PREDICTION OF SHEAR STRENGTH OF

DEEP BEAM USING GENETIC

PROGRAMMING

Department of Civil Engineering,

National Institute of Technology, Rourkela, India
PREDICTION OF SHEAR STRENGTH OF DEEP BEAM USING

GENETIC PROGRAMMING

A thesis submitted in partial fulfillment of the requirement for the degree of

Bachelor of Technology

In

Civil Engineering

By

Himanshu Kumar Singh

(110ce0059)

Final Year B.Tech student,

Supervised by- Prof. Manoranjan Barik

Department of Civil Engineering,

National Institute of Technology, Rourkela, India

 Department of Civil Engineering,

National Institute of Technology, Rourkela, India

CERTIFICATE

This is to certify that the thesis entitled”PREDICTION OF SHEAR STRENGTH OF DEEP BEAM

USING GENETIC PROGRAMMING” submitted by Himanshu Kumar Singh (110CE0059) of Civil

Engineering Department, National Institute of Technology, Rourkela is an authentic work carried out by

him under my supervision and guidance.

To the best of my knowledge, the matter included in the thesis has not been submitted to any other

institution for any degree or awards.

Date:
Prof. Manoranjan Barik
Place:Rourkela Department of Civil Engineering
National Institute of Technology
Rourkela, Odisha-769008

ii | P a g e
 ACKNOWLEDGEMENT

I would like to express my sincere gratitude to my supervisor Prof. Manoranjan Barik for providing me a

oppurtunity to work under him. His supervision helped me a lot in my research work and thesis writing.

I appreciate his ability and the support which he provided me throughout this academic year. Without

his assistance and commitment, my work might not have come to this stage of fulfilment of my bachelor

degree.

I would like to thank each and every faculty of my department for their knowledge, assistance, motivation

and care throughout the study period.

I would like to thank S Tausif Akram (B.Tech Student) and Gumpalli Sai Prasanth (B.Tech Student) for

their help in clearing my doubts whenever I faced a problem.

iii | P a g e
 ABSTRACT

This research project consists of Genetic Programming (GP) to predict an empirical model for the

convoluted non-straight relation between distinct parameters related with Reinforced Concrete (RC) deep

beam and its ultimate shear capacity. It is a manifestation of artificial intelligence and thoughts, which

is focused around the Darwinian hypothesis of evolution and genetics. The structural and size intricacy

of the empirical model advances as a component of the prediction. Model evaluated by GP is developed

specifically from experimental database accessible from prior literature. The legitimacy of the acquired

model is analyzed by comparing the GP response and the shear capacity ascertained according to

distinctive design codes. The created model produced is utilised for study of relationship between the

shear strength of deep beam and its distinct influencing parameters.

iv | P a g e
TABLE OF CONTENTS

CERTIFICATE ii

ACKNOWLEDGEMENTS iii

ABSTRACT iv

LIST OF FIGURES vii

LIST OF TABLES viii

Chapter Topic Name Page No

Chapter 1 INTRODUCTION 1

1.1 Objective 2

1.2 Introduction 2

1.3 Comparison 3

1.4 Behavior of deep beam 3

1.5 Failure Mechanism 5

Chapter 2 REVIEW OF LITERATURE 6

Chapter 3 METHODOLOGY 9

3.1 Genetic Programming (GP) 10

3.2 Matrix Laboratory (MATLAB) 13

3.3 GPTIPS 13

Chapter 4 MODELLING 14

4.1 Experimental Database 15

4.2 IS Code Computation 18

4.3 ACI Code Computation 18

4.4 Run Parameter 20

v|P a g e
Chapter Topic Name Page No

Chapter 5 RESULT & DISCUSSION 21

5.1 Input frequency analysis 22

5.2 Fitness generation 23

5.3 Error analysis 24

5.4 Scatter plot 25

5.5 Gene weightage 26

5.6 Emperical Relation 27

5.7 Comparison of result (tabular form) 28

5.8 Comparison of result (graphical form) 30

Chapter 6 CALCULATION USING DESIGN CODE 31

6.1 Input variable 32

6.2 Comparison of result (tabular form) 35

6.3 Comparison of result (graphical form) 38

Chapter 7 CONCLUSION 39

Chapter 8 REFERENCES 40

vi | P a g e
LIST OF FIGURES

Figure 1 Basic parameter for deep beam 4

Figure 2 Failure mechanism of deep beam 5

Figure 3 Flow chart for Genetic Programming 12

Figure 4 Input frequency analysis 22

Figure 5 Fitness generation 23

Figure 6 Error analysis 24

Figure 7 Scatter plot 25

Figure 8 Gene Weightage 26

Figure 9 Experimental vs GP Result 30

Figure 10 IS Code vs GP Result 38

Figure 11 ACI Code vs GP Result 38

vii | P a g e
List of tables

Table 1 Experimental result from prior literature 16

Table 2 Comparison of Experimental and GP Results 28

Table 3 Input variable 32

Table 4 Comparison of Results 35

35

viii | P a g e
 CHAPTER 1

INTRODUCTION

1|P a g e
1.1 OBJECTIVE

The objective of the present work is to develop an empirical relation between the shear strength

capacity of deep beam and parameter on which its shear strength depends utilizing the genetic

programming. The data utilized for this are collected from the earlier literature. The results

subsequently acquired are to be contrasted and those obtained using the following codes.

1. IS code.

2. ACI code

1.2 INTRODUCTION

Reinforced Concrete (RC) deep beams are utilized for load distribution within a wide range of

structures; for instance in tall buildings, offshore gravity structures, as pile caps, folded plates,

transfer girder, and foundation wall. shear walls are also considered as cantilever deep beam. Deep

beams are regularly placed on the edge of surrounded structures where they give stiffness against

horizontal loads. The American Concrete Institute (ACI) code 318-95 (condition 10.7.1) (ACI

1995) classify the beam as a deep beam if the clear span/effective-depth ratio is less than 5 for

simply supported beams. According to Indian code (Is456 clause 29), a simply supported beam is

considered as deep beam when the effective span to its overall depth is less than 2. Continous beam

are considered as deep beam when the effective span to its overall depth is less than 2.5. In Deep

Beam a non linear strain profile is noticed even in the elastic range which is not genuine in normal

beam. Strength of deep beam is administered by shear because of non linear strain profile even in

the elastic range. Conduct of shear force in deep beam are analysed by strut and tie model or tied

arch action.

