2 (5.8) and (5.9) is equivalent to a one-dimensional linear ﬁnite element (Chapter 4). [C1] and [C2] are interpolating functions. . which employs nodal displacements only. and { 1} and { 2} are nodal parameters. and considering nodal displacements only.8). the functions ƒm(x) in Eq. 2 w2 C1 1 x C2 x Hence. (5. we have 1 w1. respectively.

r s Ym[Ck] (5. My .1) and (5. the strain matrix (vector). for a strip subjected to bending. is given by Mx My Mxy w/ x 2 2 w/ y 2 2 2 w/ x y 2 { } (5. (5. For example. Mxy are moments per unit length.5)] and s is the number of nodal line parameters. (5.14).15) where Mx . 5. Other higher-order functions can be derived in a similar manner (Cheung and Tham. (5.8)].15).13) w m 1 k 1 [Nk]m{ k}m [N]{ } (5. Differentiating w. Eq.3 FORMULATION OF THE FINITE STRIP METHOD In general [see Eqs. { }. and substituting into Eq. Let [Nk]m Then.3 FORMULATION OF THE FINITE STRIP METHOD 157 which employs both nodal displacements and nodal slopes. (5.14) where [N] denotes the shape functions and { } denotes the nodal parameters.12) in which r is the number of mode shape functions [Eq. by Eq. The formulation of the ﬁnite strip method is similar to that of the ﬁnite element method (Section 4. 1998).2).12).5. the displacement function may be written as r s w m 1 Ym k 1 [Ck]{ k}m (5. we obtain the result . (5.

(4.19) [B]T[D][B] dv (5.18) where [D] is the elasticity matrix. (4.11). we obtain the stiffness matrix in the form . and (4.16) [B]m [N]m / x 2 [N]m / y 2 2 2[N]m / x y 2 2 (5. (5. the load vector is {F} A [N]T{q} dA (5.12)] for isotropic plane strain and plane stress problems.158 SPECIALIZED METHODS r { } in which [B]{ } m 1 [B]m{ }m (5. Minimization of the total potential energy gives (Cheung and Tham.17) Equation (5. (4.2 [Eqs.20). [K]{ } where [K] vol {F} (5. For a distributed load {q}. deﬁned in Section 4.21) Expanding Eq. 2000) and Eq. { }.16).25). (5.10). 1998).16) is similar to Eq. By Hooke’s law (Boresi and Chong.20) is the stiffness matrix and {F} is the load vector. is { } [D]{ } r [D] m 1 [B]m{ }m (5. the stress matrix (vector).

they can be integrated numerically using Gaussian quadrature or other numerical integration methods.23) For each strip (with s nodal line parameters).24) vol Similarly the basic element of the load vector (for distributed loads) is {Fi} T [Ni]m{q} dA (5.25) A For simple functions Eqs. (5.3 FORMULATION OF THE FINITE STRIP METHOD 159 [K] vol [[B]T[B]T 1 2 T [B]1 [D][B]1 [B]T][D][[B]1[B]2 r [B]T[D][B2] 1 [B] [D][B]2 T r [B]r] dv [B]T[D][B]r 1 dv [B] [D][B]r (5.25) can be evaluated in closed form (Cheung and Tham. the individual elements of matrix [k]mn are given by [kij]mn T [Bi]m[D][Bj]n dv (5. .22) T r vol [B] [D][B]1 [k]12 [k]r2 T r [k]11 [k]r1 where [k]1r [k]rr [k]mn vol T [B]m[D][B]n dv (5. [k11] [k]mn [ks1] [ks2] [kss] mn [k12] [k1s] In general.5.24) and (5. 1998). Alternatively.

