I.J.

Image, Graphics and Signal Processing, 2011, 1, 1-9

Published Online February 2011 in MECS (http://www.mecs-press.org/)

Inversion and Feedback Research on the

Temperature Control and Crack Prevention for
Concrete Crane Beam on Rock Wall
Yang Zhang
College of Hydraulic Science and Engineering, Yangzhou Unversity, Yangzhou 225009, China
Email: wwwzhangy@163.com

Sheng Qiang
College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China
Email: sqiang2118@163.com

Abstract—Concrete crane beam on a rock wall on a new inaccurate parameter or functions may cause wrong
structure used in underground building has become more results in numerical calculation for concrete temperature
common in recent year. But the concrete beam cracking field which may induce improper construction measures
problem always perplexes scientists and engineers. In order and dangerous crack.
to solve this, the construction information inversion and
feedback analysis method is applied. A beam section was In the past years, temperature control for mass
taken as a prototype experiment. The temperature and concrete structure such as concrete dam is emphasized
construction data was collected to inverse some necessary enough [2]. In fact the common concrete structure also has
thermal parameters. According to the characteristics of the temperature stress problem which can be found from
concrete temperature field, the basic accelerating genetic the high frequent appearing of concrete cracking.
algorithm was improved. The improved accelerating genetic
algorithm has the merits of high precision and fast Concrete crane beam on rock wall is used widely in
calculation. With this algorithm, the calculation underground structure in recent years. But the crack in
temperature and measured value are very close, which concrete beam is always a problem for designer and
shows the method is efficiency. Then inversed parameters constructor. Decreasing the length of beam is an efficient
were applied in the feedback simulation. According to the way to decrease the probability of cracking. But in some
simulation results, the proper temperature control method engineering, the cracks still appear even the beam length
was suggested. By this way, the concrete temperature was is very short. In fact, thermal stress is a very important
controlled well and the beams appear no crack in recent two potential reason to the crack [3]. To research such
year. The successful application shows that the inversion problem, the factual thermal parameters are important. In
and feedback analysis of concrete temperature field can order to reflect the true temperature field and stress field
reflect the factual performance of concrete and give of concrete crane beam, it is very necessary to get the
important direction to engineering construction. accurate thermal parameters by site measuring
temperature and back analysis [4]-[5]. In the research, the
Index Terms—inversion, genetic algorithm, feedback improved accelerating genetic algorithm (GA) is applied
analysis, crane beam on rock wall, temperature field in the back analysis.

I. INTRODUCTION II. IMPROVEMENT OF ACCELERATING GENETIC

ALGORITHM
The numerical simulation of temperature field and
stress field of concrete structure is influenced by many Comparing with the basic GA, the evolution iteration
factors, which include the material heat parameters at velocity and efficiency of the accelerating GA has been
construction site. The temperature conductivity, thermal increased. But the low capability of local search and
conductivity, surface heat transfer coefficient and premature convergence problem of the algorithm has not
adiabatic temperature of different concrete are not same. been solved radically. Besides, the problem of depositing
Even the same concrete in different circumstance excellent solutions has not solved by whether the basic
including different air temperature, humidity, wind etc, GA or accelerating GA. Therefore the improved
will make some thermal parameters various. And some accelerating GA is proposed [6]-[7]. The core of the
parameters are various with time, especially in concrete improved algorithm is at two points. The first point is to
early age [1], which should be described by functions. An classify chromosomes according to sufficiency. The
chromosome is divided into best chromosome, middle
Manuscript received March 5, 2011; revised March 26, 2011; chromosome and worst chromosome in the proportion of
accepted April 1, 2011. x1, x2 and x3. Here there is x1+x2+x3=1. Usually x1 is
Corresponding author: Sheng Qiang less than 5%, x2 is less than 85%, x3 is less than 10%.
Their value is relation to the evolutionary generation l.

