Compressive strength prediction of fly ash-based geopolymer concrete via advanced machine learning techniques

Geopolymer concrete offers a favourable alternative to conventional Portland concrete due to its reduced embodied carbon dioxide (CO2) content. Engineering properties of geopolymer concrete, such as compressive strength, are commonly characterised based on experimental practices requiring large volumes of raw materials, time for sample preparation, and costly equipment. To help address this inefficiency, this study proposes machine learning-assisted numerical methods to predict compressive strength of fly ash-based geopolymer (FAGP) concrete. Methods assessed included artificial neural network (ANN), deep neural network (DNN), and deep residual network (ResNet), based on experimentally collected data. Performance of the proposed approaches were evaluated using various statistical measures including R-squared (R2), root mean square error (RMSE), and mean absolute percentage error (MAPE). Sensitivity analysis was carried out to identify effects of the following six input variables on the compressive strength of FAGP concrete: sodium hydroxide/sodium silicate ratio, fly ash/aggregate ratio, alkali activator/fly ash ratio, concentration of sodium hydroxide, curing time, and temperature. Fly ash/aggregate ratio was found to significantly affect compressive strength of FAGP concrete. Results obtained indicate that the proposed approaches offer reliable methods for FAGP design and optimisation. Of note was ResNet, which demonstrated the highest R2 and lowest RMSE and MAPE values.

Download Full-text

Machine Learning Techniques in Concrete Mix Design

Materials ◽

10.3390/ma12081256 ◽

2019 ◽

Vol 12 (8) ◽

pp. 1256 ◽

Cited By ~ 16

Author(s):

Patryk Ziolkowski ◽

Maciej Niedostatkiewicz

Keyword(s):

Neural Network ◽

Machine Learning ◽

Artificial Neural Network ◽

Compressive Strength ◽

State Of The Art ◽

Mix Design ◽

Machine Learning Techniques ◽

Concrete Mix Design ◽

Learning Techniques ◽

Concrete Mix

Concrete mix design is a complex and multistage process in which we try to find the best composition of ingredients to create good performing concrete. In contemporary literature, as well as in state-of-the-art corporate practice, there are some methods of concrete mix design, from which the most popular are methods derived from The Three Equation Method. One of the most important features of concrete is compressive strength, which determines the concrete class. Predictable compressive strength of concrete is essential for concrete structure utilisation and is the main feature of its safety and durability. Recently, machine learning is gaining significant attention and future predictions for this technology are even more promising. Data mining on large sets of data attracts attention since machine learning algorithms have achieved a level in which they can recognise patterns which are difficult to recognise by human cognitive skills. In our paper, we would like to utilise state-of-the-art achievements in machine learning techniques for concrete mix design. In our research, we prepared an extensive database of concrete recipes with the according destructive laboratory tests, which we used to feed the selected optimal architecture of an artificial neural network. We have translated the architecture of the artificial neural network into a mathematical equation that can be used in practical applications.

Download Full-text

Comparison of Machine Learning Techniques for the Prediction of Compressive Strength of Concrete

Advances in Civil Engineering ◽

10.1155/2018/5481705 ◽

2018 ◽

Vol 2018 ◽

pp. 1-9 ◽

Cited By ~ 9

Author(s):

Palika Chopra ◽

Rajendra Kumar Sharma ◽

Maneek Kumar ◽

Tanuj Chopra

Keyword(s):

Machine Learning ◽

Compressive Strength ◽

High Accuracy ◽

Machine Learning Techniques ◽

Coefficient Of Determination ◽

Software Environment ◽

R Software ◽

Learning Techniques ◽

Compressive Strength Of Concrete ◽

Indispensable Tool

A comparative analysis for the prediction of compressive strength of concrete at the ages of 28, 56, and 91 days has been carried out using machine learning techniques via “R” software environment. R is digging out a strong foothold in the statistical realm and is becoming an indispensable tool for researchers. The dataset has been generated under controlled laboratory conditions. Using R miner, the most widely used data mining techniques decision tree (DT) model, random forest (RF) model, and neural network (NN) model have been used and compared with the help of coefficient of determination (R2) and root-mean-square error (RMSE), and it is inferred that the NN model predicts with high accuracy for compressive strength of concrete.

Download Full-text