2|P a g e
1.3 COMPARISON

DEEP BEAM SLENDER BEAM

 Plane section before bending does not  Plane section before bending remains
remain plane after bending plane after bending.

 Shear deformations become significant  Shear deformation is neglected.

compared to pure Flexure
 The stress block is non linear even at  The stress block can be considered
elastic stage linear at elastic stage

 It is subjected to two dimensional state  It is subjected to one dimensional

of stress.  state of stress.

 The resulting strain is non linear.  The strain is linear.

1.4 BEHAVIOR OF DEEP BEAM

Deep beams transfer a significant load to supports by forming a compression thrust between

load and the reaction. The diagnal compression combined with the tension bars along the beam

constitute the strut and tie model for deep beam. The force-transferring mechanism can be

recognised by tied arch action of deep beams. Crushing of a compression strut or loss of

anchorage in beam bars are responsible for the failure of deep beam. In general, shear force

governs the design of deep beam, rather than flexural. A large amount of compressive forces

are transferred to supports by “arch action" directly.

Shear strength of deep beam is controlled by the following parameters:-

 The effective span of deep beam

 Width of deep beam

3|P a g e
 Effective depth of deep beam

 Shear span of deep beam

 Characteristics compressive strength of concrete

 Yield strength of longitudinal tensile steel

 Yield strength of vertical web steel

 Yield strength of horizontal web steel

 Longitudinal tensile steel reinforcement ratio

 Horizontal web steel reinforcement ratio

 Vertical web steel reinforcement ratio

Input Parameter of deep beam

A deep beam subjected to a point load, has the input parameter as shown in fig 1 where

D=overall depth of beam

Av=shear span of beam
L= length of beam
Av= vertical Reinforcement
Ah= horrizontal Reinforcement

FIG 1

4|P a g e
1.5 FAILURE MECHANISM

Failure mechanism for deep beam in shear is based on modified coloumb failure criteria with zero

tension cutoff as proposed by A.F.Ashour for two symmetrical point loading. Two types of failure

mechanism are observed, one is symmetrical mechanism and another is unsymmetrical

mechanism. In symmetrical mechanism three rigid blocks are separated by two yield lines whereas

in unsymmetrical mechanism two rigid blocks are separated by one yield line. Unsymmetrical

mechanism occurs if the symmetry of beam are slightly disturbed. Both types of failure mechanism

are shown in fig 2

FIG 2

5|P a g e
 CHAPTER 2
REVIEW OF LITERATURE

6|P a g e
 In an attempt to develop guidelines for design Ilekar, Faith and Kara [6] choose the genetic

programming approach to develop the shear strength capacity. For development of empirical

relation, they collected shear strength of 104 specimen, from published literature. Collected

specimens include 91 number of simply supported beams and 13 number of simply supported one

way slabs tested against either 3-point bending or 4-point bending. Two types of reinforcement

were used, one was carbon FRP Bars and another was Glass FRP Bars. Specimen were not

provided with any type of shear reinforcement. They included six main parameter as input

variables. While modelling they randomly selected 56 sets for training set and 28 sets were selected

as testing set and remaining were used as validation set.

 Ashour AF [1], presented a mechanism for analysis of shear failure for simply supported

Reinforced Concrete deep beam. He modeled concrete and steel reinforcement as perfectly rigid

plastic material. He considered deep beam to be in a state of plane stress. Modified coloumb failure

criteria with zero tension cut-off were used to study the shear failure mechanism. In his analysis

he considered yield line to represent the failure zone where two rigid blocks were separated along

which in plane displacement discontinuities occur. Optimum shape of the yield line is a hyperbola

as proved by Jensen, Ashour and Morley. He observed modes of shear failure for simply supported

deep beams under 2-point loading andtwo modes of failure were observed, one is symmetrical

mode of failure and another is unsymmetrical mode of failure.

Failure mechanism consist two rigid blocks separated by the yield line. He did the different

parametric study between the input parameters of deep beam and its shear capacity. He considered

a deep beam without web reinforcement to study the effect of longitudinal tensile bar on the shear

strength of deep beam. He also study the effect of shear span-depth ratio on shear capacity of beam.

7|P a g e
 Ashour A.F, Alvarez L.F, Toropov V.V [7], in their work they used the genetic

programming to create an emperical relation to find the shear strength capacity of deep

beam.They obtained experimental database from earlier literarature to create the GP model.

They did the parametric study to known the validation of the predicted GP model on its

input parameter. To reduce the number of input parameters they converted the input

parameter and its shear strength into normalised form. Mathematical operator used in GP

modelling were addition, multiplication, division, square and negation. they choose the

population size of 1000 with mutation rate of 0.001.

8|P a g e
 CHAPTER 3

METHEDOLOGY

9|P a g e
3.1 GENETIC PROGRAMMING

The method to be used while creating the empirical relation will be genetic programming(GP).

GP is a domaininant autonomous, problem-solution approach through which computer programs

are generated to find solutions for the problems. The technique is based on the Darwinian

hyphothesis of ‘survival of the fittest’. Every result predicted by GP is compiled from two sets of

primary nodes; terminals and functions. The terminal set holds nodes that provide a framework to

the GP system while the function set contains nodes that processes values already inside the

system. There are three major evolutionary operators within a GP framework:

Reproduction: it chooses an individual from the initial population to be replicated exactly into

the subsquent generation. In reproduction a strategy is made to kill the under performed program.

There are few methods of selection from which individual is duplicated which includes fitness

measure, selection, rank selection and tournament selection.