2000. Each nodal line is free to move in the z direction and rotate about the y axis (Fig.1). .23). (5. (5.5). and (5. is (Cheung and Tham.10) are to be used.27) Ey t 3 x y 12(1 ) 12(1 ) For isotropic materials. 1998).1). 1998) Dx D1 0 D1 0 Dy 0 0 Dxy [B]m [D] (5.160 SPECIALIZED METHODS 5. 5.4 EXAMPLE OF THE FINITE STRIP METHOD In this section the ﬁnite strip method is applied to the analysis of a uniformly loaded plate simply supported on two parallel edges. (5. 6 2 (1 2x)Ym (2 3x)Ym 2 b b (1 3x 2 2x 3)Y m x(1 2x x 2)Y m 2 ( 6x 6x 2)Y m 2(1 4x 3x 2)Y m b 6 2 ( 1 2x)Ym ( 3x 1)Ym 2 b b (3x 2 2x 3)Y m x(x 2 x)Y m 2 (6x 6x 2)Y m 2(3x 2 2x)Y m b For orthotopic plates. Cheung and Tham. Eqs. The matrix [B]m .17) and (5. Thus Eqs. the elasticity matrix is (Boresi and Chong.26) where Dx D1 Ex t 3 12(1 x y) x Dy y Ex t 3 12(1 Ext 3 x y x y ) Dxy Gt 3 12 (5.

25).30) The load matrices {F}m for eccentric and concentric concentrated loads have been derived by Cheung and Tham (1998). . Eq. (5. (5.19) w1 1 {F1} {F2} 1 w2 2 (5. [k]mn reduces to [k]11 [k]22 n. to obtain the maximum moment at a known location (e.5. at the center for uniform load and under load point for concentrated load).28) By the orthogonal properties. 1998). x y and G 0 if m E 2(1 ) (5. 1998). In addition. Cheung and Tham.4 EXAMPLE OF THE FINITE STRIP METHOD 161 Ex Ey E. The load vector {Fm} can be computed by integrating Eq. the moments [Eq. 1984).. Higher-order strips have also been applied to plates (Cheung and Tham. Thus. 1998) have shown that with four strips (10 equations). It is a 4 1 matrix given by b/2 b2 /12 b/2 b2 /12 a {F} q sin 0 m y dy a (5. stiffened plates. These techniques are applicable to the analysis of curved bridges. 1998). With two degrees of freedom (w. a nodal line must pass through this location for good accuracy. considerable saving in computational labor is achieved (Yang and Chong.15)] are linear functions of x. and so on. greater accuracy was obtained than that by the ﬁnite element method with 25 nodes (75 equations). Finite strip formulations for curved and skewed plates have also been derived (Cheung and Tham.29) wr [k]rr r 1 {Fr} where the matrices [k]mm are 4 4 symmetrical matrices (Cheung and Tham.g. ) along the nodal line. Thus. In a convergence test applied to a uniformly loaded simply supported square plate. (5.

w1mn and w2mn are displacement parameters for side 1 (top) and side 2 (bottom). 1998. By linear theory.w Figure 5.v z .. Let z z/c. and Tham. that is.w y.v c z. the (x.2 Finite layers. respectively. The functions Xm and Yn are taken as terms in a trigonometric series. rectangular in planform. Then the lateral displacement component.162 SPECIALIZED METHODS 5. w. v) are linearly related to the derivatives of w. In this manner a three-dimensional problem is reduced to a one-dimensional problem with considerable saving in computer storage and computational time (Cheung. x. The resulting method is called the ﬁnite layer method (FLM). can be expressed as r t w m 1 n 1 [(1 z)w1mn zw2mn]Xm(x)Yn(y) (5. 1982b). is assumed to vary linearly in the z direction. consider Fig.2. Side 2 (bottom) . y) displacement components (u. w. The method was ﬁrst proposed by Cheung and Chakrabarti (1971). The displacement component. To illustrate the method. the philosophy of the ﬁnite strip method can be extended to layered systems. 5.31) in which. satisfying the boundary conditions. The ﬁnite layer method is useful for layered materials.5 FINITE LAYER METHOD By selecting functions satisfying the boundary conditions in two directions. Cheung et al.u b a Side 1 (top) y.u b a x.

we may write r t v m 1 n 1 r t [(1 [(1 m 1 n 1 z)v1mn z)u1mn zv2mn]Xm(x)Y n(y) (5. The displacement is u {ƒ} v r t [N]mn{ }mn m 1 n 1 [N]{ } (5.35) w where { }mn [u1mn. u2mn and v2mn are displacement parameters of sides 1 and 2.32) (5. w2mn]T (5. v1mn.37) The stress–strain relationship is . 2000) xx yy { } zz xy yz zx u/ x v/ y w/ z 1 –( u/ y v / x) 2 1 –( v / z w/ y) 2 1 –( w/ x u/ z) 2 r t [B]mn{ }mn m 1 n 1 [B]{ } (5.33) Thus. v2mn.36) The strain–displacement relationship is (Boresi and Chong.5.31). u2mn. w1mn. as with Eq. v1mn.5 FINITE LAYER METHOD 163 u v A B w x w y (5. (5.34) zu2mn]X m(x)Yn(y) u in which u1mn.

