Concrete has become a major construction material all around the world with over ten billion tons consumed annually. One of the major issues to be kept monitored during manufacture of concrete is its initial setting time; that is to say, the time needed for the initiation of fresh concrete’s solidification. Information on the initial setting time of fresh concrete and/or cement paste are essential in construction scheduling, as well as in management of the haulage and placement of concrete. Conventional civil engineering laboratory tests for the determination of setting time are time and resources consuming in nature. This study aims to propose an intelligent model that will provide efficient prediction of setting time of cement pastes. An artificial Neural Network (ANN) model was proposed for the setting time predictions in this study; and its prediction performance was investigated systematically by using two training functions, under two different train:test data distributions together with five varying hidden neuron values. Setting time of cement pastes was predicted considering 12 input parameters. The results obtained indicates that the prediction accuracy of the employed ANN model is satisfactory; since it yielded remarkably high values of correlation coefficient and low mean square error such as 0.998 and 0.0003, respectively.
Civil Engineering; Artificial Neural Network; Cement pastes; Initial setting time; Feedforward backpropogation; Train:test data distributions; Hidden neurons.
Cite this paper
Pinar Akpinar, Mariya A. Abubakar. (2019) Intelligent Prediction of Initial Setting Time for Cement Pastes by Using Artificial Neural Network. International Journal of Control Systems and Robotics, 4, 37-45
Copyright © 2019 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0