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
2 Inversion and Feedback Research on the Temperature Control and Crack Prevention for
Concrete Crane Beam on Rock Wall

The best chromosome is copied directly, the middle the spatial domain discrete and the time domain
chromosome participates the cross operation, the worst differential for (3). Then the finite element recurrence
chromosome participates the mutation operation. Then the equation of backward difference for temperature field can
quasi progeny population is generated. The point is to be obtained:
solve the problem of depositing excellent solutions and
improve the local search capability. The second point is to  1  1
introducing the niche elimination operation. It is to unite [H ] + [ R ]  {Tn +1} − [ R ]{Tn } + {Fn +1} = 0 (4)
the quasi progeny population and the first NR individuals  ∆ tn  ∆tn
which are remembered before the classifying operation, in which, [ H ] is heat conductivity matrix; [ R] is
then the paired Hamming distance of the new population heat conductivity supplementary matrix; {Tn } and {Tn +1}
are compared. Let NT=NR+n. Define the Hamming
distance as: are node temperature array; {Fn +1} is node temperature
m
load array; n is time step ordinal number; ∆tn is time
sij = Vi − V j = ∑ (v
k =1
ik − v jk ) 2 step length. According to recurrence (4), the node
temperature {Tn +1} of next time step can be derived from
i = 1, 2,..., NT − 1; j = i + 1,..., NT (1)
Suppose S is the control threshold value. If Sij<S, the known current node temperature {Tn } .
compare the value of individual sufficiency between To control the temperature in concrete, pipe cooling
{Vi,Vj}. The individual of less sufficiency is given a are widely used, such as in the structure of nuclear power
greater penalty function to decrease its sufficiency plants [9]-[10], and in civil building concrete structure for
furthest. Thus the elimination rate of the punished air-conditioning or other purposes [11]-[13]. But the
individual is great in the evolution course. Then the unsteady temperature field including cooling pipe is
population variety is kept and the premature convergence always a hard problem to simulate.
is removed.
Three-dimensional (3D) finite element method is a
Otherwise, the common judge condition of the powerful measure to solve this problem, and has been
population convergence is improved. Suppose the object studied for a long time. Early studies, such as pseudo 3D
function value of the first NS individuals (usually and 3D finite element methods, were developed and used
NS=(5~10％)⋅n) after the operation of the lth and (l+1)th for a long time[14]-[15]. However, the pseudo 3D method
generation and sorted from superiority to inferiority are was converted from a simple two-dimensional (2D)
separately f1l , f2l ,..., f NS
l
and f1l +1 , f 2l +1 ,..., f NS
l +!
. Record analysis method and cannot correctly apply boundary
conditions for the atmosphere and cooling water
~ ~
EPS = n1 f1 + n2 f 2 (2)
convection.
Then, some calculation methods based on heat transfer
In which: by heat balance principle is developed, such as Jin Keun
(NS )
NS
~ Kim method [16]. The water temperature distribution
f 1 = NS ⋅ f 1 l + 1 − ∑j =1
f j
l +1
⋅ f 1l +1
along the pipe can be calculated using these methods, and
the boundary conditions of the atmosphere can also be
( )
NS
~
f 2 = ∑ f jl +1 − f jl f jl +1 applied. However, the temperature gradient properties are
j =1 simplified. Also based on the heat balance principle, an
n1— the control coefficient of premature convergence explicit iterative algorithm was developed solve it [17]-
index for the same generation population; [18], which will be applied in the inversion and simulation
n2— the control coefficient of evolution convergence for the crane beam.
for the different generation population. A typical concrete block section containing a cooling
pipe, as shown in figure 1, can be described by some cube
III. THE PRINCIPLE OF FINITE ELEMENT METHOD FOR finite elements around the pipe. Every surrounding
THREE DIMENSIONAL UNSTEADY TEMPERATURE FIELD element has a free face towards the pipe, as shown in
figure 2, which is the third boundary condition in thermal
According to the heat balance principle, at the arbitrary calculation to simulate the heat exchange between
point in the calculation area R, an unsteady temperature cooling water and concrete.
field T ( x，y ，z，t ) must satisfy the following heat
transfer equation:
∂T  ∂ 2T ∂ 2T ∂ 2T  ∂θ
= a 2 + 2 + 2  + (3)
∂t  ∂x ∂y ∂z  ∂τ
In which, T is temperature (℃); a is temperature
conductivity coefficient (m2/h); θ is adiabatic
temperature of concrete (℃); t is time (d); τ is concrete
age (d). Figure 1. Sketch of pipe cooling in concrete