4.40) Hence.164 SPECIALIZED METHODS { } [ xx yy zz xy yz zx ]T [D][B]{ } (5.38) Proceeding as in Section 5. we obtain the stiffness matrix (Cheung and Tham.4. In formulating the ﬁnite prism method (FPM). we have the conditions a 0 b 0 c 0 [B]T [D][B]rs dx dy dz mn 0 for mn rs (5. (5. 1998) r ƒ m 1 ƒm(x. z) is a function of (x. the off-diagonal terms are zero. for prismatic members with rectangular planform.6 FINITE PRISM METHOD Similarly to Eq.39) For a simply supported rectangular plate. the three-dimensional problem can be reduced to a twodimensional problem by expressing the displacement function ƒ as (Cheung and Tham. z)Ym (5. and the stiffness matrix reduces to [B]T [D][B]11 11 T [B]12[D][B]12 a b 0 c 0 [K] 0 [B]T [D][B]1t 1t T [B]21[D][B]21 [B]T [D][B]rt rt dx dy dz (5. z) only and Ym are trigonometric functions of y.42) where ƒm(x. including nodal dis- .41) The remaining formulation is similar to that of Sections 5. 1998) a b 0 c [S] 0 0 [B]T[D][B] dx dy dz (5. 5.3 and 5. which satisfy the boundary conditions in the y direction.1).

if k k .5. the displacement of a point in the element. Using Eq. however. or curvilinear). the shape functions are given as follows: Corner nodes: Ck Midside nodes: 1 – 4 k (1 k )(1 k ) (5. z) sin km y (5. for a prismatic member with two ends simply supported.3).44) In general.6 FINITE PRISM METHOD 165 placements and shape functions as in the ﬁnite element method. the lateral (out-of-plane) displacement component is r w m 1 wm(x. k the nodal coordinates of the element. and ( k . Therefore. the displacement of a point within the element can be expressed in terms of the nodal displacements k as s k k k 1 (5. k . k) functions associated with a particular coordinate system (such as Cartesian. Then we can represent the local coordinates in terms of nodal coordinates in the form s k k 1 k (5. Similarly.46) and wkm are the mth term nodal displacements at the kth node.42).43) where s refers to the number of nodes of the element. k k . let .47) . it is convenient to use nodal coordinates. skew.45) in which s wm(x. be the local coordinates of an element. 5. the element is termed isoparametric. For an isoparametric six-node (ISW 6) model (Fig. k the nodal displacements. the Ck are the shape functions for the two-dimensional element (in the – plane). z) k 1 C kwkm (5. (5.

v) are r 6 u m 1 k 1 r 6 C kukm sin km y C kvkm cos km y m 1 k 1 (5.3 and 5.52) where . Ck 1 –(1 2 k )(1 2 ) (5. It is (Chong et al. 1982b) P Kijmn P T Bim PD PBjn d(vol) (5.3 ISW’6 model.49) The corresponding in-plane displacement components (u.4.51) v The stiffness matrix is developed in a manner similar to the development in Sections 5.48) Therefore..166 SPECIALIZED METHODS Figure 5. the lateral (out-of-plane) displacement component for the isoparametric six-node model is r 6 w C kwkm sin km y m 1 k 1 (5.50) (5.

