Computational & Technology Resources
an online resource for computational,
engineering & technology publications
engineering & technology publications
Civil-Comp Conferences
ISSN 2753-3239
ISSN 2753-3239
CCC: 5
PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON SOFT COMPUTING, MACHINE LEARNING AND OPTIMISATION IN CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING
PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON SOFT COMPUTING, MACHINE LEARNING AND OPTIMISATION IN CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING
Edited by: P. Iványi, J. Logo and B.H.V. Topping
Paper 3.2
A Comparison of Neural Networks and Random Forest for predicting the subsurface tensile strength of cementitious composites containing waste materials
S. Czarnecki and M. Moj
Cathedral of Materials Engineering and Construction Processes, Wroclaw University of Science and Technology, Poland
Full Bibliographic Reference for this paper
S. Czarnecki, M. Moj, "A Comparison of Neural Networks and
Random Forest for predicting the subsurface
tensile strength of cementitious composites
containing waste materials", in P. Iványi, J. Logo, B.H.V. Topping, (Editors), "Proceedings of the Sixth International Conference on
Soft Computing, Machine Learning and Optimisation in
Civil, Structural and Environmental Engineering", Civil-Comp Press, Edinburgh, UK,
Online volume: CCC 5, Paper 3.2, 2023, doi:10.4203/ccc.5.3.2
Keywords: neural networks, random forest, cementitious composites, waste materials, floors, subsurface tensile strength.
Abstract
In this article the accurate model of predicting the eco-friendly mortar’s subsurface
tensile strength is presented. These eco-friendly mortars were made by substituting in
the mortars the mass cement by waste materials: fly ash, granite flour and ground
granulated blast furnace slag. These mortars were tested using ultrasonic pulse
velocity method and based on the results of these tests the dataset were built.
Estimation of the subsurface tensile method were done using hybrid combination of
ultrasonic pulse velocity method and soft computing techniques. The accuracy of this
method were proved by the very high values of the coefficient of determination around
0.9 and very low values of the mean average percentage error around 5%. These
method might be suitable for use in existing structures where experimental destructive
test are problematic.
download the full-text of this paper (PDF, 9 pages, 595 Kb)
go to the previous paper
go to the next paper
return to the table of contents
return to the volume description