Crossover: it is a recombination technique, where two parent results are picked and parts of their

subtree are exchanged in light of fact that each function holds the property ‘closure’ (each tree

member can transform all possible argument values), every crossover operation ought to bring a

legal structure. It follows the folowing principle:

1. Two trees are selected from the population lot.

2. One node is randomly selected from each trees

3. Selected nodes subtrees are exchanged to bring two children of new population

Mutation: it is responsible for irregular changes in a tree before it is brought into the next

population. Dissimilar to crossover, it is abiogennetic and works on one single individual.

Throughout mutation process either all functions or terminals are separated undereneath an

10 | P a g e
 arbitrarily determined node and a new limb is randomly generated or a single node is exchanged

with each other.

 Perspective to portray GP as far as the structures that experiences adaptation are

 Initial structure generation

 Fitness measure test, which assess the structure

 Operation which change the structure

 The state (memory) of the framework at each stage

 The system for terminating the process

 The system for designating an output, and the parameters that control the process

 GP has the five following component

 Function set

 Terminal set

 Fitness function

 Control parameter

 Stop condition

Parameters that control the shear strength of deep beams are input variables and its shear capacity

are the output variable. Initially GP model use single gene and two lengths of head and is incresed

after the subsequent run.

11 | P a g e
FLOW CHART

FIG 3
12 | P a g e
3.2 MATLAB

Matrix Laboratory commonly Known as Matlab is a high level programming language which gives

an interective environment for numerical computation,data analysis, simulation, visualisation and

programming. With the help of matlab one can develop algorithm, can do the data analysis and

can generate models and application. Matlab can be used for a range of application such as image

processing, financial computation, and control system.

3.3 GPTIPS

Genetic Programming tool box for use with MATLAB also known as GPTIPS is a machine

learning method which is biologically inspired. It is done by randomly creating a population of

program represented by tree structures and then to modify the population crossover and mutation

is performed to create a new population. The process is iterated until the program comes with the

best result. It can be configured to produce multigene individual which is one of the important

features of GPTIPS.

13 | P a g e
 CHAPTER 4

MODELLING OF DEEP BEAM

14 | P a g e
4.1 EXPERIMENTAL DATABASE

The following table shows the experimental results obtained from earlier literature and its input

parameter named x1 through x8 where

X1= width of deep beam (in mm)

X2=depth of deep beam (in mm)

X3= shear span ratio

X4= characteristics strength of concrete (in N/mm2)

X5=characteristics strength of longitudinal steel (in N/mm2)

X6=longitudinal reinforcement steel ratio (in %)

X7=characteristics strength of web steel (in N/mm2)

X8=web reinforcement steel ratio (in %)

15 | P a g e
TABLE 1

EXPERIMENTAL RESULT FROM PRIOR LITERATURE

SL. No X1 X2 X3 X4 X5 X6 X7 X8 Shear Force(EXP)

mm mm N/mm2 N/mm2 % N/mm2 % KN
1 100.000 375.000 1.100 43.790 444.980 0.600 445.380 0.000 90.000
2 100.000 375.000 1.100 43.680 444.980 0.600 445.380 0.500 105.000
3 100.000 375.000 1.100 43.740 444.980 0.600 445.380 0.750 125.000
4 100.000 375.000 1.100 43.810 444.980 0.600 445.380 1.250 150.000
5 100.000 375.000 1.100 43.700 444.980 0.600 445.380 2.250 160.000
6 100.000 375.000 1.100 43.790 445.980 2.400 445.280 0.000 185.000
7 100.000 375.000 1.100 43.680 445.980 2.400 445.280 0.500 292.000
8 100.000 375.000 1.100 43.740 445.980 2.400 445.280 0.750 315.000
9 100.000 375.000 1.100 43.810 445.980 2.400 445.280 1.250 320.000
10 100.000 375.000 1.100 43.700 445.980 2.400 445.280 2.250 330.000
11 100.000 375.000 1.100 43.000 445.380 0.800 445.280 0.000 98.250
12 100.000 275.000 1.100 43.270 444.120 0.600 445.280 0.000 84.000
13 100.000 225.000 1.100 47.550 444.120 0.600 445.280 0.000 88.000
14 98.000 270.000 1.130 24.913 430.940 2.150 437.350 0.280 161.240
15 98.000 270.000 1.130 24.040 430.940 2.360 437.350 0.280 148.341
16 98.000 270.000 1.130 21.420 430.940 2.460 437.350 0.280 141.224
17 98.000 270.000 1.130 27.396 430.940 2.460 437.350 0.280 170.937
18 98.000 270.000 1.130 28.040 430.940 2.670 437.350 0.280 184.080
19 98.000 270.000 1.130 28.913 430.940 2.150 437.350 0.630 174.495
20 98.000 270.000 1.130 26.339 430.940 2.360 437.350 0.630 170.581
21 98.000 270.000 1.130 27.120 430.940 2.460 437.350 0.630 171.915
22 98.000 270.000 1.130 25.512 430.940 2.670 437.350 0.630 161.907
23 98.000 270.000 1.130 24.547 430.940 2.150 437.350 1.250 161.018
24 98.000 270.000 1.130 25.649 430.940 2.360 437.350 1.250 172.716
25 98.000 270.000 1.130 27.764 430.940 2.460 437.350 1.250 178.543
26 98.000 270.000 1.130 26.569 430.940 2.670 437.350 1.250 168.134
27 98.000 270.000 1.363 29.419 430.940 2.150 437.350 0.240 147.451
28 98.000 270.000 1.363 26.799 430.940 2.360 437.350 0.240 143.559
29 98.000 270.000 1.363 27.764 430.940 2.460 437.350 0.240 140.334
30 98.000 270.000 1.363 26.017 430.940 2.670 437.350 0.240 153.345

16 | P a g e
TABLE 1 (CONTINUED)