The geometric transformation from the natural coordinate (x. and FPM) are especially adaptable for personal computers (Rhodes. with signiﬁcant savings in computational labor. AND FPM 167 Ci m y sin x L 0 0 P 0 Ci m m y sin L L 0 Ci m y cos x L Ci m y cos x L 0 0 0 Ci m y sin z L 0 Ci m m y cos L L Ci m y sin z L (5. FLM.52). and the stiffness matrix can be obtained accordingly (Cheung and Tham. AND FPM Owing to their narrow bandwidth and reduction in dimensions.7 APPLICATIONS AND DEVELOPMENTS OF FSM. FLM. ﬁnite layer. where the longitudinal beam stiffness could be derived separately and added to the structure stiffness. 1998). the superscript P indicates the ﬁnite prism model. ) can be carried out as in the standard ﬁnite element method. In Eq. 1987) and minicomputers. and ﬁnite prism methods (FSM.5. (1987) and Graves-Smith (1987). the ﬁnite strip. 1969a). One of the ﬁrst reported applications of the method was to orthotropic right box girder bridge decks (Cheung. P 5. z) to the local coordinate ( .53) Bim Ci m m y cos L L 0 Ci m y sin x L D is the elasticity matrix for isotropic materials. FLM. (5. Cheung suggested that the ﬁnite strip method could be used in a composite slab-beam system. Developments and applications are presented by Wiseman et al. This procedure is a reﬁnement to systems in which the action of longitudinal beams is .7 APPLICATIONS AND DEVELOPMENTS OF FSM.

. (1986) and to arbitrarily shaped general plates by Li . This condition was veriﬁed by comparison with theoretical and model test results for simply supported slab-beam models with varying slab thicknesses (Cheung et al. Other formulations for increasing the accuracy or the range of applicability of the ﬁnite strip method have been published. allowing extension of the ﬁnite strip method to plate bending problems with irregular boundaries. 1985. and the accuracy of the solutions was demonstrated for various boundary conditions. The other reﬁnement introduced an auxiliary nodal line within the strips. Although this requirement limited the analysis to plates with uniform properties across the section. to curved slabs by Cheung et al. convergence was shown to be faster for appropriate plate systems. Bucco et al. (1979) proposed a deﬂection contour method which allows the ﬁnite strip method to be used for any plate shape and loading for which the deﬂection contour is known. Two reﬁnements aimed at increasing the accuracy of ﬁnite strips in the in-plane transverse direction were developed by Loo and Cusens (1970).168 SPECIALIZED METHODS approximated by assuming orthotropic properties for the slab. 1984) used X-spline functions. Cheung further asserted that the beam would have to coincide with a strip nodal line. The plates were divided into strips where point and patch loads were applied. the cubic B-spline method was extended to skewed plates by Tham et al (1986). and techniques for approximating a plate with more irregularly shaped sides and ends were discussed. Brown and Ghali (1978) used a subparametric ﬁnite strip for the analysis of quadrilateral plates. 1986) for orthotropic sectorial plates and for rectangular plates under various loadings. An analytical method using a separation-of-variables procedure and series solution for the resulting differential equations was developed by Harik and his associates (1984. After initial application to ﬂat plates and box girder bridges by Cheung and Fan (1983). The buckling of irregular plates has been investigated by Chong and Chen (1986). A drawback of these reﬁnements is that they both increase the size of the stiffness matrix by 50%. (1982a). The solution was shown to converge correctly for a trapezoidal plate. Yang and Chong (1982. Other applications of the ﬁnite strip method have used spline functions for the boundary component of the displacement function. 1970). One reﬁnement ensued curvature compatibility at the nodal lines. A force method for introducing rigid column supports in the strips was also described. Extensive development of the ﬁnite strip method using the cubic B-spline as the boundary component of the displacement function was done by Cheung et al.