When finite element method is used to solve a

temperature field, variation principle is applied to make

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
 Inversion and Feedback Research on the Temperature Control and Crack Prevention for 3
Concrete Crane Beam on Rock Wall

IV. ENGINEERING APPLICATION

A. Computation Parameter and Condition

A crane beam on rock wall is usually in a rock cavern.
The rock is the base of the crane beam. The parameter of
the rock is also important in calculation. According to
detecting and test, the thermal parameters of the rock are
gained, seen in table I.
Figure 2. Heat exchange faces
TABLE I
THERMAL PARAMETERS OF SURROUNDING ROCK
By studying the heat exchange between water and
concrete, the heat transfer through the pipe surface within Heat
Specific heat
Linear Temperature
time dt can be expressed as: Rock type conductivity Coefficient conductivity
λ
c(kJ/(kg.℃))
(kJ /( m ⋅ h⋅ C ))
0 －5
(10 /℃) a(m2/h)
∂T
dQc = ∫∫ qi ds dt = −λ ∫∫ ds dt (5) Basalt 7.8372 0.886 0.68 0.0032165
Γ0 Γ0
∂n The merit of a crane beam on rock wall is saving the
where λ is the thermal conductivity of the concrete, Γ0 concrete pillar. But at the same time, extra long bolt must
is the interface between the concrete and pipe, and n is be applied in the beam and rock wall. The length of bolt
the interface normal vector. is about 8~12m, the horizontal space between two bolts is
The heat absorbed by water flowing through the about 0.7m. Because the number of bots and its function
section of pipe within time dt is on load which influences the stress in beam evidently, the
bolts should be considered in numerical calculation. Bolt
dQ1 = cw ρ w ∆Tw qw dt (6) is man-made homogeneous material. It is simple to get its
where cw, ρw and qw stand for the specific heat, density thermal parameters, seen in table II.
of water and flux, respectively, and ΔTw is the water
temperature difference between the inlet and outlet of the TABLE II
THERMAL PARAMETERS OF BOLT
pipe section.
Given that the volume of the pipe is very small, and Heat
Specific heat
Linear Temperature
the water body heat increment in the pipe section can be Bolt Type conductivity Coefficient conductivity
λ
c(kJ/(kg.℃))
neglected. From (5), (6), the water temperature increment (kJ /( m ⋅ h⋅ C ))
0 －5
(10 /℃) a(m2/h)
within the pipe section can be found: 40Si2MnV 163.29 0.46 1.65 0.04522

−λ ∂T Concrete thermal parameters are the most important

∆Twi =
cw ρ w qw ∫∫0 ∂n ds (7) parameters in the temperature field simulation. Because
part of the thermal parameters that will not change in
Γ
Because the inlet water temperature is known, the construction and running, can be gained by experiment in
water temperature at any pipe section along the pipe can lab [8], such as the parameters shown in table III. Some
be calculated step by step. If there is a water pipe divided other parameters are necessary to be inversed, such as
into several pipe sections and the water temperature boundary conditions.
increment of one pipe section is ΔTwj and the inlet water
temperature is Tw0, then, TABLE III
PARTS OF THERMAL PARAMETERS OF CONCRETE
i
Twi = Tw 0 + ∑ ∆Tw j
Heat Linear Temperature
(8) conductivity Specific heat
Coefficient conductivity
c(kJ/(kg.℃))
j =1 λ (kJ /( m ⋅ h⋅ 0 C ))
－
(10 5/℃) a(m2/h)
Because the unknown temperature gradient on the 9.6868 0.874 0.85 0.00434
interface between the concrete and pipe affects the water
temperature distribution along the pipe, it is a non- B. The Prototype Experiment
linearity problem of the boundary conditions and cannot The first built crane beam is taken as the prototype
be solved directly. Therefore, an iterative method should experiment. Some temperature sensors are set in it before
be used. concrete pouring. Once the beam concrete is poured,
For the first iteration, the water temperature along temperature measuring work begins. The external factors,
water pipe is assumed equal to the inlet water temperature. such as pouring temperature, pipe cooling course, air
The approximate solution of the temperature field can be temperature, surface covering condition, etc, should be
calculated with (4). Then, the water temperature also recorded during all the measuring work. Figure 3
distribution along the water pipe can be calculated with (7) shows the excavated rock wall before the crane beam is
and (8). The process is then repeated until the water poured. Figure 4 shows the crane beam in construction.
temperature distribution along the water pipe and the The beam segment in the picture left half has been poured
concrete temperature field reaches a steady-state solution. and is protected by form board. The picture right half in
figure 4 shows other beam segments which are just steel
reinforcement cages. The white line is the cooling water

