Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 6

Chapter 4

METHODOLOGY

4.1 Model Provisions


A satisfactory spring mass analogue to characterize basic dynamics for two mass model
of elevated tank was proposed by Housner (1963) after the chileane earthquake of 1960,
which is more appropriate and is being commonly used in most of the international
codes including GSDMA guideline. The pressure generated within the fluid due to the
dynamic motion of the tank can be separated into impulsive and convective parts. When a
tank containing liquid with a free surface is subjected to horizontal earthquake ground
motion, tank wall and liquid are subjected to horizontal acceleration. The liquid in the
lower region of tank behaves like a mass that is rigidly connected to tank wall, termed as
impulsive liquid mass. Liquid mass in the upper region of tank undergoes sloshing
motion, termed as convective liquid mass. For representing these two masses and in order
to include the effect of their hydrodynamic pressure in analysis, two-mass model is
adopted for elevated tanks.
In spring mass model convective mass (mc) is attached to the tank wall by the spring
having stiffness (Kc), whereas impulsive mass (m i) is rigidly attached to tank wall. For
elevated tanks two-mass model is considered, which consists of two degrees of freedom
system. Spring mass model can also be applied on elevated tanks, but two-mass model
idealization is closer to reality. The two- mass model is shown in Fig. 4.1. where, m i, mc,
Kc, hi, hc, hs, etc. are the parameters of spring mass model and charts as well as empirical
formulae are given for finding their values. The parameters of this model depend on
geometry of the tank and its flexibility. The two-mass model was first proposed by G. M.
Housner (1963) and is being commonly used in most of the international codes. The
response of the two-degree of freedom system can be obtained by elementary structural
dynamics.

29
Fig. 4.1 Two Mass Model for Elevated Tank
However, for most of elevated tanks it is observed that both the time periods are well
separated. Hence, the two-mass idealization can be treated as two uncoupled single
degree of freedom system as shown in Fig. 4.1 (b). The stiffness (Ks) is lateral stiffness of
staging. The mass (ms) is the structural mass and shall comprise of mass of tank container
and one third mass of staging as staging will acts like a lateral spring. Mass of container
comprises of roof slab, container wall, gallery if any, floor slab, floor beams, ring beam,
circular girder, and domes if provided. Staging part of elevated water tanks follows the
Provisions given by Criteria for design of RCC staging for overhead water tanks (First
[18]
revision of IS 11682): Draft Code This draft standard lays down criteria for analysis,
design and construction of reinforced cement concrete staging of framed type with
columns.

4.2 Consideration of GSDMA (Gujarat State Disaster Management Authority)


Guideline
This guideline has been prepared in accordance with generally recognized engineering
principles and practices. Many international codes, standards and guidelines have been
referred. Seismic analysis of water tanks considering hydrodynamic effect has been
explained in detail. Following important provisions and changes have been incorporated
as compared to that of IS 1893:1984.
i. Analysis of ground supported tanks.
ii. For elevated tanks, the single degree of freedom is replaced by two degree of
freedom idealization.

30
iii. Bracing beam flexibility is explicitly included in calculation of lateral stiffness of
tank staging.
iv. The effect of convective and impulsive hydrodynamic pressure distribution in the
analysis.
v. Effect of vertical ground acceleration on hydrodynamic pressure and pressure
due to wall inertia.
vi. Sloshing effect of water and maximum sloshing wave height.
vii. P-Delta effect for elevated water tank.