(1986). Characteristic functions for strips with varying end conditions have also been developed (Cheung and Cheung. and their book (Loo and Cusens. the longitudinal displacement function for a structure with pinned ends and no internal supports is a sine function. They represented longitudinal displacement functions as eigenfunctions of a continuous beam with the same span and relative span stiffnesses as the plate structure. The strips used were simply supported with respect to out-of-plane displacement and had similar in-plane longitudinal displacement functions. The application of the ﬁnite strip method to folded plates by Cheung (1969c) extended the method to three-dimensional plate structures.5. (1982a. The primary difference between this technique and the ﬁnite strip method for right bridge decks is the use of a polar coordinate system for the displacement function. 1978) contains an extensive treatment of the subject with many examples. This system is an acceptable approximation for many structures. A radial shell element was subsequently developed (Cheung and Cheung. Cheung and Chan (1978) used folded plate elements to study 392 theoretical bridge models in order to suggest reﬁnements to American and Canadian design codes. It was noted that a model in this coordinate system could be used to approximate a rectangular slab by specifying a very large radius for the plate. AND FPM 169 et al. the ﬁnite strip method has been readily adapted to free vibration and buckling problems. 1971b). These end conditions suggest a real plate system having end diaphragms or supports with inﬁnite in-plane stiffness. continuous over rigid supports. Cusens and Loo (1974) have done extensive work in the application of the ﬁnite strip method to the analysis of prestressed concrete box girder bridges. (1971). The formulation for frequency analysis was outlined by Cheung et al.b) to study free vibrations and to calculate the effects of different face temperatures on . Because of its efﬁciency and ease of input. For comparison. The analysis of cylindrically orthotropic curved slabs using the ﬁnite strip method was described by Cheung (1969c). The ﬁnite strip method has become popular for the analysis of slab girder and box girder bridges. including roof systems resting on walls and box girder bridges with plate diaphragms. 1971a) which allowed the analysis of curved box girder bridges supported by rigid end diaphragms. has been presented by Delcourt and Cheung (1978). Finite strips and prisms were combined by Chong et al. Shallow shells were analyzed by Fan and Cheung (1983) using a spline ﬁnite strip formulation based on the theory of Vlasov (1961).7 APPLICATIONS AND DEVELOPMENTS OF FSM. FLM. A study of folded plates.

Cheung et al. Yoshida (1971) was among the ﬁrst to develop the geometric stiffness matrix for buckling of rectangular plates. 5. Both local and overall buckling of H-columns with residual stress under axial load were studied by Yoshida and Maegawa (1978). Chong and Chen (1986) investigated the buckling of nonrectangular plates by spline ﬁnite strips. . The panel facings were modeled by ﬁnite strips and the panel core by ﬁnite prisms.54) P where is the frequency. The eigenvalue problem for free vibrations is characterized by the equation (Chong et al. Buckling of columns with I-sections was Figure 5. The ﬁnite strip method has been used as the basis for parametric and theoretical studies of column buckling. SM m is the in-plane mass matrix of the bendS B ing strip.4). PMm is P the mass matrix of the ﬁnite prism. 1982a) 2 S P Mm S P m 2 S B Mm S B m S 2 P S P m Mm S P m S B m P KP m KB m Km P m 0 (5. K m is the bending stiffness matrix of the bending P B strip. S m and S m are the in-plane and bending interpolation parameters of the bending strip respectively.170 SPECIALIZED METHODS foam-core architectural sandwich panels (Fig. M m is the bending mass matrix of the bending strip. 1998) for the analysis of local stability in plates and stiffened panels.4 Finite strips and prisms in a sandwich panel. Mahendran and Murray (1986) modiﬁed the standard ﬁnite strip displacement function to include out-of-phase displacement parameters. SK m is the in-plane stiffness matrix S B of the bending strip. (1982b) used the ﬁnite layer method to derive buckling loads of sandwich plates. PKm is the stiffness matrix of the ﬁnite prism. Przemieniecki (1972) used the same displacement function as in the ﬁnite strip method (Cheung and Tham. and P m are the interpolating parameters of the ﬁnite prism.. allowing shearing loads to be applied to the strips in the buckling analysis.

. an enhancement of the ﬂexibility and applicability of the ﬁnite strip method. (1991) developed a cubic B-spline ﬁnite strip method for the analysis of thin plates. P. REFERENCES Boresi. and Chong. P. Arbitrarily shaped sections were mapped approximately into circular cylindrical shells in which ﬁnite strips can be applied for discretization. was developed by Puckett and Gutkowski (1986). The compound strip method was extended to curved plate systems and was used for dynamic analysis by Puckett and Lang (1986). Sridharan and Graves-Smith (1981). This formulation added both longitudinal and transverse beams at any location in a strip and included the effect of column supports. The postbuckling behavior of columns has been the subject of papers by Graves-Smith and Sridharan (1979). 2000). With the compound strip method. 1990). eliminated the repetitive solution passes required for each redundant in a ﬂexibility analysis. Spline ﬁnite strips have also been used to analyze non-prismatic space structures (Tham.REFERENCES 171 studied by Hancock (1981). Hasegawa and Maeno (1979) determined the effects of stiffeners on cold-formed steel shapes using ﬁnite strips. A. The ﬁnite strip method was used for the free vibration and buckling analysis of plates with abrupt thickness changes and a nonhomogeneous winkler elastic foundation (Cheung et al. the analysis of continuous beams can be performed using simple displacement functions previously suitable only for single spans. K. which included the added stiffness of all elements associated with a beam. New York. A stiffness formulation. The interaction of the two modes of buckling for other prismatic columns was also investigated by Sridharan and Benito (1984) using the ﬁnite strip method.. allowing the total strip stiffness to be calculated by adding the element stiffnesses. and Graves-Smith and Gierlinski (1982). Wiley. Oscillatory convergence (Gibb’s phenomenon) is avoided. (1991). The strain energies for ﬂexure and torsion of the beams as well as for axial and ﬂexural deformation of the columns were derived in terms consistent with the strip displacement function. Spline series with unequal spacing allow local discretization reﬁnement near patch and concentrated loads. The compound strip method. using ﬁnite strips and a coordinate transformation. 2000. Gutkowski et al. Transient responses of noncircular cylindrical shells subject to shock waves were studied by Cheung et al.. Elasticity in Engineering Mechanics.