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
4 Inversion and Feedback Research on the Temperature Control and Crack Prevention for
Concrete Crane Beam on Rock Wall

transit pipe from a pool. Figure 5 shows the form board disposed in the mesh. The pipe cooling course will be
which is named VISA mould. The mould will keep the calculated with the explicit pipe model. The bolt layout
surface of concrete smooth and compact. It can also and the pipe layout with their grid are shown in figure 7
provide some insulation effect, so it is an important and figure 8 separately. The calculation time step length
boundary condition in the numerical simulation, which is same as the interval of measuring. The air temperature,
should be inversed correctly. pipe inlet water temperature, water flux are same as the
measured value.
The sensors are set at the center section location. The
location of five sensors is shown in figure 9. The inner
sensors are N1, N2 and N3. Sensor D4 and C5 are used to
measure the temperature near concrete surface. D4 is 5cm
to the beam top and C5 is 5cm to the beam side.

Figure 3. Rock wall before beam built

Figure 6. Finite element mesh for inversion

Figure 4. Crane beam on rock wall in construction

Figure 7. Bolt layout in beam

Figure 5. VISA mould of the crane beam

There are always some errors in the recorded data

because of human or machine. All the data should be
analyzed and the unreasonable data must be eliminated
before they are used in the inversion.
C. The Mesh for Inversion and the Sensor Location
Figure 6 shows the whole finite element mesh for
inversion, which includes part of surrounding rock, one
segment of crane beam, bolts in rock wall and beam. The
whole mesh includes 35784 nodes and 32411 elements.
To simulate the factual situation, cooling pipe is also Figure 8. Pipe layout and its gird in beam

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
 Inversion and Feedback Research on the Temperature Control and Crack Prevention for 5
Concrete Crane Beam on Rock Wall

(2) The beginning time of pipe cooling is late in the

prototype experiment. The cooling start at the 14th hour
after the concrete poured. Seen from figure 11, the
maximum water temperature difference between pipe inlet
and outlet is 12.80℃. It is a too high value. Because in the
first 14 hours, concrete hydration reaction is taking place
rapidly while no pipe cooling is working. When the pipe
starts cooling, much heat quantity has gathered inside the
concrete beam. After the temperature peak in concrete,
the water temperature difference between pipe inlet and
outlet decreases and becomes stable. The inversion
results also show such rule.

I
nlet O
utle
t
35

tu /℃
re
30
Figure 9. Sensor location on beam section

mpera
25

rte
D. Experiment and Inversion Results Analysis
20

te
The measured data is input as known condition in the
Wa 15
inversion program. After many times optimization, the
0 2 4 6 8
best parameters are gained as follows.
T
ime/d
Concrete adiabatic temperature: θ0 ＝40.00 ℃
Concrete adiabatic temperature curve:
Figure 11. Water temperature curves of cooling pipe inlet and outlet
θ = 40.00 × (1 − e−0.60τ ) ℃
1.02

Heat transfer coefficient of form board： (3) Figure 12~16 show the measured and calculated
temperature curves of different sensors. The calculated
β1 ＝9.18 kJ/(m2·h·℃) data is gained on the base of inversed parameter. Most of
Heat transfer coefficient of wet felt： the data fits well which reflects the inversed parameters
β 2 ＝22.61 kJ/(m2·h·℃) are reliable and can be applied in the future concrete
temperature forecast and control of the crane beam.
Heat transfer coefficient of pipe wall：
β3 ＝45.42 kJ/(m2·h·℃)
Ca
lcu
lati
on Me
asu
ring
Heat transfer coefficient of uncovered concrete： 45
β 4 ＝32.37 kJ/(m2·h·℃) 40
re/℃

