High-Performance Concrete Compressive Strength Prediction Based Weighted Support Vector Machines

Abstract

Concrete is the safest and sustainable construction material wh ich is most widely used in the world as it provides superior fire resistance, gains strength over time and gives an extremely long service life. Unfortunately high performance concrete is undoubtedly one of the most innovativ e materials in construction. Its Designing involves the process of selecting suitable ingredients of concrete (water, cement, fine and aggregates and a number of additives like mineral and chemical ad mixture) and determining their relat ive amounts with the objective of producing a high performance concrete of the required, strength, durability, and workab ility as economically as possible. Their proportions have a high influence on the final strength of the product. These relations do not seem to follow a mathematical formu la and yet their knowledge is crucial to optimize the quantities of raw materials used in the manufacture of high performance concrete. Therefore, it would be important to have a tool to numerically model such relationships, even before pro cessing. In this aspect the main purpose of this paper is to predict the compressive strength of the high performance concrete by using classification algorith ms. For building these models, training and testing using the available experimental results for 1030 specimens produced with 8 d ifferent mixture proportions are used. The result fro m this study suggests that weighted Support Vector Machines (wSVM) based models perform remarkably well in predict ing the compressive strength of the concrete mix.

Investigation on factors affecting early strength of high-performance concrete by Gaussian Process Regression

“…Concrete is a widely used material in building structures because of its many unique features [ 1 ]. Along with the development of construction engineering technologies, the development of new concrete technology also plays a significant role.…”

“…The concept of HPC is used to describe concrete that is manufactured with high quality, carefully selected raw material ingredients. The optimal proportion of the mixture is then mixed, poured, solidified, and deal with the highest technical standards [ 1 , 4 ]. Therefore, HPC has outstanding properties compared to conventional concrete such as high compression strength, very high tensile strength, high elastic modulus, sustainability and stability under adverse impacts of aggressive erosion of environment, and many other characteristics, very useful in building transport infrastructure [ 1 , 3 , 5 ].…”

“…The optimal proportion of the mixture is then mixed, poured, solidified, and deal with the highest technical standards [ 1 , 4 ]. Therefore, HPC has outstanding properties compared to conventional concrete such as high compression strength, very high tensile strength, high elastic modulus, sustainability and stability under adverse impacts of aggressive erosion of environment, and many other characteristics, very useful in building transport infrastructure [ 1 , 3 , 5 ]. As a result, HPC is increasingly used for highway applications, including new construction, repair and rehabilitation, bridges, tunnels, and high-rise buildings [ 6 ].…”

“…As a result, it is predicted that HPC would grow more popular in the next decades, mostly due to its high levels of sustainability and durability. However, the addition of these components in HPC makes calculating the HPC mixing ratio and HPC behavior model significantly more complicated than similar processes for conventional cement [ 1 ]. The HPC mixed design method was proposed by ACI [ 8 ], Aitcin [ 5 ], Islam Laskar and Talukdar [ 9 ] with the purpose of obtaining a combination of component materials and proportions respectively to create HPC with improved properties.…”

“…In another study, Melda Yucel and Ersin Namli [ 23 ] used eight different ML and Extreme learning machines (EML) techniques drafted from Random Forest (RF) and ANN algorithms to compare the predictive compressive performance of HPC. Mustapha and Mohamed [ 1 ] proposed a Support Vector Machine (SVM) model that performed well in forecasting concrete’s compressive strength. However, these works showed the relationship between concrete components that influence the compressive strength without establishing the relevance of each kind of material that affects compressive strength.…”