Shanghai. 1968b.. Assoc. S. C. Fan. Cheung. S.. MI. 97(EM3). London... Analysis of slab and girder bridges by the ﬁnite strip method. Rocky.. Y. pp. K. D. 1039–1044. Proc. K. ASCE. S.. S. K. and Chong. P. 182–203. J. Cheung. 94(EM6). 95–104. and Ghali. Cheung. T. Proc. and Chakrabarti. 29. Inst.172 SPECIALIZED METHODS Brown. T. Bridge Struct. 1–19. Y. Y. 174–185. Finite strip analysis of quadrilateral plates in bending. Cheung. K.. L. Y. Eng. Assoc. Mazumdar. 1968a.. Cheung. Tham. G. and Tham. K. Cheung. Crosby Lockwood. Struct. CRC Press. Comput.. K. 481–484. Sound Vib. Struct. Spline ﬁnite strip in structure analysis. 41–51.. 104(EM2). K. 1365–1378. V. Cheung. 1970. Proc.. 1969c. 1. Cheung. Eng. 5. 2.. K.. Y. 19–37.. Int. Y. Proc.. 5. K. 1998. layered plates. Eng. Div. 131–134. 1978. J. K. M. 2963–2979. and Chan. K. J... Proc. The ﬁnite strip method in the analysis of elastic plates with two opposite simply supported ends. Y. Int. Y. Y. and Sved. Div. Int. Civ. Analysis of curved box girder bridges by ﬁnite strip method. and Evans.. 1982b. Eng. 1969b. ASCE. M.. S. ASCE. R. Cheung. Build. S. and Reddy. Orthotropic right bridges by the ﬁnite strip method. Application of the ﬁnite strip method combined with the deﬂection contour method to plate bending problems. 1981. ASCE. 1971a. 31(I). Frequency analysis of certain single and continuous span bridges. Can. Proc. FL.. C. Static analysis of right box girder bridges by spline ﬁnite strip method. Finite strip evaluation of effective ﬂange width of bridge girders. Folded plate structures by the ﬁnite strip method.. Y. SP-26.. Mech. 1979. C. K. K.. Cheung. and Cheung. Mech. 311–323. 95(ST2). Buckling of sandwich plate by ﬁnite layer method. 704–709. K. Cheung. Eng. Proc. pp. and Wu. S. A. American Concrete Institute. J. Eng. 1982a.. Finite strip method in structural and continuum mechanics. Gordon & Breach. in Developments in Bridge Design and Construction. J. Cheung. S. Div. and Ghali. Div. Q. 1983. The Finite Strip Method.. 75. Civ. 1971. Natural vibrations of thin. Publ. pp. 188–199. J. Mech. Y. L. 1–7. . Cheung. Cheung. Cheung. Y. M. M. M. and Fan. Detroit. The analysis of cylindrical orthotropic curved bridge decks. Cheung. M. K. Cheung. Y. 1971. Finite strip method analysis of elastic slabs. C. in Proc. 1969a. A. G. Struct. Analysis of simply supported thick. 40. 18.. Civ. 1971b. Eng. (eds. Proc.). J. Pt. Bridge Struct. Struct. G. L. Y. D. Cheung. 325–337. London. Sci. 15(2). Comput. Bannister. Eng.. G. 827–830.. ﬂatwalled structures with different boundary conditions. K. Y. J. Boca Raton. J. K. Inst. 1978. Bucco. International Conference on Finite Element Methods... 1... Y. H. and Cheung.

