Response of A Continuous Guideway On Equally Spaced Supports Traversed by A Moving Vehicle

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Journal of Sound and Vibration (1979) 67(2), 245-25 I

RESPONSE OF A CONTINUOUS GUIDEWAY ON EQUALLY


SPACED SUPPORTS TRAVERSED BY A MOVING VEHICLE

J. GENIN AND Y. I. CHUNG

School of Mechanical Engineering,


Purdue University, West Lafayette, Indiana 47901, U.S.A.

(Received 28 March 1979, and in revisedform 26 June 1979)

An algorithm is developed to analyze the dynamic response of a continuous guideway


resting on equally spaced supports being traversed by a moving vehicle. A critical speed,
one at which the amplitude of the guideway displacement becomes large, is determined.
The critical speed corresponds to the slowest possible speed at which a transverse wave can
propagate longitudinally in the guideway structure. A parametric study is also presented.

1. INTRODUCTION

In a previous study of the vehicle-guideway interaction problem we investigated the


dynamic response of the system when the guideway structure was composed of a series of
simply supported beams on equally spaced supports [l]. In this paper we look at the case
where the guideway is a continuous beam, again on equally spaced supports (see Figure
1).

Figure 1. Continuous beam on equally spaced supports.

Several studies exist on guideway structures modeled as a beam resting on an elastic


foundation in which a moving co-ordinate system freezes the guideway response per-
mitting a pseudo-static approach to the problem [2]. Also, several studies exist on the
steady state aspect of our guideway model in which wave motion plays an important role
in its response to external forces [3]. When such a guideway is subjected to the loading of
a discrete moving vehicle, the resulting boundary value problem is highly complex.
In what follows the behavior of the continuous guideway structure resting on equally
spaced supports subjected to a moving vehicle is investigated. An algorithm is derived
based upon the finite difference approach developed by Genin, Ginsberg and Ting [4,5).
Our method accounts fully for the inertial interaction between the vehicle and the
guideway, the significance of which was shown in our previous study [l]. Although this
245
0022-460X/79/220245+ 07 $02.00/O 0 1979Academic Press Inc. (London) Limited
246 J. GENIN AND Y. I. CHUNG

study is intended primarily for a high speed ground transportation system, there are many
other applications for the analysis contained herein.

2. EQUATIONS OF MOTION

The vehicle and its suspension system are modeled in Figure 2. For greater flexibility in
the resulting study the vehicle is assumed to have sprung and unsprung masses with a
primary and secondary suspension system. In terms of the quantities shown in Figure 2
the equations of motion for the vehicle are

M, d2Y,ldt2 + c,(dY,/dr-dY,ldr) + k(y, -yz) = 0, (1)


MS d2y,/dt2 + Mu d2y2/dt2 + k,(y2.-y,) = 0, 67
where
YlG) = YoV) + Y(X,f)l,=,(t)r (3)
ye(t) being the irregularities of the guideway and y(x, t)~,=,l~r~ the deflection of the
guideway directly under the vehicle. The primary suspension may be either a wheeled or a
levitated system and the secondary suspension may be a conventional (passive) or
hydraulic (active) suspension system. Thus our model covers most conceivable vehicle
types.

--L&L&L-
Figure 2. Schematic diagram of a vehicle model.

For the guideway motion, the integral formulation is

where c(x, t) is the influence function for the particular guideway model being in-
vestigated. A derivation of the influence function for an infinite beam on periodic supports
is omitted here, for brevity.
The loading function f(t) contains the vehicle dynamic force, which includes the inertial
loading as well as the static force : that is,
f(t) = (M,+M,,)g - MS d2y3/dt2 - Mu d2y2/dt2. (5)
The weighting function in equation (4), w(x, t), describes the time varying distribution of
the vehicle loading over the guideway. Hence, the complete interaction between the
vehicle and the guideway is contained in the foregoing formulation.
RESPONSE OF A VEHICLE-TRAVERSED GUIDEWAY 247
In order to facilitate the analysis, we shall now non-dimensionalize all the parameters
and variables as follows :

