|About this Abstract
||Materials Science & Technology 2019
||Ceramics and Glasses Simulations and Machine Learning
||Prediction of Compressive Strength and Modulus of Elasticity of Concrete Using Machine Learning Models
||Taihao Han, Aditya Kumar
|On-Site Speaker (Planned)
Machine learning (ML) techniques are increasingly used to simulate and predict quantitative composition-property relationships of concrete. This poster presents a comprehensive study using multilayer perceptron artificial neural network (MLP-ANN), support vector machine (SVM), and random forest (RF) algorithms to predict the different properties of cement system. The RF constructs multiple decision trees to vote for the final predicted results. The SVM utilizes the hyperplanes of data sets in a high or infinite dimensional space to determine the relationship between inputs and outputs. The MLP-ANN consists of several neuron layers- an input layer, an output layer, and one more hidden layers- to compute the final prediction. The accuracy of predicted capabilities of these techniques is investigated by using concrete data from different literatures and our lab experimental results. The metrics used for evaluation of prediction accuracy of these algorithms included five different statistical parameters and a composite performance index (CPI).