4.3 Fluid-Structure Interaction


As proposed by GSDMA guideline, spring mass analogue of tank-liquid system forms the
basis for evaluating hydrodynamic pressure in most of the design codes across the globe.
During lateral base excitation seismic ground acceleration causes hydrodynamic pressure
on the tank wall which depends on the geometry of tank, height of liquid, properties of
liquid and fluid-tank interaction. Proper estimation of hydrodynamic pressure requires a
rigorous fluid-structure interaction analysis. In the mechanical analogue of tank-liquid
system, the liquid is divided in two parts as, impulsive liquid and convective liquid. Some
researchers have attempted numerical study for tanks of other shapes (Joshi, 2000 and
Damatty, 2000). Similarly, for tanks with internal obstructions, there are some analytical
and numerical studies (Aslam, 1979, Choun, 1996 and Pal, 1998). There are some
experimental studies on the effect of obstructions on the sloshing frequency (Kimura et al.
1995, Armenio 1996, and Reed et al. 1998).Finite element model (FEM) is used to model
the elevated tank system. The analysis of elevated tank under seismic load of Fluid-
Structure-interaction problems can be investigated by using different approaches such as
[22]
added mass (Westergaard, 1931) or velocity potential, Lagrangian (Wilson and
Khalvati, 1983), Eulerian (Zienkiewicz and Bettes, 1978), and Lagrangian Euclidian
approach (Donea, et al 1982). These analyses can be carried out using FEM or by the
analytical methods. The added mass approach as shown in Figure 4.2 can be investigated
by using some of conventional FEM software such as ANSYS V.12, SAP 2000, STAAD
Pro and LUSAS. The general equation of motion for a system subjected to an earthquake
excitation can be written as in standard symbols,

(1)

31
In which M, C and K are mass, damping and stiffness matrices and are the
acceleration, velocity and displacement respectively, and is the ground acceleration. In
the case of added mass approach the form of equation (1) become as below:

(2)
In which M* is the new mass matrix after adding hydrodynamic mass to the structural
mass, while the damping and stiffness matrices are same as in equation (1).

Fig. 4.2 FEM Model for Fluid-Structure-Interaction


Added Mass Approach

Fig. 4.3 (a) Westergaard Added Mass Concept (b)


Normal and cartesian directions Westergaard Model’s method was originally developed
for the dams but it can be applied to other hydraulic structure, under earthquake loads i.e.
tanks. In this paper the impulsive mass has been obtained according to GSDMA guideline
technique and is added to the tanks walls according to Westergaard Approach as shown in
Figure 4.3 using equation 3. Where, is the mass density, h is the depth of water and A i

is the area of curvilinear surface. ….. (3)

32
4.4 Selection of Elements
(24)
ANSYS offers various elements for modeling and analysis, out of which three
elements are used for present FEA model. The tank roof system is represented by shell
and beam elements .Tank wall is modeled by elastic shell elements. The contents are
represented by 3-D fluid elements. The fluid element is particularly well suited for
calculating hydrostatic pressures and fluid/solid interactions.
The basic equation solved in a typical undamped modal analysis is the classical eigen
value problem: [K]{Φi} = ωi2 [M] {Φi}
where:
[K] = stiffness matrix
{Φi} = mode shape vector (eigenvector) of mode i
ωi2 = natural circular frequency of mode i ( ωi2 is the eigen value)
[M] = mass matrix After application of boundary conditions the modal analysis was run
using reduced method. About 241 modes were extracted and significant mode were
chosen for the comparison of participation factors, modal coefficients, and mass
distribution percentages. Larger mass distribution percentages usually indicate important
modes in the corresponding dynamic response analysis. Modal analysis study was carried
out by changing the capacity of the tank, percentage of volume of water and height of the
staging.

4.5 Response Spectrum Analysis


A response spectrum represents is a graph of response versus frequency, where the
response might be displacement, velocity, acceleration, or force. The models which were
used for modal analysis were again run using Single Point Response Spectrum method.
The tank was excitated by response spectrum in X-direction and Z-direction. The real
data Bhuj earthquake 2001( India) is given

33
Fig. 4.4 Response Spectrum of Bhuj Earthquake (2001) in X direction and Z direction[2]

Table No. 4.1 Bhuj Earthquke Disaggregated Seismic Spectra


Frequency (cps) 0.10 0.11 0.13 0.17 0.20 0.25 0.33 0.50 0.67
Acceleration (g) 0.002 0.003 0.003 0.005 0.006 0.006 0.010 0.021 0.031
2
mm/s

Frequency (cps) 1 1.11 1.25 1.43 1.67 2.00 2.50 3.33 4.5
Acceleration (g) 0.047 0.070 0.088 0.105 0.110 0.130 0.150 0.200 0.255
2
mm/s

Frequency (cps) 5.00 6.67 10.00 11.11 12.50 14.29 16.67 20.00 25.00
Acceleration (g) 0.200 0.165 0.153 0.140 0.131 0.121 0.111 0.100 0.100
2
mm/s

34

You might also like