y,, = g13/48EI, j = YlYst, j, = YllYst, F2 = YzlYst, j3 = Y,lYs,,


fi = (MS + M,,)/ml, P = Mu/M, + Mu),

i, = q/2 &&qK, 6” = (1/71u)Jk,Io,

0, = W4 Jk,IWs+ M,) > 8, = lw,/m, ii = nvpo,,


g = 48/n41ik2, i, = (1/rd2 c&2 JmEI. 5 = mJ1, z = mltp, c”= (E1/13)c.

In the foregoing, o1 is the fundamental natural frequency of a simply supported guideway


span of length 1, and y,, is the maximum deflection of a simply supported guideway span
when a unit load is applied at the midpoint of the span. Here the guideway is assumed to
be uniform and smooth. Further, we shall reduce the system to study the case where the
vehicle is traveling at a constant speed and the length of the vehicle is small compared to
the span length. With these assumptions, equations (lH5) can be written in non-
dimensional form as
(1-p) d2j3/dz2 + 21, &,(dj,/dz -dj,/dT) + &,,2(j3-j2) = 0, (6)
(1 -p) d2j3/dz2 + p d’j,/dz* + @(j2 -jJ = 0, (7)
jr = j(r,r), (8)

(9)

3. NUMERICAL ALGORITHMS

It obviously is not practical to carry out the integration over an infinite range. The
method of approximation employed here consists of neglecting the inertia force of the
spans remote from the vehicle. This is reasonable since one is basically interested in the
guideway response near the vehicle and, equally important, the interference between any
two spans diminishes as the distance between the two spans increases. In that it causes no
computational inconvenience, the influence function will be used in its exact form.
For the numerical computations each span is divided into N intervals of length rc/N,
and the time is similarly divided. Thus, the integration is reduced to the summation of
values at discrete node points [6]

where the fk’s are the weighting coefficients whose values depend on the choice of method
for numerical integration. For the numerical differentiation of the time variable we chose
a backward difference scheme. Specifically,

(11)
248 J. GENIN AND Y. I. CHUNG

The application of the numerical schemes (10) and (11) to equations (6H9) yields

+ h,,"{J3(Tj)-~2(zj)) = O, (12)

(13)

Equations (12), (13) and (14) may be transformed into recursion relations by solving for
jj(tj, zj), jj2(rj) and jj3 (rj) in terms of other variables. Thus one obtains

(16)

+ n42i;Iv"2c"ij
(’
x/2b.i(1-P)P,+3,j+~~~+2,j}

In the foregoing the 5;s and zis indicate the jth nodal points of the position and time
co-ordinates, respectively. Also Jij, JjN+ 2,j ) yN + 3,j and Zij have been used as abbreviations
for jj(ti, rj), j2(rj), y3(zj) and E(ci, <j). s,(Zj) has been expressed as Jjj.
RJZSF’ONSEOF A VEHICLE-TRAVERSED GUIDEWAY 249

For convenience in the computer programming, the recursion relations (1% (16) and
(17) may be represented by the single matrix equation
Q

Aj~ = Gj + C B,~_,,
where

Yl,j

A,, = Z +
rj=

n’B’{(;~b,,
+%(;)a+,
I-! _ijN+
1,j

9N + 2,j

jN+3,j

A,, =.41&2(~~b,Q,
... ...

A
12
=
0
[ 0
... s”i,” . . . 0
0 0 1’

A 22 =

I
,u

0
; ‘b,, + fi,z (l-14
0 ‘b,
;

B,, = -n3fi2{(;)‘br +$f$q}I’, B,, = -x4ii&Y2(;)ibrQ,

I.
-(l-d(Eyb.
+X,% (!)a,
(1-p) ; ‘b,
0

In the foregoing Z is the identity matrix and the other matrices are defined as follows:

1
h,fl .* PLc”lj C1-P121j
p= ; *. ; i .

