Predicting High-Performance Concrete Compressive Strength with Machine Learning

dc.contributorNakshatrala, Kalyana Babu
dc.contributor.authorTran, Tommy
dc.description.abstractRecent technological advances in computer hardware and software enable complex computations and allow handling intensive data sets. In particular, the development of machine learning tools has enabled us to extract science and patterns from larger heterogeneous data. Herein, we utilize machine learning to address problems in civil engineering—predict the compressive strength of high-performance concrete based on the composition of the ingredients. Although there are prior efforts on using Artificial Neural Networks to predict the compressive strength, herein we employ more advanced and modern machine learning techniques and algorithms—Keras, Tensorflow, and scikit-learn Python libraries—to make predictions. Specifically, we will use supervised machine learning methods—deep neural networks and support vector machines—for modeling the performance of HPC. The dataset used contains 1,030 batches of HPC and was obtained from the University of California, Irvine Machine Learning Repository. Using numerical experiments, we show that machine learning is an effective tool for predicting concrete properties, and neural networks of the multilayer perceptron type giving higher predictability than support vector machines for this particular dataset.
dc.description.departmentCivil and Environmental Engineering, Department of
dc.description.departmentHonors College
dc.relation.ispartofSummer Undergraduate Research Fellowship
dc.rightsThe author of this work is the copyright owner. UH Libraries and the Texas Digital Library have their permission to store and provide access to this work. Further transmission, reproduction, or presentation of this work is prohibited except with permission of the author(s).
dc.titlePredicting High-Performance Concrete Compressive Strength with Machine Learning


Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
4.69 MB
Adobe Portable Document Format