Prediction of compressive strength of self-compacting concrete containing bottom ash using artificial neural networks
Introduction
Concrete is essentially a mixture of paste and aggregate. The paste, comprised of cement and water, binds the aggregate into a hard mass; the paste hardens because of the chemical reaction of the cement and water called hydration. In concrete mix design and quality control, the uniaxial compressive strength of concrete is considered as the most valuable property, which in turn is influenced by a number of factors. Various factors affect the concrete mix design like to designate a concrete as HPC, it should possesses, in addition to good strength, several other favourable qualities. The water/cement (w/c) ratio in the concrete is lower than normal concrete which requires special additives in the concrete, along with a superplasticizer to obtain good workability. Usually special cements are also required. The type of aggregate is important to obtain high strength. The grading of the aggregate influences the workability. The order in which the materials are mixed is also important for the workability of the concrete. Strength performance remains the most important property of structural concrete, from an engineering viewpoint. The strength of the concrete is determined by the characteristics of the mortar, coarse aggregate, and the interface. For the same quality mortar, different types of coarse aggregate with different shape, texture, mineralogy, and strength may result in different concrete strengths. The tests for compressive strength are generally carried out at about 7 or 28 days from the date of placing the concrete. The testing at 28-days is standard and therefore essential and at other ages can be carried out if necessary. If due to some experimental error in designing the mix, the test results fall short of required strength, the entire process of concrete design has to be repeated which may be a costly and time consuming. The same applies to all types of concrete, i.e. normal concrete, self-compacting concrete, ready mixed concrete, etc. It is well recognized that prediction of concrete strength is important in modern concrete constructions and in engineering judgments.
The successful development of self-compacting concrete (SCC), which is defined as the type of high performance concrete, filling all corners of formwork without vibration, and having good deformability, high segregation resistance and no blocking around reinforcement, must ensure a good balance between deformability and stability. It requires manipulation of several mixture variables to ensure acceptable flowable behaviour and proper mechanical properties. Also, absence of theoretical relationships between mixture proportioning and measured engineering properties of SCC makes it more complex.
Within last decade, researchers have explored the potential of artificial neural networks (ANNs), a nonlinear modelling approach, in predicting the compressive strength of the concrete, due to its ability to learn input–output relation for any complex problem in an efficient way. Artificial neural network (ANN) does not need specific equation form. Instead, it only needs sufficient input–output data. It can also continuously retrain new data to adapt new data conveniently. ANNs have been investigated to deal with the problems involving incomplete or imprecise information. The capability of artificial neural network to act as universal function approximators has been traditionally used to model problems in which the relation between dependent and independent variables is not clearly understood. When the number of components increases, the relationship between variables becomes usually complex and the use of a nonlinear modelling approach is required. In recent years, ANNs have been applied to many civil engineering applications with some degree of success. ANNs have been applied to geotechnical problem like prediction of settlement of shallow foundations [1]. Researchers have also used ANN in structural engineering [2]. Some researchers have recently proposed a new method of mix design and prediction of concrete strength using neural network [3], [4]. Also, several works were reported on the use of neural network based modelling approach in predicting the concrete strength [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. Some attempts have been made to describe the compressive strength properties using traditional regression analysis tools and statistical models [15], [16], [17]. However, the development of neural network models for predicting the strength of SCC has not been fully investigated. Thus, it was required to develop some suitable methodology to estimate the compressive strength of self-compacting concrete based on its constituents at the time of design.
Therefore, the objective of the present study was to examine the potential of ANN for predicting the 28-day compressive strength of SCC mixtures, with data obtained from literature. These models were further applied to prediction of strength at 7, 28, 90 and 365 days to the data obtained experimentally. The complex relationship between mixture proportions and engineering properties of SCC was generated based on data obtained experimentally. It was observed that the neural network could effectively predict compressive strength in spite of intricate data and could be used as a tool to support decision making, by improving the efficiency of the process.
Section snippets
Artificial neural network
Artificial neural network exhibit analogies to the way arrays of neuron function in biological learning and memory. The fundamental building blocks are units (‘nodes’) comparable to neurons, weighted connections that can be likened to synapses in biological systems. Nodes are simple information processing elements. The number of nodes in ANNs and the connection patterns of the nodes can vary. The total number of nodes in the input and output layers coincide with the number of input and output
Database
The model’s success in predicting the behaviour of SCC mixtures depends on comprehensiveness of the training data. Availability of large variety of experimental data was required to develop the relationship between the mixture variables of SCC and its measured properties. The basic parameters considered in this study were cement content, sand content, coarse aggregate content, fly ash content, water-to-powder ratio and superplasticizer dosage. A database of 80 mixes from the literature was
Training and testing of neural networks
Training means to present the network with the experimental data and have it learn, or modify its weights, such that it correctly reproduces the strength behaviour of mix. However, training the network successfully requires many choices and training experiences. After a number of trials, the values of the network parameters considered by this study are as given in Table 4.
Results and analysis
The acceptance/rejection of the model developed are determined by its ability to predict the strength of SCC. Also, a successfully trained model is characterized by its ability to predict strength values for the data it was trained on. A 10-fold cross validation is used to predict the strength for the data set used in this study. The cross validation is the method of accuracy of a classification or regression model. The input data set is divided into several parts (a number defined by the
Conclusions
Artificial neural networks are viable computational models for a wide variety of problems including prediction problems. The neural network can be used for a particular problem when deviation in the available data is expected and accepted and also when a defined methodology is not available as in the case of present study.
- •
This study presents the application of neural network to predict the compressive strength of SCC based on several parameters. SCC is different from conventional concrete such
References (24)
- et al.
Concrete strength prediction by means of neural network
Constr Build Mater
(1997) Modeling of strength of high-performance concrete using artificial neural networks
Cem Concr Res
(1998)- et al.
Neural network prediction of unconfined compressive strength of coal fly ash–cement mixtures
Cem Concr Res
(2003) - et al.
Neural networks for predicting properties of concretes with admixtures
Constr Build Mater
(2001) - et al.
An optimal neural network and concrete strength modeling
J Adv Eng Softw
(2002) Prediction of concrete strength using artificial neural networks
Eng Struct
(2003)Medium strength self-compacting concrete containing fly ash: modelling using factorial experimental plans
Cem Concr Res
(2004)- et al.
Self-compacting concrete incorporating high volumes of class F fly ash preliminary results
Cem Concr Res
(2001) - et al.
Parameter estimation in ground-water hydrology using artificial neural networks
J Comput Civil Eng
(2003) Simulating structural analysis with neural network
J Comput Civil Eng
(1994)
HPC strength prediction using artificial neural network
J Comput Civil Eng
Application of neural networks for proportioning of concrete mixes
ACI Mater J
Cited by (210)
Supplementary cementitious materials in blended cement concrete: Advancements in predicting compressive strength through machine learning
2024, Materials Today CommunicationsTo determine the compressive strength of self-compacting recycled aggregate concrete using artificial neural network (ANN)
2024, Ain Shams Engineering JournalUsing artificial intelligence methods to predict the compressive strength of concrete containing sugarcane bagasse ash
2023, Construction and Building MaterialsPrediction of compressive strength of cementitious grouts for semi-flexible pavement application using machine learning approach
2023, Case Studies in Construction Materials