Nonlinear Beam Deflection - transverse force

Method of derivation:
Curvature = Moment / Flexural Rigidity
Elliptic Integrals
Principle of Elastic Similarity
End result:
x, & y as a function of F
F = K1q + K2*q3 => F=F(q)

Identification of terms

EI

s = distance along the beam from node 1 to node 2

r(s) = radius of curvature at s
(s) = angle of the beam at s with respect to the horizontal
M(s) = moment at s
E = modulus of elasticity
I = moment of inertia about the z-axis (out of plane)
x and y are the in-plane horizontal and vertical axes with the origin at node 1.
Fo = external force at node 2
L = beam length
= projected beam shortening
L

ds

dy

dx
2
L-

Fig 1

Beam Curvature due to M, F

The curvature at each point s along the beam is defined to be

2

1
d y
dy
curvature

1
2
r (s)
dx
dx
2

d ( s ) M ( s )

ds
EI

If an external transverse force is applied at the node 2, the curvature at s is

M ( s ) Fo ( L x )

EI
EI

(1)

Change of variables
Equation (1) contains two independent variables, s and x. We eliminate x by
differentiating (1) with respect to s, then re-integrate it.
Noting dx/ds=cos() , differentiating (1) gives

d 2 Fo dx Fo

cos
2
EI ds
EI
ds

(2)

d 2
1 d
Applying the identity ds 2 d 2 ds , the integration of (2) leads to

1 d

2 ds

Fo

sin C
EI

(3)

Boundary conditions
To solve for the integration constant in (3) we apply the following boundary
condition at the end node 2

d
ds

o
sL

(4)

since the moment vanishes at node 2. Therefore, equation (3) becomes

ds

2 Fo
sin o sin
EI
0

ds L
0

(5)

d
2 Fo
sin o sin
EI

Now, we want to end up with a solution of the form

0 0 Fo , EI , L

(6)

Transformation to Elliptic Integrals

Elliptic integrals have been well characterized (see M. Abramowitz, A. Stegun,
Handbook of Mathematical Functions). Transforming equation (6) to an elliptic
integral form provides us with their benefits.
Introducing two new variables p and defined as

1 sin
(7)
1 sin o
We seek substitutions for d, sin(), and sin() in equation (6).
1
p 1 sin o
2
2

sin 2

and

(8)

We find the differential element d by the differentiation of equation (8) with

respect to and then using the trigonomic identity

cos 1 sin 2 2 p sin 1 p 2 sin 2

This leads to

4 p 2 sin cos
2 p sin 1 p sin
2

2
sin

2
p
1
And from (7) & (8) we see that
o

(11)

(9)

(10)

sin 2 p 2 sin 2 1
(12)

The Elliptic Integral Form

Substitution of equations (10-12) into equation (6) results in
2 ( p )

F0 L
d

2
2
EI
1 ( p ) 1 p sin
2

(13)

where the lower and upper limits of the integral are determined from equation (11)
1

0 1 sin

2 p

0 2

(14)

Associating equation (13) with the complete and incomplete elliptic integrals of
the first kind, we have

F0 L2
K ( p) F p,1
EI

(15)

Realization of Force and Displacement, SIMO

Given 0, its required force is found by plugging (7) & (14) into (15).
Equation (15) is the nondimensionalized external force at node 2.
After finding the nondimensional force, the relationship dy ds sin
gives us the nondimensional transverse deflection, LL

Equation (1) with the BCs

1
2

x 0

EI
sin d
F0 L2 sin( 0 ) sin( )

0 , and

d
ds

x 0

M0

EI

(16)

2 F0 sin 0
F0 ( L x )
EI

gives the nondimensional projected beam shortening as

EI
2
1
2
(
2
p
1)
2
L
F0 L

(17)

Nonlinear Stiffness
The outputs of equations (15-17) as functions of the input 0 are plotted as Os
below. The solid curves are 3rd order, piecewise continuous, polynomial fits.
To obtain nonlinear stiffness, we first assume that the curves can be approximated
by a third order polynomial of the form
2
3
F0 L2
q
q
q


(18)
A B C D
EI
L
L
L
were q stands for , x, y, etc. Seeing that the solution has odd symmetry, we only
need to keep constants B and D, which also eases iterative computations.
Absorbing the material and geometric terms into B and D, respectively K 1 and K2,
we find that
3

F0 K1,i q K 2,i q

(19)

The coefficients of these polynomial curves are associated with the linear
stiffnesses K1,i and the cubic nonlinearities K2,i.
In order to maintain accuracy, (19) is applied in a continuous piecewise fashion by
dividing the total physical range into, say, 4 intervals, were i(q)=[1,4]. (see plot)

Plot of Nonlinear Stiffness vs Theory

F0 L2
EI

0
/2

Fig 2

Principle of Elastic Similarity

Using the geometric properties of the above elliptic integrals together with the
principle of elastic similarity we find an intuitive relationship. This type of
analysis was first reported by [R. Fay, A new approach to the analysis of the
deflection of thin cantilevers, Journal of Applied Mechanics 28, Trans. ASME,
83, Ser. E (1961) 87].
From similarity:
From Fig 1, using your imagination, assume that the beam,
which remains anchored at node 1, continues arching to the
left of node 1 until it reaches a new node 0. Node 0 is such
that it is oriented normal to the horizontal axis (see Fig 3).
The length of the section from node 0 to 1 is L 2. Also
imagine that the shape of L2 is the shape that would be
required if L1 is to remain unaltered when node 1 is no longer fixed.

Geometric Properties
From geometry: From the geometric properties of the elliptic functions, we
make the following identifications.

L L1 L2 K ( p) / k
sin( B / 2) p sin( B )

F0
EI

p sin( / 2)

/ 2 0
1 B
d
L2
k 0 1 p 2 sin 2 ( )
1
L1 L L2 K ( p) F ( p, B )
k
h 2p/k
L1 h cos( B )

{[ 2 E ( p, B ) F ( p, B )] [2 E ( p ) K ( p )]} / k b

Identification of terms

F0

B=/2
1

(x,y)

L1

L2

b
2

L 1 -

0
h
Fig 3

Nonlinear Beam Deflection - Fx, Fy, & M

Here we attempt answer the question:
Given the angles of node2, determine x, y, Fx, Fy, and M.

Method of derivation:
By intuition, jumping straight to elastic similarity and
the geometric properties of elliptic integrals.
End result:
x, & y as a function of Fx, Fy, and M
F = K1q + K2*q3 => F=F(q)

A New Approach to Analyzing Nonlinear Beams with

Coupled Fx, Fy, and M Forces.
The method presented above will now be extending to a much more
general situation.
The geometric quantities listed below are self-described in a new
geometric representation shown in Fig 4.
Given 0 and B, the following holds

L3 h cos( 0 )
0 B
p sin( B / 2) / sin( B ) sin

L2 k F ( p, 0 ) F ( p, B )
M L3 F0
Fx F0 cos( B )
Fy F0 sin( B )

F0 L2
EI

The Modified Geometric Representation

F0

B
1

L2

L1

0
2

L3

0
h=2p/k

B
Fig 4