Based on the site application of the temperature 35

ratu

control method, comparing the inversed results with 30

e

measured data, the following recognition can be gained.

mp

25
Te

(1) From the air temperature curve in cavern (figure 20

10), the maximum air temperature is 22.69℃, minimum 15

0 1 2 3 4 5 6 7 8
air temperature is 19.06℃ during the first 8 days age. Con
cret
eage
/d
And the maximum air temperature difference in one day
is only 2.50℃. It explains why there is no greater wave
on the temperature curves of surface sensor D4 and C5. Figure 12. Measured and calculated temperature curves of sensor N1
Another way, the day and night temperature difference in
cavern is evidently less than outside.
Ca
lcul
atio
n Me
asur
ing
45
23
℃

40
℃
re/

e/

22
35
tur
atu

21
ra

30
mer

pe

20 25
Airte

em
T

19 20
15
18
0 1 2 3 4 5 6 7 8
0 2 4 6 8
Con
cret
eag
e/d
Ti
me/d

Figure 10. Air temperature curve in cavern Figure 13. Measured and calculated temperature curves of sensor N2

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
6 Inversion and Feedback Research on the Temperature Control and Crack Prevention for
Concrete Crane Beam on Rock Wall

C
alcu
lati
on Me
asur
ing 20.0℃. The water cooling begins after the pouring with
the water temperature of 13.0℃ and the flux of 3.00m3/h.
5
0 The cooling keeps 7 days and the water direction is
4
5 changed every 12 hours. The mould and felt is removed
/℃

at the age of 14th day.

re

4
0
atu

3
5 As for case 2, the mould and felt is removed at the age
mper

3
0 of 7th day. The other calculation situation is same as case
Te

2
5 1.
2
0 As for case 3, the mould and felt is removed at the age
0 2 4 6 8 of 30th day. The other calculation situation is same as case
Co
ncre
teage
/d 1.
From the calculation result of case 1, the top
Figure 14. Measured and calculated temperature curves of sensor N3 temperature of point N1, N2 and N3 is 33.47℃, 35.09℃,
36.02℃, and the occur time is the 1.75th day, the 2.00th
day, the 2.00th day separately. The temperature curve of
Ca
lcul
atio
n Me
asur
ing
40
N3 is shown in figure 17. The internal max tension stress
never exceeds the allowable tensile strength. For example,
℃

35
e/

the max tension stress of point N3 is 1.16MPa occurred at

tur

30 the 60th day, which is not more than the allowable tensile
ra

strength of 1.25MPa, as shown in figure 18. The line in

pe

25
figure 18 with the sign of 2.0 σa is the allowable tensile
em
T

20
strength curve with a safety factor of 2.0. In the next
15 stress curve figures, the sign is the same meaning. From
0 1 2 3 4 5 6 7 8 figure 18, it is obviously that the surface stress always
Con
cret
eag
e/d keeps lower than the allowable tensile strength.

