Development of Artificial Neural Network to Predict the Concrete Strength
In recent decades, a number of machine learning algorithms has proved themselves as a vital need for a broad range of applications in the structural health domain. Here in this work, a machine learning based Artificial Neural Network model has been developed to predict the strength of the concrete from 1030 cases, donated to the UCI machine learning repository. As a result, a number of topologies of the model are developed whose performance evaluations are done through the errors and correlation factors associated with each one of them. Apart from this, a comparative analysis of the predicted strength with the real is also done at the end to signify the performance of the model with much less error and strong correlation factor. The proposed model will help in the prediction of the concrete strength by broadening neural network application in such problems and avoiding the computational burden on highly combative analytical physics based approaches.
KeywordsArtificial neural network (ANN) Machine learning Topologies Correlation factor
- 1.Aggarwal, R., Kumar, M., Sharma, R.K., Sharma, M.K.: Predicting compressive strength of concrete. Int. J. Appl. Sci. Eng. 13, 171–185 (2015)Google Scholar
- 2.Chopra, P., Sharma, R.K., Kumar, M.: Prediction of compressive strength of concrete using artificial neural network and genetic programming. Adv. Mater. Sci. Eng. (2016)Google Scholar
- 3.de Melo, V.V., Banzhaf, W.: Predicting high-performance concrete compressive strength using features constructed by Kaizen Programming. In: Brazilian Conference on Intelligent System (2015)Google Scholar
- 8.Singh, V.P., Kotiyal, Y.C.: Prediction of compressive strength using artificial neural network. Int. J. Civ. Environ. Eng. 7 (2013)Google Scholar