31 98.000 270.000 1.363 25.557 430.940 2.150 437.350 0.420 128.992

32 98.000 270.000 1.363 25.328 430.940 2.360 437.350 0.420 131.216
33 98.000 270.000 1.363 23.305 430.940 2.460 437.350 0.420 126.101
34 98.000 270.000 1.363 29.051 430.940 2.460 437.350 0.420 149.898
35 98.000 270.000 1.363 26.339 430.940 2.670 437.350 0.420 145.227
36 98.000 270.000 1.363 21.651 430.940 2.150 437.350 0.630 130.771
37 98.000 270.000 1.363 27.212 430.940 2.150 437.350 0.770 158.949
38 98.000 270.000 1.363 25.328 430.940 2.360 437.350 0.770 158.349
39 98.000 270.000 1.363 25.649 430.940 2.460 437.350 0.770 155.013
40 98.000 270.000 1.363 27.535 430.940 2.670 437.350 0.770 166.133
41 98.000 270.000 1.363 22.800 430.940 2.150 437.350 1.250 153.456
42 98.000 270.000 1.693 25.649 430.940 2.150 437.350 0.180 118.984
43 98.000 270.000 1.693 29.189 430.940 2.360 437.350 0.180 123.432
44 98.000 270.000 1.693 30.247 430.940 2.460 437.350 0.180 130.994
45 98.000 270.000 1.693 29.051 430.940 2.670 437.350 0.180 122.320
46 98.000 270.000 1.693 26.477 430.940 2.150 437.350 0.310 124.099
47 98.000 270.000 1.693 25.649 430.940 2.360 437.350 0.310 103.638
48 98.000 270.000 1.693 25.741 430.940 2.360 437.350 0.310 115.314
49 98.000 270.000 1.693 27.259 430.940 2.460 437.350 0.310 124.544
50 98.000 270.000 1.693 27.672 430.940 2.670 437.350 0.310 124.099
51 98.000 270.000 1.693 28.040 430.940 2.150 437.350 0.560 140.779
52 98.000 270.000 1.693 22.064 430.940 2.360 437.350 0.560 124.989
53 98.000 270.000 1.693 24.363 430.940 2.460 437.350 0.560 127.658
54 98.000 270.000 1.693 25.328 430.940 2.670 437.350 0.560 137.221
55 98.000 270.000 1.693 26.109 430.940 2.150 437.350 0.770 146.462
56 98.000 270.000 1.693 24.731 430.940 2.360 437.350 0.630 128.547
57 98.000 270.000 1.693 25.649 430.940 2.460 437.350 0.770 152.344
58 98.000 270.000 1.693 24.684 430.940 2.460 437.350 0.770 152.566
59 98.000 270.000 1.693 28.316 430.940 2.670 437.350 0.770 159.461
60 98.000 270.000 1.693 21.420 430.940 2.670 437.350 0.420 89.405
61 73.500 671.000 0.379 28.683 286.830 0.520 279.940 2.450 238.858
62 73.500 551.000 0.461 32.728 286.830 0.630 279.940 2.450 224.179
63 73.500 424.000 0.599 28.316 286.830 0.800 279.940 2.450 189.485
64 73.500 301.000 0.844 28.316 286.830 1.090 279.940 2.450 164.131
65 73.500 181.000 1.403 28.867 286.830 1.730 279.940 2.450 89.405
66 73.500 666.000 0.381 34.797 286.830 0.520 303.380 0.860 249.088
67 73.500 544.000 0.467 24.823 286.830 0.630 303.380 0.860 224.179
68 73.500 424.000 0.599 26.477 286.830 0.800 303.380 0.860 215.283
69 73.500 304.000 0.836 30.339 286.830 1.090 303.380 0.860 139.667
70 73.500 179.000 1.419 26.844 286.830 1.730 303.380 0.860 99.635
71 73.500 632.000 0.402 24.731 279.940 1.140 279.940 0.610 239.302
72 73.500 512.000 0.496 25.649 279.940 1.240 279.940 0.610 208.166
73 73.500 392.000 0.648 26.844 279.940 1.410 279.940 0.610 172.582
74 73.500 274.000 0.927 29.235 279.940 1.700 279.940 0.610 127.213
75 73.500 159.000 1.597 30.063 279.940 2.340 279.940 0.610 77.840
17 | P a g e
4.2 IS CODE COMPUTATION

From [9]

Shear capacity of section of deep beam(Vu) = Vc+Vs

Vc= Shear capacity carried by concrete

Vs= Shear capacity carried by steel

1. Vc= C1(1-0.35(av/d))ftbD

Where,

C1 = 0.72 (a constant)

Av = shear span

D = total depth

b = width

ft = tensile strength of concrete

2. Vs =C2∑Ai*(yi/D)*sin2α

Where,

C2 = 225 (for fe415)

A1= horrizontal reinforcement area

A2= vertical reinforcement area

α= angle between bar considered and critical diagnol crack

n= no. of bars including tension steel cut by assumed crack

yi= depth from top of beam to the point where bar intersects the critical diagonal crack line

D= total depth of the beam

18 | P a g e
4.3 ACI CODE COMPUTATION

[ From 9]

Shear capacity of section (Vu) = Vc+Vs

Vc= Shear capacity carried by concrete

Vs= Shear capacity carried by steel

1. Vc= 0.13fck0.5(bd)

Where,

d = effective depth

b = width

fck = compressive strength of concrete

2. Vs =0.85[fyd*{(Av/S1)*(1+f)/12+(Ah/S2)(11-f)/12}]

Where,

f= Ln/d

Ah= horrizontal reinforcement area

Av= vertical reinforcement area

S1= Spacing of vertical reinforcement

S2= Spacing of horizontal reinforcement

19 | P a g e
4.4 RUN PARAMETER

There are large number of test results in prior literature for R C deep beam. Out of them, results of

75 deep beams are taken to create the GP Response. Out of 75 numbers of data 52 nos. data are

randomly selected and used as training indices, and 25 data are randomly selected and used as test

indices. Out of 52 data which are selected for training indices 33 data are used for validation

purpose. Initially 100 population size were selected and increased to 1000 for better result. The

mathematical operators used while creating the GP response are addition, multiplication, sin,

square, substraction and exponential. Rate of mutation and crossover used are 0.1 and 0.85

respectively.The input parameters should be in the following range to obtained the best result.