Figure 15. Measured and calculated temperature curves of sensor D4

Ca
lcul
atio
n Me
asur
ing
45
40
e/℃

35
ratur

30
pe

25
em
T

20 Figure 17. Temperature curve of inside point N3 in case 1

15
0 1 2 3 4 5 6 7 8
Con
cret
eag
e/d

Figure 16. Measured and calculated temperature curves of sensor C5

(4) Under the situation of lack of test data, some

thermal parameters can be gained by inversion calculation.
It shows adequately the importance and necessity of
concrete thermal parameter inversion.
E. Feedback Analysis of the Crane Beam Figure 18. First principle stress σ1 curve of inside point N3 in case 1
The feedback calculation is applied for the beam
constructed later on the base of inversion. According to Figure 19 shows the temperature comparison curves
the site situation in the first several months, it was found of top face point D4 calculated in case 2 and case 3.
that the factual pouring temperature and cooling water Figure 20 shows the first principle stress comparison
temperature is relative steady. So their sensitive analysis is curves of top face point D4 calculated in case 2 and case
not necessary to carry out. In the following simulation 3, including the allowable stress curve. Figure 21 shows
cases, the duration time of surface preservation is the temperature comparison curves of side face point C5
researched. calculated in case 2 and case 3. Figure 22 shows the first
principle stress comparison curves of side face point C5
At first the calculation conditions are presented. calculated in case 2 and case 3, including the allowable
As for case 1, the thermal parameters of concrete are stress curve.
same as the inversion calculation. The wet felt on the top
of beam is also simulated. The pouring temperature is

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
 Inversion and Feedback Research on the Temperature Control and Crack Prevention for 7
Concrete Crane Beam on Rock Wall

value, the stress goes down because the temperature

difference between internal and surface becomes less. In
the later age, the surface tension stress keeps lower than
the allowable tensile strength.
From the calculation result of case 3, the temperature
decreasing amplitude is less when removal occurred at the
30th day, which is only 0.5℃ in 1.0d. Wherever the top
face or the side face, the surface tension stress can always
satisfy the allowable tensile strength.
Figure 19. Temperature curves of D4 in case 2 and case 3 Because the heat exchange coefficient of side face is
lower than the top face, the removal of covering materials
makes more influence to the side face temperature and
stress, especially removing at 7th day. The temperature
decreasing amplitude of side face point C5 drops is 3.60℃
in 0.5d after the removal and the corresponding stress
increases from 0.65MPa to 1.12MPa, which exceeds the
current allowable tensile strength. In the later 2 days, the
stress keeps raising to 1.20MPa, which is still more than
the allowable tensile strength. While the temperature and
stress does not change suddenly if the removal occurs at
30th day, and the stress does not exceed the allowable
tensile strength in the whole construction time.
Figure 20. First principle stress σ1 curves of D4 in case 2 and case 3
The above analysis shows that the preservation
material and covering time have much influence on the
concrete surface temperature and stress. As for the
preservation material with lower heat exchange coefficient,
the removal time should be later.
At the same time, the extent and time of surface
preservation also depend on the extent and time of water
cooling. If the extent of water cooling at early age is more,
the surface covering removal can be earlier. Judging from
the site situation, the pipe input water temperature is not
controlled strictly, and parts of the pipe linker have water
Figure 21. Temperature curves of C5 in case 2 and case 3 leakage which leads to the flux unsteady. Under such
situation, the surface covering time can not be shortened.
It should be more than 14.0d.

V. CONCLUSIONS
The prototype experiment of concrete crane beam on
rock wall is carried out to get the temperature of different
points inside the beam. Then the improved accelerating
genetic algorithm and temperature field finite element
program are applied to inverse some important thermal
parameters of concrete. Comparing the measured data
with calculated data, they are very close. It shows the
Figure 22. First principle stress σ1 curves of C5 in case 2 and case 3
inversion method is effective. The inversed parameters are
applied in the feedback simulation. According to the
From the calculation result of case 2, the max internal
simulation results, the proper temperature control method
temperature and stress do not change obviously
is suggested. On the base of above research working, the
contrasting with case 1. But the surface temperature and
following conclusions are drawn.
stress change much. When the covering materials are
removed at the 7th day, surface temperature decreases (1) The prototype experiment for concrete temperature
suddenly. It changes from 22.17℃ to 18.93℃ in only 0.5d. is simple and feasible, which can be easily operated. The
After the removal, surface temperature waves apparently experiment results can satisfy the feedback analysis
with the day and night air temperature difference. The calculation and engineering construction.
surface tension stress increases suddenly form 0.63MPa at
the 7th day to 1.06MPa at the 7.5th day. The value is near (2) The back analysis and simulation in construction is
the current allowable tensile strength 1.09MPa. Then the a gradually optimizing course. It can reflect the factual
stress keeps rising as the temperature decreasing until it performance of concrete and give important direction to
reaches 1.15MPa at the 8.5th day which exceeds the engineering construction.
current allowable tensile strength 1.12MPa. After the top

