Comparison of artificial neural network and fuzzy logic models for prediction of long-term compressive strength of silica fume concrete

Abstract

In this study, an artificial neural network (ANN) and fuzzy logic (FL) study were developed to predict the compressive strength of silica fume concrete. A data set of a laboratory work, in which a total of 48 concretes were produced, was utilized in the ANNs and FL study. The concrete mixture parameters were four different water–cement ratios, three different cement dosages and three partial silica fume replacement ratios. Compressive strength of moist cured specimens was measured at five different ages. The obtained results with the experimental methods were compared with ANN and FL results. The results showed that ANN and FL can be alternative approaches for the predicting of compressive strength of silica fume concrete.

Introduction

It is reported that, energy saving, high cost of cement and pressure of environmental lobbyists strengthen the use of industrial waste materials, such as fly ash, silica fume, ground granulated blast-furnace slag and rice husk ash to blend with cement and concrete [1]. Among these materials, silica fume is a by-product of silicon metal or siliconalloymetal factories. Although the silica fume was a waste of industrial materials, it became the most valuable by-product between the pozzolanic materials due to its very active and high pozzolanic property. Currently, it is widely used in concrete or cement as an admixture [2].

The use of silica fume reduces the workability of fresh concrete or mortar due to its very high specific surface area, however, it improves many of the properties of hardened concrete or mortar [3], [4], [5]. Ozturan [6], Mehta [7], Toutanji and Bayasi [8] and Massazza [9] reported that the behavior of silica fume in concrete is physiochemical. The physical phase is in the refinement of the void system of cement paste and particularly the transition zone. The chemical phase consists of the pozzolanic reaction that transforms the weak calcium hydroxide crystals into the strong calcium silicate hydrate gel. The results of these actions of silica fume provide significant improvements in compressive and flexural strengths along with improvement in durability. The use of silica fume in concrete increases the resistance of concrete to acid and sulfate attack; it improves durability of concrete by reducing porosity and permeability of cement paste matrix [2], [10], [11], [12]. It makes the concrete more resistant to abrasive forces, and reduces the expansion generated by alkali-aggregate reaction [13].

On the other hand, compressive strength of a concrete is a major and important mechanical property, which is generally obtained by crushing the concrete specimens after a standard curing of 28 days. Conventional methods of predicting 28-day compressive strength of concrete are generally based on either Abrams water–cement ratio rule or maturity concept of concrete [14].

Lee [15] reported that, for many years, researchers have proposed various methods, which are generally based on maturity concept of concrete, to predict the concrete compressive strength.

Several studies independently have shown that concrete strength development is determined not only by the w/c ratio, but that it is also influenced by the content of other ingredients. Therefore, although experimental data have shown the practical acceptability of this rule within wide limits, a few deviations have been reported. The current empirical equations presented in the codes and standards for estimating compressive strength are based on tests of concrete without supplementary cementitious materials. The validity of these relationships for concrete with supplementary cementitious materials (silica fume, fly ash, blast-furnace slag, etc.) should be investigated. The more we know about the concrete composition versus strength relationship, the better we can understand the nature of concrete and how to optimize the concrete mixture [14], [15], [16].

Over the last two decades, a different modeling method based on fuzzy logic (FL) or neural networks (NNs) has become popular and has been used by many researchers for a variety of engineering applications. NNs are a family of massively parallel architectures that solve difficult problems via the cooperation of highly interconnected but simple computing elements (or artificial neurons). Basically, the processing elements of a neural network are similar to the neurons in the brain, which consist of many simple computational elements arranged in layers [17]. Fuzzy control theory can be applied on linear and nonlinear systems. It does not need to handle the tedious mathematical models of controlled body. It needs only to set a simple controlling method based on engineering experience. Therefore, it is particularly useful in complicated structural control system. The compressive strength can be calculated using the models built with FL and NNs. It is convenient and easy to use these models for numerical experiments to review the effects of each variable on the mix proportions [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28].

The aim of this paper is to construct a NN and FL model to predict the compressive strength of silica fume concrete using concrete ingredients and time. The NN model with one hidden layer was constructed, training, testing and validation stages have been performed using the available test data of 240 different concrete mix-designs used. The NN and FL models had six input parameters and one output parameter. The obtained results from compressive strength tests were compared with predicted results.