Struct. C. Chong. 49. T. C. 24(3). J. 1982. C. 1974. P. Graves-Smith. Graves-Smith. Y. and Cheung. L. 251–273. A. 11(4). K.. J. K. 81(4). Div.. Y. Inst. Buckling of stiffened webs by local edge loads. 1983. Rhodes. in Recent Technical Advances and Trends in Design. 81. and Xiong. W. and Walker. G. Y.. G. Au. Free vibration of formed sandwich panel by ﬁnite-prism-strip method. Elastic collapse of thin-walled columns. Stockholm. pp. 718–729. Chong.. Yuan. Tham. S.. and Tham. K. International Association Bridge and Structural Engineers. Tham. IABSE International Colloquium on Thin-Walled Metal Structures in Buildings. 255–263. Hancock.. J. J. Y. J. K. P. 413–435. Eng. Cusens.. 1991. Gutkowski. Proc. P. R.. Delcourt. Eng.. ASCE. S. 1–16.. pp. P. 165–180. Research. Analysis of shallow shells by spline ﬁnite strip method. Interactive buckling in I-section columns. Applications of the ﬁnite strip method in the analysis of concrete box bridges. 305–318. Struct.. ASCE. C. 1357–1366.. Proc. 11(4). in Proc. Buckling of irregular plates by splined ﬁnite strips. 107. 15.. 3.. 1991. Cheung.. Essex. Sandwich panels with cold-formed thin facings. 339–348. and Cheung. Y.. Y. Vol... 1982a.. F. G. Elsevier Applied Science. Proc. J. Thin-Walled Struct.. J. 57. Y. Application of spline-ﬁnite strip method in the analysis of curved slab bridge. 1979. 2.. 36. Comput. A. Proc. T. K.). Barking. J. Finite Strip Method for the free vibration and buckling analysis of plates with abrupt changes in thickness and complex support conditions. Plate bending analysis by unequally spaced splines. (eds. J. 321– 324. Graves-Smith. K. Pt. J. and Puckett. (eds. and Loo. Struct. C. Chong. T. Civ.. Finite strip analysis of continuous folded plates. 111–124. and Li. 2. pp. Y. J. T.. L. and Chen. Struct.. K. 2000. and Gierlinski. Eng. Cheung. Keynote paper. and Sridharan. 1981. . Vol. Civ. and Zheng. R.. K.. 1987. L. M. Fan. R. Chong.. Thermal behavior of formed sandwich plate by ﬁnite-prism-method.. K. 575– 582. 1986. C. Transient response of cylindrical shells with arbitrary shaped sections. 89– 110. K. R.). Chen. International conference at the University of Strathclyde. Glasgow.. A. 534–536. Z. May. 1978. Y. 1986. J. 1982b. T. J. A.. Pt.REFERENCES 173 Cheung. Thin-Walled Struct. and Walker. and Cheung. Proc. K. Z. K. 1986. and Construction. Y. Thin-Walled Struct. Div. Sound Vib. Inst. England. R. D.. The ﬁnite strip analysis of structures. AIAAJ. Cheung. Rhodes. 108(ST6). Sweden. G. C. in Developments in Thin-Walled Structures. 5. L.