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
8 Inversion and Feedback Research on the Temperature Control and Crack Prevention for
Concrete Crane Beam on Rock Wall

(3) In the inversion of concrete thermal parameters Ignalina Nuclear Power Plant”, Nuclear Engineering and
course, the improved accelerating genetic algorithm has Design. Vol. 237, pp. 848–857, 2007.
the merits of high precision and fast calculation. It [10] Silin N., Masson V.P. and et al. “Hydrodynamic transition
overcomes the shortage of traditional optimization method, delay in rectangular channels under high heat flux”, Annals
most of which show a series deceleration when the of Nuclear Energy. Vol. 37, pp. 615–620. 2010.
inversion parameters become more, easily fall into the [11] Koschenz M. and Dorer V. “Interaction of an air system
local extreme value, be hard to converge because of error with concrete core conditioning” Energy and Buildings.
transfer. It is worthy to be applied widely in civil Vol. 30, pp. 139-145, 1999.
engineering. [12] Tan B.K., Wong T.N. and Ooi K.T. “A study of liquid flow
(4) The feedback analysis shows that the preservation in a flat plate heat pipe under localized heating”,
material and covering time have much influence on the International Journal of Thermal Sciences. Vol. 49, pp. 99-
108, 2010.
concrete surface temperature and stress. As for the
preservation material with lower heat exchange coefficient, [13] He J. and Hoyano A. “Experimental study of cooling
the removal time should be later. effects of a passive evaporative cooling wall constructed of
porous ceramics with high water soaking-up ability”,
(5) The extent and time of surface preservation is Building and Environment. Vol. 45, pp. 461–472, 2010.
related to the extent and time of water cooling. If the [14] Kawaraba H., Kanokogi T., Tanabe T. “Development of
extent of water cooling at early age is more, the surface the FEM program for the analysis of pipe cooling effects
covering removal can be earlier. on the thermal stress of massive concrete”, Transactions of
the Japan Concrete Institute. Vol. 8, pp. 125-130, 1986.
[15] Machida N, Uehara K. “Nonlinear thermal stress analysis
ACKNOWLEDGMENT
of a massive concrete structure”, Computers & Structures.
The authors wish to thank S. K. Chen, L. Guo, Z. Q. Vol. 26, pp. 287-296, 1987.
Xie, M. D. Zhu. This work was supported in part by a [16] Kim J.K., Kim K.H. and Yang J.H. “Thermal analysis of
grant from Wuhan University Open Research Fund hydration heat in concrete structures with pipe-cooling
Program of State Key Laboratory of Water Resources and system”, Computers & Structures. Vol. 79, pp. 163-167,
Hydropower Engineering Science (No. 2009 B060). 2001.
[17] Y M Zhu, J R He, Z Q Xiao. “Calculation and
REFERENCES experimental comparison of temperature field of embedded
cooling water pipes and their application”. Journal of
[1] J H Hattel, J A Thorborg. “A numerical model for Hohai University, vol. 31, pp. 626-630, June 2003.
predicting the thermo mechanical conditions during [18] Y M Zhu, Z Q Xu, J R He. “A calculation method for
hydration of early-age concrete”. Applied Mathematical solving temperature field of mass concrete with cooling
Modeling, vol. 27, pp. 1-26, 2003. pipes”. Journal of Yangtze River Scientific Research
[2] S Qiang, Y M Zhu, Y L Zhao, H L Rao. “Study on Institute. Vol. 20, pp. 19-22, February 2003.
concrete temperature control and crack prevention
measures of Jinping I super high arch dam”. Proceedings
of the 1st International Conference on Long Time Effects
and Seepage Behavior of Dams. Vol. 1, pp. 564-568, May Yang Zhang was born in Hubei Province, Chinese on
2008. January 15, 1979. Zhang earned the bachelor’s degree of civil
engineering from Wuhan University in Hubei Province, Chinese
[3] B F Zhu. Thermal Stresses and Temperature Control of in 2000, and the master’s degree of hydraulic structure from
Mass Concrete. Second ed. Beijing:China Electric Power Wuhan University in 2003, and the doctor’s degree of hydraulic
Press, 1999, pp.8-17. structure from Hohai University in Jiangsu Province in 2009.
[4] G J Stolz. “Numerical solution to an ivnerse problem of Zhang’s major field of study is rock stabilization analysis and
heat conduction for simple shampes”. J. Heat Transfer, vol. mass concrete crack prevention.
82(C), pp. 20-26, 1960. She became a lecturer at Yangzhou University in Jiangsu
[5] Y F Ma, Y M Zhu, Y Z Liu. “Feedback study of Province from 2009. Her main works include:
temperature control and crack prevention of Jiangtanghu “Research on 3D p-version hierarchical FEM for unsteady
pump concrete sluice during construction”. Water Power, temperature field”. Rock and Soil Mechanics, Chinese. Vol. 30,
vol. 32, pp. 33-35, January 2006. pp. 487-491, February 2009.
[6] Y M Zhu, Y J Liu. “Determination of thermal parameters “Analysis of p-version adaptive finite element method for
of concrete by reverse analysis of test results”. Chinese three dimensional unsteady temperature field”. Journal of
Journal of Geotechnical Engineering, vol. 24, pp. 175-177, Yangtze River Scientific Research Institute, Chinese. Vol. 26, pp.
February 2002. 51-55, June 2009.
“Study on static analysis of 3D high-order DDA”. Chinese
[7] M Zhou, S D Sun. Principle and Application of Genetic Journal of Rock Mechanics and Engineering, Chinese. Vol. 29,
Algorithm. Beijing: National Defence Industry Press, 2000, pp. 558-563, March 2010.
pp.4-7. The current and previous research interests are mass concrete
[8] R Kohn, M Vogelins. “Determining conductivity by crack prevention and numerical algorithm.
boundary measurements”. Common Pure Appl. Math, vol. Dr. Zhang is not a member of any professional society.
137, pp. 289-298, March 1984.
[9] Dundulis G., Uspuras E. and et al. “Evaluation of pipe
whip impacts on neighboring piping and walls of the Sheng Qiang was born in Jiangsu Province, Chinese on July
20, 1977. Qiang earned the bachelor’s degree of hydraulic

