Deep learning-based concrete compressive strength prediction with modified resilient backpropagation training
by M. Adams Joe; J. Sahaya Ruben; M. Prem Anand; M. Anand
International Journal of Intelligent Engineering Informatics (IJIEI), Vol. 12, No. 3, 2024

Abstract: This article proposes a novel approach for predicting concrete compressive strength using deep learning techniques. It overcomes limitations of traditional methods like memory footprint, training time, and computational requirements for predicting concrete compressive strength. Currently various filter pruning techniques are used to compress models by removing irrelevant information, but they cannot decrease memory consumption due to their large parameters. So, the entropy-based filter pruning is suggested to reduce the complexity of the model by decreasing the parameters. Then for training the CNN model, the modified resilient backpropagation technique (MRPROP) is suggested, because the previous backpropagation techniques take more time for training and also it loss the accuracy. This MRPROP improve the efficiency and convergence of CNN training and also it updates the models weight. The proposed approach demonstrated superior performance in mean squared error, root mean squared error, loss function, and regression analysis, as per the experimental results.

Online publication date: Fri, 26-Jul-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Engineering Informatics (IJIEI):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com