Input parameter Range

Min Max
compressive strength of concrete 21 48
width of beam 73.5 100
depth of beam 159 671
shear span of beam 0.3785 1.6927
tensile steel reinforcement ratio 0.52 2.67
charcteristics stress of tensile steel 279.94 445.98
vertical shear reinforcement ratio 0 2.45
characteristics stress of web steel 279.94 445.98

20 | P a g e
 CHAPTER 5

RESULT &DISCUSSION

21 | P a g e
5.1 INPUT FREQUENCY ANALYSIS

The graphical input frequency analysis of single model or of a user specified fraction of the

population is used to provide the identification of input variables that are significant to the output.

The most significant input variable is X4, where X4 is the characteristics compressive strength of

concrete.

FIG 4

22 | P a g e
5.2 FITNEESS GENERATION

It is the plot of log of fitness found corrosponding to its generation and best fitness found at 398

generation

FIG 5

23 | P a g e
5.3 ERROR ANALYSIS
In this the predicted output are compared with the actual values on training set, test set and

validation test. Maximum deviation from actual in training set error is 7.3059, in test set error is

15.3164 and in validation set error is 7.2325.

FIG 6

24 | P a g e
5.4 SCATTER PLOT

It is the scatter plot of predicted output value and the actual value. It shows that the GP results are

close to the Experimental results.

FIG 7

25 | P a g e
5.5 GENE WEIGHTAGE

As number of gene involved while creating the model is 4 and the most significant gene is the bias

gene with gene 2 as least significant.

FIG 8

26 | P a g e
5.6 EMPERICAL RELATION

Y=0.5976X5+0.5976X4X6+0.06164e(X1+X4+SIN(X8))square(X5+X6)+0.2096

(X3+sin(sin(x8))-2.895)(x1+x3-x5+0.4603)-

0.09253square(sin(x8))(x42+X3)(sin(X8)-X4(X3-X6)+4.303)+0.2668

Where,

Y= shear strength of deep beam (in KN)

X1= width of deep beam (in mm)

X2=depth of deep beam (in mm)

X3= shear span ratio

X4= characteristics strength of concrete (in N/mm2)

X5=characteristics strength of longitudinal steel (in N/mm2)

X6=longitudinal reinforcement steel ratio (in %)

X7=characteristics strength of web steel (in N/mm2)

X8=web reinforcement steel ratio (in %)

27 | P a g e
5.7 COMPARISION OF RESULTS (TABULATION FORM)
TABLE 2
RESULT COMPARISON BETWEEN EXPERIMENTAL AND GP

SL. No X1 X2 X3 X4 X5 X6 X7 X8 Shear Force(EXP) Shear force(GP)

mm mm N/mm2 N/mm2 % N/mm2 % KN KN

1 100.000 375.000 1.100 43.790 444.980 0.600 445.380 0.000 90.000 88.935
2 100.000 375.000 1.100 43.680 444.980 0.600 445.380 0.500 105.000 108.566
3 100.000 375.000 1.100 43.740 444.980 0.600 445.380 0.750 125.000 125.379
4 100.000 375.000 1.100 43.810 444.980 0.600 445.380 1.250 150.000 152.509
5 100.000 375.000 1.100 43.700 444.980 0.600 445.380 2.250 160.000 151.836
6 100.000 375.000 1.100 43.790 445.980 2.400 445.280 0.000 185.000 180.849
7 100.000 375.000 1.100 43.680 445.980 2.400 445.280 0.500 292.000 259.420
8 100.000 375.000 1.100 43.740 445.980 2.400 445.280 0.750 315.000 297.680
9 100.000 375.000 1.100 43.810 445.980 2.400 445.280 1.250 320.000 325.569
10 100.000 375.000 1.100 43.700 445.980 2.400 445.280 2.250 330.000 323.917
11 100.000 375.000 1.100 43.000 445.380 0.800 445.280 0.000 98.250 94.198
12 100.000 275.000 1.100 43.270 444.120 0.600 445.280 0.000 84.000 90.541
13 100.000 225.000 1.100 47.550 444.120 0.600 445.280 0.000 88.000 83.817
14 98.000 270.000 1.130 24.913 430.940 2.150 437.350 0.280 161.240 164.030
15 98.000 270.000 1.130 24.040 430.940 2.360 437.350 0.280 148.341 155.816
16 98.000 270.000 1.130 21.420 430.940 2.460 437.350 0.280 141.224 141.408
17 98.000 270.000 1.130 27.396 430.940 2.460 437.350 0.280 170.937 167.298
18 98.000 270.000 1.130 28.040 430.940 2.670 437.350 0.280 184.080 169.327
19 98.000 270.000 1.130 28.913 430.940 2.150 437.350 0.630 174.495 173.458
20 98.000 270.000 1.130 26.339 430.940 2.360 437.350 0.630 170.581 166.686
21 98.000 270.000 1.130 27.120 430.940 2.460 437.350 0.630 171.915 168.409
22 98.000 270.000 1.130 25.512 430.940 2.670 437.350 0.630 161.907 159.468
23 98.000 270.000 1.130 24.547 430.940 2.150 437.350 1.250 161.018 167.184
24 98.000 270.000 1.130 25.649 430.940 2.360 437.350 1.250 172.716 169.671
25 98.000 270.000 1.130 27.764 430.940 2.460 437.350 1.250 178.543 181.308
26 98.000 270.000 1.130 26.569 430.940 2.670 437.350 1.250 168.134 175.935
27 98.000 270.000 1.363 29.419 430.940 2.150 437.350 0.240 147.451 148.875
28 98.000 270.000 1.363 26.799 430.940 2.360 437.350 0.240 143.559 141.063
29 98.000 270.000 1.363 27.764 430.940 2.460 437.350 0.240 140.334 143.923
30 98.000 270.000 1.363 26.017 430.940 2.670 437.350 0.240 153.345 133.592
31 98.000 270.000 1.363 25.557 430.940 2.150 437.350 0.420 128.992 145.499
32 98.000 270.000 1.363 25.328 430.940 2.360 437.350 0.420 131.216 140.854
33 98.000 270.000 1.363 23.305 430.940 2.460 437.350 0.420 126.101 130.767
34 98.000 270.000 1.363 29.051 430.940 2.460 437.350 0.420 149.898 153.171
35 98.000 270.000 1.363 26.339 430.940 2.670 437.350 0.420 145.227 140.488