[- cN+l,l f1 '. PEN + 1.j pEN+I,j


250 J. GENIN AND Y. I. CHUNG

4. ANALYSIS AND RESULTS

In the computations, the trapezoidal rule and a second order backward difference
scheme were used for the numerical integration and differentiation, respectively. The
number of increments in each span determines the amount of computation time required
as well as the accuracy of the solution. Since the computing time increases roughly by the
order of N3.5 where N is the number of increments selected, the choice of an acceptable
number of increments is important for an effective analysis. In our study each span was
divided into seven segments and the results displayed reasonable accuracy when com-
pared to those obtained by using a liner division.

I I I 1 I
0 0.5 I.0 1.5

Vehicle speed, i;
Figure 3. Guideway maximum deflection versus vehicle speed for varying mass ratios.

Figure 3 shows a graph of guideway maximum deflection with respect to the vehicle
speed for four values of mass ratio. A mass ratio of zero amounts to the pure force
approximation. Note that an increase in vehicle mass tends to lower the critical speed.

5. CONCLUDING REMARKS

It is of interest to compare the behavior of the guideway structure analyzed here to that
of a continuous guideway resting on an elastic foundation.
For the guideway analyzed here the speed at which a transverse wave can propagate along
the guideway can be obtained as follows. If a wave form y = sin a(x - ut) is substituted into
the Bernoulli-Euler beam equation
El a4y/ih4 + m a2ypt2 = 0

one obtains EIa4-ma2v2 = 0, or v = am. Then, noting that a = n/l, 241, . . . , one
arrives at D,,,~”
= (n/1) m. In terms of a non-dimensional speed, U”,in= 1.0.
In a subsequent study we showed that this critical speed of the traversing vehicle is
largely dictated by the vehicle’s speed and the natural frequency of the guideway. The
vehicle configuration has little effect on the onset of the critical condition. Hence, the
design of a high speed continuous guideway may require that special attention be given to
this phenomenon.
RESPONSE OF A VEHICLE-TRAVERSEDGUIDEWAY 251

0.001 / / I 1
0.01 0.1 I.0

Vehicle speed, 7

Figure 4. Passenger car heave acceleration versus vehicle speed for (a) a continuous guideway and (b) a simply
supported guideway. fi = 0.5, Jo= 0.1, [ = 0.3, R = 0.6, r = 0.2.

With the foregoing analysis in mind, another important performance criterion to look
at for the vehicle-guideway system is the heave acceleration of the passenger compart-
ment of the vehicle. Figure 4 compares the performance of a continuous guideway model
with that of a multi-span simply supported guideway model, analyzed in reference [l]. As
expected, the continuous guideway has greater rigidity than that of a simply supported
guideway except for the speed range near the critical speed.

REFERENCES

1. Y. I. CHUNG and J. GENIN 1978 American Society of Mechanical Engineers Journal of Dynamics
Systems, Measurement and Control 100, 326-332. Stability of a vehicle on a multispan simply
supported guideway.
2. J. T. KENNEY 1954 Journal of Applied Mechanics, Transactions of the American Society of
Mechanical Engineers 76, 359-364. Steady state vibrations of beam on elastic foundation for
moving load.
D. J. MEAD1970 Journalof Soundand Vibration 11,181-197. Free wave propagation in periodically
supported, infinite beams.
J. GENIN, J. H. GINSBERGand E. C. TING 1975 Journalof Soundand Vibration 38,15-26. A complete
formulation of inertial effects in the guideway-vehicle interaction problem.
E. C. TING, J. GENIN and J. H. GINSBERG1974 Journal of Sound and Vibration 33,49-58. A general
algorithm for moving mass problems.
M. G. SALVADORIand M. L. BARON1961 Numerical Method in Engineering. Englewood Cliffs, New
Jersey: Prentice Hall.

You might also like