Affiliation:
1. Adichunchanagiri Institute of Technology,Chikkamagluru ,Karnataka
Abstract
Abstract
The building construction industry is continuously exploring innovative methods to advance the properties of concrete and ensure its durability and strength. Self-compacting concrete (SCC) has gained considerable attention due to its ability to flow and compact under its weight, resulting in reduced labour requirements and improved structural integrity. In present investigate the prediction of critical properties of self-compacting concrete using an artificial neural network approach (ANN), and the compressive and split tensile strength of SCC were predicted in the current study. The data from seven sets of a total of concrete samples and the experimentally generated data for self-compacting concrete were used in this study. ANNs are regarded as nonlinear statistical data modeling tools for identifying patterns or modeling complex interactions between inputs and outputs. Eight input parameters are employed in the current ANN model to forecast the compressive and split tensile strength of self-compacting concrete. Cement, M-sand, coarse aggregate, GGBS, Micro silica (MS), water, Super Plasticizer (SP), and Age (days) are among these materials in variable proportions. The concrete's compressive and split tensile strengths are both included in the single output parameter. The study covered the significance of various input parameters for utilizing neural networks to predict strengths at various ages. The compressive strength of SCC predicted by ANN is a perfect match with experimental data and extremely low Mean Square Errors (M.S.E.). Additionally, the ANN model's effectiveness is higher than that of multiple regression analysis (MRA).The ANN model has greater promise than the MRA model for creating an ideal mix percentage for predicting the compressive strength and split tensile strength of concrete without suffering a significant loss in material or time.
Publisher
Research Square Platform LLC