28 | P a g e
TABLE 2 (Continued)

36 98.000 270.000 1.363 21.651 430.940 2.150 437.350 0.630 130.771 145.170
37 98.000 270.000 1.363 27.212 430.940 2.150 437.350 0.770 158.949 156.770
38 98.000 270.000 1.363 25.328 430.940 2.360 437.350 0.770 158.349 150.137
39 98.000 270.000 1.363 25.649 430.940 2.460 437.350 0.770 155.013 149.704
40 98.000 270.000 1.363 27.535 430.940 2.670 437.350 0.770 166.133 156.993
41 98.000 270.000 1.363 22.800 430.940 2.150 437.350 1.250 153.456 140.233
42 98.000 270.000 1.693 25.649 430.940 2.150 437.350 0.180 118.984 114.296
43 98.000 270.000 1.693 29.189 430.940 2.360 437.350 0.180 123.432 125.545
44 98.000 270.000 1.693 30.247 430.940 2.460 437.350 0.180 130.994 129.965
45 98.000 270.000 1.693 29.051 430.940 2.670 437.350 0.180 122.320 126.158
46 98.000 270.000 1.693 26.477 430.940 2.150 437.350 0.310 124.099 125.005
47 98.000 270.000 1.693 25.649 430.940 2.360 437.350 0.310 103.638 119.017
48 98.000 270.000 1.693 25.741 430.940 2.360 437.350 0.310 115.314 119.404
49 98.000 270.000 1.693 27.259 430.940 2.460 437.350 0.310 124.544 124.712
50 98.000 270.000 1.693 27.672 430.940 2.670 437.350 0.310 124.099 125.743
51 98.000 270.000 1.693 28.040 430.940 2.150 437.350 0.560 140.779 143.184
52 98.000 270.000 1.693 22.064 430.940 2.360 437.350 0.560 124.989 123.946
53 98.000 270.000 1.693 24.363 430.940 2.460 437.350 0.560 127.658 128.949
54 98.000 270.000 1.693 25.328 430.940 2.670 437.350 0.560 137.221 129.069
55 98.000 270.000 1.693 26.109 430.940 2.150 437.350 0.770 146.462 148.986
56 98.000 270.000 1.693 24.731 430.940 2.360 437.350 0.630 128.547 136.553
57 98.000 270.000 1.693 25.649 430.940 2.460 437.350 0.770 152.344 143.674
58 98.000 270.000 1.693 24.684 430.940 2.460 437.350 0.770 152.566 140.236
59 98.000 270.000 1.693 28.316 430.940 2.670 437.350 0.770 159.461 155.202
60 98.000 270.000 1.693 21.420 430.940 2.670 437.350 0.420 89.405 97.267
61 73.500 671.000 0.379 28.683 286.830 0.520 279.940 2.450 238.858 236.9481
62 73.500 551.000 0.461 32.728 286.830 0.630 279.940 2.450 224.179 196.346
63 73.500 424.000 0.599 28.316 286.830 0.800 279.940 2.450 189.485 193.8205
64 73.500 301.000 0.844 28.316 286.830 1.090 279.940 2.450 164.131 153.0452
65 73.500 181.000 1.403 28.867 286.830 1.730 279.940 2.450 89.405 96.5815
66 73.500 666.000 0.381 34.797 286.830 0.520 303.380 0.860 249.088 199.0221
67 73.500 544.000 0.467 24.823 286.830 0.630 303.380 0.860 224.179 238.8321
68 73.500 424.000 0.599 26.477 286.830 0.800 303.380 0.860 215.283 201.1875
69 73.500 304.000 0.836 30.339 286.830 1.090 303.380 0.860 139.667 145.1783
70 73.500 179.000 1.419 26.844 286.830 1.730 303.380 0.860 99.635 97.1695
71 73.500 632.000 0.402 24.731 279.940 1.140 279.940 0.610 239.302 235.2198
72 73.500 512.000 0.496 25.649 279.940 1.240 279.940 0.610 208.166 210.1324
73 73.500 392.000 0.648 26.844 279.940 1.410 279.940 0.610 172.582 175.2592
74 73.500 274.000 0.927 29.235 279.940 1.700 279.940 0.610 127.213 126.358
75 73.500 159.000 1.597 30.063 279.940 2.340 279.940 0.610 77.840 76.7757

29 | P a g e
5.8 COMPARISON OF RESULTS (GRAPHICAL FORM)

It is the result comparison between value obtained from GP response and actual value obtained

from earlier literature.

EXPERIMENTAL VS GP
350.000 350.000

300.000 300.000

250.000 250.000

200.000 200.000

150.000 150.000

100.000 100.000

50.000 50.000

0.000 0.000
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75

Shear Force(EXP) KN Shear force(GP) KN

X-AXIS= SL. NO.

Y-AXIS= SHEAR STRENGTH IN KN/m2

FIG 9

30 | P a g e
 CHAPTER 6
CALCULATION USING DESIGN CODE

31 | P a g e
6.1 INPUT VARIABLE
The input parameter tabulated below are used to calculate the shear strength using IS code, ACI
code and emperical relation developed by GP Response

TABLE 3
INPUT VARIABLE

SL.NO. b D Av/d fck fy long.steel fy vert. steel

mm mm N/mm2
x1 x2 x3 x4 x5 x6 x7 x8
1 80 350 1.100 30 415 0.28 415 0.00
2 85 400 1.100 30 415 0.42 415 0.00
3 90 450 1.100 30 415 0.60 415 0.00
4 100 500 1.100 30 415 0.60 415 0.00
5 80 550 1.100 30 415 1.20 415 0.00
6 85 600 1.100 30 415 1.80 415 0.00
7 90 350 1.100 30 415 2.40 415 0.00
8 100 400 1.100 30 415 0.60 415 0.00
9 100 450 1.100 30 415 0.60 415 0.50
10 100 500 1.100 30 415 0.60 415 0.75
11 100 550 1.100 30 415 0.60 415 1.25
12 80 600 1.100 30 415 0.60 415 2.25
13 85 400 0.763 30 415 2.40 415 0.00
14 90 400 0.763 30 415 2.40 415 0.50
15 100 400 0.763 25 415 2.40 415 0.75
16 100 400 0.763 25 415 2.40 415 1.25
17 100 400 0.763 25 415 2.40 415 2.25
18 100 400 0.763 25 415 0.80 415 0.00
19 100 350 0.871 25 415 0.80 415 0.00
20 100 400 0.763 25 415 0.80 415 0.00
21 80 450 0.678 25 415 0.80 415 0.00
22 85 500 0.610 25 415 0.80 415 0.00
23 90 550 1.670 25 415 0.80 415 0.00
24 100 600 0.508 35 415 0.60 415 0.00
25 100 350 0.871 35 415 0.60 415 0.00
26 100 300 1.227 35 415 0.60 415 0.00
27 100 250 1.472 35 415 0.60 415 0.00
28 100 343 1.073 35 415 2.15 415 0.28
29 100 343 1.073 35 415 2.36 415 0.28
30 100 343 1.073 35 415 2.46 415 0.28

32 | P a g e
TABLE 3 (Continued)

31 80 343 1.073 35 415 2.46 415 0.28

32 85 343 1.073 35 415 2.60 415 0.28
33 90 343 1.073 35 415 2.15 415 0.63
34 100 350 1.051 35 415 2.36 415 0.63
35 100 400 0.920 30 415 2.46 415 0.63
36 100 450 0.818 30 415 2.67 415 0.63
37 100 500 0.736 30 415 2.15 415 1.25
38 100 550 0.669 30 415 2.36 415 1.25
39 100 600 0.613 30 415 2.46 415 1.25
40 100 343 1.073 30 415 2.60 415 1.25
41 100 343 1.332 30 415 2.15 415 0.24
42 100 343 1.332 30 415 2.36 415 0.24
43 100 343 1.332 30 415 2.46 415 0.24
44 80 343 1.332 30 415 2.67 415 0.24
45 85 343 1.332 30 415 2.15 415 0.42
46 90 343 1.332 30 415 2.36 415 0.42
47 100 343 1.332 30 415 2.46 415 0.42
48 100 343 1.332 30 415 2.46 415 0.42
49 100 350 1.306 25 415 2.67 415 0.42
50 100 400 1.143 25 415 2.15 415 0.63
51 100 450 1.016 25 415 2.15 415 0.77
52 100 500 0.914 25 415 2.36 415 0.77
53 100 550 0.831 25 415 2.46 415 0.77
54 100 600 0.762 25 415 2.67 415 0.77
55 100 343 1.332 25 415 2.15 415 1.25
56 100 343 1.332 25 415 2.15 415 0.18
57 80 343 1.332 25 415 2.36 415 0.18
58 85 343 1.332 35 415 2.46 415 0.18
59 90 343 1.332 35 415 2.67 415 0.18
60 100 343 0.741 35 415 2.15 415 0.31
61 100 343 0.741 35 415 2.36 415 0.31
62 100 343 0.741 35 415 2.36 415 0.31
63 100 343 0.741 35 415 2.46 415 0.31
64 100 343 0.741 35 415 2.60 415 0.31
65 100 343 0.741 35 415 2.15 415 0.56
66 100 350 0.726 35 415 2.36 415 0.56

33 | P a g e
TABLE 3 (Continued)

67 100 400 0.635 35 415 2.46 415 0.56

68 100 450 0.564 35 415 2.60 415 0.56
69 100 500 0.508 30 415 2.15 415 0.77
70 100 550 0.462 30 415 2.36 415 0.63
71 100 600 0.423 30 415 2.46 415 0.77
72 100 343 0.741 30 415 2.46 415 0.77
73 100 343 0.741 30 415 2.67 415 0.77
74 100 343 0.741 30 415 2.67 415 0.42
75 100 625 1.200 30 415 0.52 415 2.45
76 100 613 1.200 30 415 0.63 415 2.45
77 100 350 1.200 30 415 0.80 415 2.45
78 100 400 1.200 30 415 1.09 415 2.45
79 100 450 1.200 30 415 1.73 415 2.45
80 100 500 1.200 30 415 0.52 415 0.86
81 100 550 1.200 30 415 0.63 415 0.86
82 100 600 1.200 30 415 0.80 415 0.86
83 100 368 1.050 25 415 1.09 415 0.86
84 100 245 1.050 25 415 1.73 415 0.86
85 100 635 1.050 25 415 1.14 415 0.61
86 100 613 1.050 25 415 1.24 415 0.61
87 100 490 1.050 25 415 1.41 415 0.61
88 100 368 1.050 25 415 1.70 415 0.61
89 100 245 1.050 25 415 2.34 415 0.61
90 100 300 1.100 25 415 0.60 415 0.63
91 100 325 1.100 25 415 0.60 415 0.77
92 100 350 1.100 35 415 0.60 415 0.77
93 100 400 1.100 35 415 2.15 415 0.77
94 100 450 1.100 35 415 2.36 415 0.42
95 100 500 1.100 35 415 2.46 415 2.45
96 100 550 1.100 35 415 2.46 415 2.45
97 100 600 1.100 35 415 2.60 415 2.45
98 100 500 1.100 35 415 2.15 415 2.45
99 100 525 1.100 35 415 2.15 415 2.45
100 100 550 1.100 35 415 2.36 415 0.86
101 100 575 1.100 35 415 2.46 415 0.86
102 100 600 1.100 35 415 1.60 415 0.86

34 | P a g e
6.2 COMPARISON OF RESULTS (TABULAR FORM)

SHEAR CAPACITY
sl.no Shear capacity (IS Code) Shear capacity(ACI Code) shear capacity using gp
KN KN KN

1 42.233 17.089 222.523

2 56.623 98.621 206.732
3 75.722 119.340 190.537
4 93.894 149.175 158.612
5 112.488 132.600 218.243
6 165.510 154.976 203.055
7 118.026 89.505 189.513
8 74.376 116.025 158.612
9 108.773 153.765 158.500
10 135.471 184.450 161.100
11 180.647 230.421 174.755
12 212.688 247.452 235.482
13 108.213 98.621 264.388
14 127.607 117.783 232.805
15 143.986 118.955 180.717
16 158.462 133.800 184.077
17 187.414 163.490 183.901
18 75.947 96.688 240.566
19 67.093 82.875 219.746
20 75.947 96.688 240.566
21 67.559 88.400 320.381
22 78.839 105.666 319.919
23 96.643 124.313 141.693
24 120.298 212.713 236.326
25 69.288 116.025 156.543
26 56.737 96.688 104.706
27 44.287 77.350 84.046
28 131.345 122.199 161.898
29 138.753 122.199 169.243
30 142.281 122.199 172.956

35 | P a g e
31 113.825 97.759 234.799
32 125.137 103.869 225.388
33 129.323 119.970 195.534
34 153.080 136.137 174.357
35 165.023 137.063 176.118
36 182.358 154.564 191.740
37 194.150 194.592 208.502
38 208.643 213.004 222.823
39 218.410 231.416 234.613
40 178.055 136.777 171.524
41 136.880 105.998 129.608
42 145.735 105.998 130.467
43 149.952 105.998 130.979
44 127.046 84.798 193.684
45 122.800 95.751 179.206
46 137.993 101.384 165.220
47 157.543 112.649 134.356
48 157.543 112.649 134.356
49 165.674 98.481 114.164
50 167.168 121.022 143.266
51 185.637 141.374 155.605
52 205.454 156.318 161.828
53 218.621 171.262 169.919
54 235.502 186.207 174.965
55 176.326 127.129 128.128
56 131.205 87.592 117.677
57 112.048 70.074 174.265
58 127.767 101.974 187.659
59 143.253 107.972 180.197
60 111.554 121.073 215.398
61 116.458 121.073 223.134
62 116.458 121.073 223.134
63 118.792 121.073 227.039
64 122.061 121.073 232.748
65 117.391 127.327 207.565
66 123.604 130.149 218.107

36 | P a g e
 67 134.598 150.310 237.769
68 145.623 170.470 256.977
69 141.707 171.991 234.462
70 150.050 185.343 249.043
71 161.525 207.404 253.193
72 125.518 116.392 199.041
73 130.421 116.392 201.627
74 122.249 107.636 209.048
75 298.245 343.655 161.951
76 300.363 336.738 159.329
77 169.495 185.154 155.780
78 210.269 213.972 151.133
79 273.484 242.790 147.163
80 139.643 192.152 157.013
81 161.827 213.024 155.025
82 189.853 233.897 152.338
83 115.059 116.687 181.596
84 87.481 73.069 162.848
85 193.081 196.904 179.493
86 192.677 189.604 176.528
87 160.651 148.788 171.613
88 128.566 108.304 163.595
89 95.764 67.488 147.530
90 71.091 86.841 191.603
91 82.389 99.509 192.933
92 96.475 141.376 134.408
93 178.293 164.335 162.148
94 195.929 172.479 167.148
95 337.679 289.269 179.159
96 374.472 320.130 179.159
97 420.750 350.991 187.744
98 320.494 289.269 163.087
99 337.936 304.700 163.087
100 270.376 237.869 171.988
101 290.035 249.560 176.959
102 245.265 261.251 144.739

37 | P a g e
6.3 COMPARISON OF RESULTS (GRAPHICAL FORM)

IS CODE VS GP
450.000 350.000
400.000
300.000
350.000
250.000
300.000
250.000 200.000

200.000 150.000
150.000
100.000
100.000
50.000
50.000
0.000 0.000
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101

Shear capacity (IS Code) KN shear capacity using gp KN

FIG 10

ACI CODE VS GP
400.000

350.000

300.000

250.000

200.000

150.000

100.000

50.000

0.000
10

46

82
1
4
7

13
16
19
22
25
28
31
34
37
40
43

49
52
55
58
61
64
67
70
73
76
79

85
88
91
94
97
100
103

Shear capacity(ACI Code) KN shear capacity using gp KN

FIG 11

38 | P a g e
CONCLUSION

An empirical model was predicted to determine the shear strength capacity of deep beam using

Genetic Programming (GP). Good validation of results were obtained between experimental

results and predicted results. GP Prediction would have been more accurate if the experimental

results were more. From the empirical relationship, shear strength capacity of some imput variables

were computed and result were validated with the output obtained using IS Code and ACI Code.

The effect of shear span to depth ratio, horizontal steel ratio and vertical web steel ratio are most

significant on the shear strength capacity of deep beam.

39 | P a g e
REFERENCES

1. Ashour AF. Shear capacity of reinforced concrete deep beams. Struct Eng J ASCE

2000;126:1045–52

2. Koza JR. Genetic programming: on the programming of computers by means of natural

selection. London: MIT Press; 1992.

3. Kong FK, Robins PJ, Kirby DP, Short DR. Deep beams with inclined web reinforcement. ACI

J 1972;69:172–6.

4. Smith KN, Vantsiotis AS. Shear strength of deep beams.ACI Struct J 1982;79:201–13.

5. Londhe R.S, Shear strength analysis and prediction of reinforced concrete transfer beams in

high-rise buildings. Structural Engineering and Mechanics, Vol. 37, No. 1 (2011) 39-59

6. Ilker Fatih Kara, Prediction of shear strength of FRP-reinforced concrete beams without

stirrups based on genetic programming. Advances in Engineering Software 42 (2011) 295–304

7. Ashour A.F, Alvarez L.F, Toropov V.V . Empirical modelling of shear strength of RC

deep beams by genetic programming. Computers and Structures 81 (2003) 331–338

8. ACI Committee 318. Building code requirements for structural concrete (ACI 318-99) and

commentary (ACI318R-99). Detroit: American Concrete Institute; (1999).

9. Varghese P.C, Advanced reinforced concrete design. New Delhi: Phi learning private limited.

(2012)

10. IS 456. “Indian Standard for Plain and Reinforced Concrete.” Code of practice, bureau of

Indian standards, New Delhi.(2000)

40 | P a g e