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9
 Inversion and Feedback Research on the Temperature Control and Crack Prevention for 9
Concrete Crane Beam on Rock Wall

structure from Wuhan University in Hubei Province, Chinese in

1999, and the master’s degree of hydraulic structure from
Wuhan University in 2002, and the doctor’s degree of hydraulic
structure from Wuhan University in 2005. Qiang’s major field
of study is rock stabilization analysis and numerical algorithm.
He attended Hohai University in Jiangsu Province as a
lecturer from 2005. And he became an associate professor at
2009. Her main works include:
“Three dimensional elasto-viscoplastic composite element
model for discontinuous rock masses”. Chinese Journal of Rock
Mechanics and Engineering, Chinese. Vol. 23, pp. 3390-3396,
October 2004.
“Preliminary research on composite element model of
temperature field for discontinuous rock masses”. Chinese
Journal of Rock Mechanics and Engineering, Chinese. Vol. 27,
pp. 2094-2100, October 2008.
“Application of three-dimensional hierarchical FEM to
calculation of temperature field in concrete inverted siphon
engineering”. Journal of hydraulic engineering, Chinese. Vol.
40, pp. 94-100, January 2009.
The current and previous research interests are mass concrete
crack prevention and numerical algorithm.

Dr. Qiang is not a member of any professional society. His

doctor dissertation was awarded the Excellent Doctor
Dissertation of Hubei Province in 2006. “Research on the
pumping concrete temperature control and crack prevention
method for large scale channel in pump station and application”
was awarded Dayu Hydraulic Science and Technology Prize in
2008. “Research on the crack reason and prevention method for
sluice concrete and application” was awarded the Science and
Technology Prize of Huaihe Water Resource Commission in
2008.

Copyright © 2011 MECS I.J. Image, Graphics and Signal Processing, 2011, 1, 1-9