M. A. A. M. 1985.. Matrix analysis of local instability in plates. 112.. 1517–1532. 1989. G. 1986. I. 112(1). 1990. W. J. and Cusens. (ed. J. H. Eng. T. 554–568. G. C. Analytical solution to orthotropic sector. Tehachapi. J. J. 112(1). and Benito. McGraw-Hill. E. A reﬁned ﬁnite strip method for the analysis of orthotropic plates. 1978. in Recent Technical Advances and Trends in Design. 1986.. A. A. and Salamoun. Mech. K.. Proc. Compound strip method for the frequency analysis of continuous sector plates. Civ. Sridharan. International conference at the University of Strathclyde. W. 1972. and Lang. 1981. 5(5).. S. J. Sridharan. and Cusens. and Chong. A simple microcomputer ﬁnite strip analysis. 869–888. P. 1986. J. J. Rhodes. 4. pp. R. C. Postbuckling analyses with ﬁnite strips. Eng. Puckett. 112(5). A. 107(EM5). ASCE. Viewpoint Press. Mech. 235–246. Harik. Methods Eng. and Walker. and Maeno. Wiseman. R. 43–54. Eng. Eng. New York. and Graves-Smith. Mahendran. Loo. Compound strip method for the buckling analysis of continuous plates. Div. A. 1987. Columns: static and dynamic interactive buckling... 110. G. J. 209–216. G. Puckett. 5. Elastic buckling analysis of ideal thin walled structures under combined loading using a ﬁnite strip method. A. J. pp. L.. J. L. R. Thin-Walled Struct.. Glasgow.. Elements of Vibration Analysis. and Pashanasangi.. J.. Mech. Y. Mech. Tham. Hasegawa.174 SPECIALIZED METHODS Harik. Przemieniecki. in Dynamics of Structures. Int. Analytical strip solution to rectangular plates. 8(3). and Murray. 1986. and Lang. 385–402.). J. J. The Finite-Strip Method in Bridge Engineering. I. Li. 105–118. Eng. Design of edge stiffened plates. S. M. VA.. J.... 1986. A. 276–291. Harik. Eng. 165–182. Int. Compound strip method for analysis of plate systems. Inst. K. Numer. Eng. Application of spline ﬁnite strip method in the analysis of space structures. Y. Puckett. L. J. C. 10(3).. Y. Mech. and Gutkowski.. Meirovitch. 48. and Construction. J.. Spline ﬁnite strip analysis of general plates. S. Mech.. 111. 1979. and Tham. R. 679–680. Thin-Walled Struct. 121–138. Proc.. Puckett. R. 1375–1389.. American Society of Civil Engineers. J.. Research. Eng. E.. N. 110. Struct. 85–91.. 1984. L. (eds. I. E. J.. J. Thin-Walled Struct. J. Eng. Loo. stiffened panels and columns. Rhodes. Thin-Walled Struct. G. Struct. J. 1986. Y. 49–65. 1984.. CA. 1987. D. 329–362.. J. Reston.). 1970. J. S. Cheung.. Compound strip method for continuous sector plates. Curved bridge desks: analytical strip solution. 2nd ed... Roesset. . L.

1971. K. New York. and Maegawa. Lo. Jpn. and Woinowsky-Krieger. Arch. Schechtman. A. L. 1982. 1982. 130.S.. J. 6. and Chong. Reston. Reinf. China. Tham. Blackwell Science. Y. Y. Comput.. Y.C. Mater. 1–27. K. D. Z. 1986. 1984. 22(1). Struct. Timoshenko.. K. MA. H. 127–132. H. and Leung. 1... P. 161–171. Bending of skew plates by spline-ﬁnite-strip method.). Buckling analysis of plate structures by ﬁnite strip method.. 1996. S. K. American Society of Civil Engineers. pp. McGraw-Hill. Proc. Yang. Vlasov. D. and Cheng. K. Puckett. Proc. 1987. Department of Commerce. and Chong. P. S. Hong Kong. Yoshida. VA. 824– 829. Theory of Plates and Shells. J. Yang.BIBLIOGRAPHY 175 Tham. K. V. Struct. (ed. 1971. Shanghai. H. and Cheung. Malden. 16–28. Local and member buckling of H-columns. Finite strip method with X-spline functions. Cheung.. Flexural bending and axial compression of architectural sandwich panels by ﬁnite-prism-strip methods. 1978. University of Hong Kong. Chong. Y. K. M. Plast. Y. Finite Element Implementation. Li. Compos. Washington.. Comput. J. K... P. A. J. Soc.. Nav. International Conference on Finite Element Methods. G. L.. P. H. J. Roesset. L. 2nd ed.. . On ﬁnite strip method.. Y. pp. Yoshida. Department of Civil Engineering. Recent developments of the ﬁnite strip method.. 18(1). BIBLIOGRAPHY Cheung. in Dynamics of Structures. Struct. U. Wiseman. 31–38. 292–309.. Mech.. T. H. G.. Y. Y. P. and Chong. 1961. Thin-Walled Elastic Beams (trans.. M. W. S.

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd