The aim of this study of concrete which have two different architectures in ANN system to evaluate the effect of Fly ash, Metakaolin, Silica fume, Bottom ash.
ANN nonlinear information devices, which are built from interconnected elementary processing devices called neurons.
An ANN is configured for a specific application, such as data classification, through a learning process.
ANN’s are a type of artificial intelligence that attempts to imitate the way a human brain works.
Analysis of durability of high performance concrete using Artificial Neural Network (ANN)
1. Analysis of durability of high performance
concrete using Artificial Neural Network (ANN)
Presented By:-
Mr. Priyank Bharat Panchmiya
Guided By:-
Prof. V.M. Pandit Sir
3. INTRODUCTION:-
• In view of the global sustainable development, it is imperative that
supplementary cementing materials be used in replace of cement in
concrete industry.
• Most supplementary cementing materials are silica fume, fly ash,
blast-furnace slag.
• Silica fume increase the strength of the concrete but affect the
workability of the concrete. & Fly ash increase the workability of
concrete but affect the strength of concrete.
• The aim of this study of concrete which have two different
architectures in ANN system to evaluate the effect of Fly ash,
Metakaolin, Silica fume, Bottom ash.
4. LITERATURE REVIEW :-
NAME OF SCIENTIST
USE ANN
YEAR DESCRIPTION
1. Nooraei et al. 2007 Compressive strength of concrete after 28 day. The material
use in concrete are cement, water, silica fume, super
plasticizer, fine aggregate & coarse aggregate using ANN
2. Atici et al. 2009 The material use in concrete is blast furnace slag & fly ash.
After these compressive strength of concrete estimate using
ANN
3. Serkan subas 2009 Estimate ability of the utilizing different amount of the class C
fly ash on the mechanical properties of cement using ANN
4. Seyed et al. 2011 Studied the application of ANN to predict compressive
strength of high strength concrete
5. Vijay et al. 2013 Predicted the compressive strength of concrete using ANN
6. Sakshiguta et al. 2013 Predicted the compressive strength of concrete containing
nano-silica using ANN
7. Wankhade et al. 2013 Predicted the compressive strength of concrete using ANN
5. METHODOLOGY:-
1) EXPERIMENTAL INVESTIGATION:-
• M30 grade of concrete were used in investigation.
• Mix design was done based on IS 1062-2009(17).
• The concrete mix proportion 1:1:73:3:2 with w/c 0.45.
• 12 high performance concrete (HPC) were prepared.
• The conventional concrete mix CC & combinations of HPC mixes given in table 1.
• Volume of water is 172.8lit/m3 & Coarse aggregate is 1220lit/m3 kept constant
while mixing.
• The mix combination and mix proportions are given in table 1 & 2.
• The selected 4 HPC mixes are having maximum compressive strength at 28 days.
8. 2) PREPARATION OF TEST SPECIMEN :-
• Concrete cubes were casted for all five mixes .In total 71 were casted for all mixes.
• All the materials were thoroughly mixed in dry state by machine so as to obtain uniform colour.
• The required percentage pf superplasticizer was added to the water calculated for the particular
mix.
• The slump test were carried out.
• After that specimen cast in steel mould & concrete were compacted on vibrating table.
• The specimen were demoulded after 24 hours & cured in water for 28 days.
• The results were confirming by IS 516-1959(16).
• The cube were tested using CTM of capacity 2000KN.
9. 3) ARTIFICIAL NEURAL NETWORK (ANN):-
• ANN nonlinear information devices, which are built from interconnected
elementary processing devices called neurons.
• An ANN is configured for a specific application, such as data classification,
through a learning process.
• Ann’s are a type of artificial intelligence that attempts to imitate the way a
human brain works.
10. 4) FEED FORWARD NEURAL NETWORK:-
• In a feed forward neural network, the artificial neurons are arranged in layers,
and all the neurons in each layer have connections to all the neurons in the next
layer.
• However, there is no connection between neurons of the same layer or the
neurons which are not in successive layers.
• The feed forward network consists of one input layer, one or two hidden layers
and one output layer of neurons.
14. RESULTS AND DISCUSSION:-
• In the training of ANN-I and ANN-II models, various experimental data are
used.
• In the ANN-I and ANN-II models, 71 data of Experimental results were
used for training
• All results obtained from experimental studies and predicted by using the
training results of ANN-I models for 28, 56, 90 and 120 days
15. CONCLUSIONS:-
• In this Study, using these beneficial properties of artificial neural networks
in order to predict the 28, 56, 90 and 120 days compressive strength values
of concrete containing Industrial Byproducts with attempting experiments
were developed two different architectures namely ANN-I and ANN-II.
• In two models developed on ANN method, a multilayer feed forward neural
network in a back propagation algorithm were used. In ANN-I model, one
hidden layer were selected.
• In ANN-II model, two hidden layers were selected. In the first hidden layers
10 neurons and in the second hidden layer 10 neurons were determined.
• The compressive strength values predicted from training for ANN-I &
ANN-II models were very close to the experimental results.
16. REFERENCES:-
1) Dr. B. Vidivelli
(Professor, Department of Civil & Structural Engineering)
2) A. Jayaranjini
(Research Scholar,Department of Civil & Structural Engineering,
Annamalai University, Tamilnadu, India)