
Research Article
A Distinct Artificial Feed Forward Neural Network (AF2NN) Model for Predicting Compressive Strength of Geo-Polymer Concrete
@INPROCEEDINGS{10.1007/978-3-031-77075-3_10, author={D. Anantha Lakshmi and R. VelKennady}, title={A Distinct Artificial Feed Forward Neural Network (AF2NN) Model for Predicting Compressive Strength of Geo-Polymer Concrete}, proceedings={Cognitive Computing and Cyber Physical Systems. 5th EAI International Conference, IC4S 2024, Bhimavaram, India, April 5--7, 2024, Proceedings, Part-I}, proceedings_a={IC4S}, year={2025}, month={2}, keywords={Geo Polymer Concrete (GPC) Industrial Waste Management Bottom Ash Fine Aggregates Artificial Feed Forward Neural Network (AF2NN) and Machine Learning}, doi={10.1007/978-3-031-77075-3_10} }
- D. Anantha Lakshmi
R. VelKennady
Year: 2025
A Distinct Artificial Feed Forward Neural Network (AF2NN) Model for Predicting Compressive Strength of Geo-Polymer Concrete
IC4S
Springer
DOI: 10.1007/978-3-031-77075-3_10
Abstract
Geo-Polymer Concrete (GPC) utilizes leftover resources, offers a practical solution for sustainable building. Still, the development of blend designs and the application of GPC may be seriously hampered by the expensive and laborious fabrication and assessment procedures for GPC. The basic properties of GPC are influenced based on the type of precursor substance and concentration of hydroxide activators, and the liquid to solid ratio. Artificial Feed Forward Neural Network (AF2NN) can be a successful method for investigating and forecasting GPC properties in order to minimize both cost and time. In this study, an AF2NN model based on machine learning has been applied to determine the fly-ash-based GPC's compressive qualities, wherein bottom ash has been used in place of fine aggregates. Both internal laboratory-level GPC testing and data from the scientific literature are used as inputs. The GPC specimen parameters are used as features for input in the AF2NN algorithm to forecast compressive strength as the results, with the least amount of error possible. After a performance examination and comparison of such models using the metrics of mean squared error (MSE) and coefficient of correlation, an AF2NN model for precise prediction of the compressive property of fly-ash and bottom-ash based GPC is now being developed.