Design Of Normal Concrete Mixes Using Neural Network Model

The most important factor in determining the quality of concrete is its strength. In order to achieve the required strength, a right proportion of materials in concrete
such as water, cement, sand and course aggregate, need to be identified. The present mix design methods such as AC1 and DoE methods, which involve numerous calculations, design charts and table look-up are seem to be tedious and
lengthy. The purpose of this project is to develop a simpler and generalized concrete mix design method using neural network techniques. A procedure for developing work models using back propagation networks is presented,
and a number of issues related to data preparation are described to facilitate the development of efficient application. The findings of this project show that the
application of neural network is capable of providing solutions to the civil engineering problem, particularly in designing the concrete mixes.