Prediction Of Concrete Compressive Strength Research Articles

Prediction of Compressive Strength of Concrete Using Artificial Neural Network and Genetic Programming

An effort has been made to develop concrete compressive strength prediction models with the help of two emerging data mining techniques, namely, Artificial Neural Networks (ANNs) and Genetic Programming (GP). The data for analysis and model development was collected at 28-, 56-, and 91-day curing periods through experiments conducted in the laboratory under standard controlled conditions. The developed models have also been tested on in situ concrete data taken from literature. A comparison of the prediction results obtained using both the models is presented and it can be inferred that the ANN model with the training function Levenberg-Marquardt (LM) for the prediction of concrete compressive strength is the best prediction tool.

Advancing concrete strength prediction using non-destructive testing: Development and verification of a generalizable model

Accurate prediction of concrete compressive strength is imperative for investigating the in-situ concrete quality. To avoid destructive testing, developing reliable predictive models for concrete compressive strength using nondestructive tests (NDTs) is an active area of research. However, many of the developed models are dependent on calibration and/or concrete past history (e.g. mixture proportion, curing history, concrete mechanical properties, etc.), which reduces their utility for in-situ predictions.This paper develops predictive models for concrete compressive strength that are independent of concrete past history. To this end, ultrasonic pulse velocity (UPV) and rebound hammer (RH) tests were performed on 84 concrete cylindrical samples. Next, compressive strengths were determined using destructive testing on these cylinders, and predictive models were developed using NDT results. Furthermore, to ensure generalizability to new data, all models were tested on independent data collected from six different research papers. The results support combined usage of UPV and RH in a quadratic polynomial model structure. Therefore, the final model was proposed based on combining models from a threefold cross-validation of the experimental data. This model predicted the independent data with very good accuracy. Finally, a concrete quality classification table using combined RH and UPV is proposed based on a variant of machine learning k-means clustering algorithm.

Predicting Compressive Strength of Concrete

This paper deals with development of regression models for prediction of compressive strength of concrete. The compressive strength of concrete is predicted using four variables, namely, water-binder ratio, fine aggregate-binder ratio, coarse aggregate-binder ratio and binder content. Linear regression models have been developed for variations in fly ash replacements (0 and 15 percent), Zones of aggregates (A, B and C) and curing ages (28, 56 and 91 days). An effort has also been made to modify the linear models using a two step approach. First step is to develop quadratic models (termed as full models) by identifying the suitable combinations of the four variables described above. The second step is to select the best minimal subset of the predictors in full models using Mallow's Cp statistic. The proposed quadratic regression models yielded coefficient of determination R 2 ≥ 0.99 in almost every case except for concrete with Zones A and B of aggregates without fly ash for 91 days curing period and for concrete with Zone C of aggregates without fly ash for 56 days curing period. Concrete is a strong building material composed of sand, gravel, cement and water. Additional admixtures are also added sometimes to enhance certain properties of fresh or hardened concrete. In the construction industry, among the various properties of the concrete determination of compressive strength has received a large amount of attention because concrete is primarily meant to withstand compressive stresses. The compressive strength of concrete is controlled by proportioning of cement, coarse and fine aggregates, water and various admixtures. Prediction of concrete compressive strength has been an active area of research and a considerable number of studies have been carried out in this direction. Many attempts have been made to develop a suitable mathematical model which is capable of predicting compressive strength of concrete at various ages with acceptable accuracy. Multiple regression models have been used by researchers to improve accuracy of predictions of concrete compressive strength. Nipatsat and Tangtermsirikul (1) developed equations for estimating compressive strength of concrete containing fly ash with curing ages from 3 days up to 1 year. Kazberuk and Lelusz (2) proposed regression models to predict compressive strength of concrete with fly ash replacement percentages up to 30%. Namyong et al. (3) proposed the regression equations for predicting compressive strength of in-situ concrete. They have selected

Assessment of concrete compressive strength prediction models

Theoretical and phenomenological models for predicting the compressive strength of concrete proposed in literature have been reviewed. The applicability and accuracy of these models in predicting the compressive strength of concrete at 28 days and at different ages were investigated using experimental data reported in the literature. Assessment of the models using visual display and numerical methods have revealed that the Average Paste Thickness (APT) compressive strength model, which accounts for the type of cement, cement degree of hydration, mixture proportions, aggregate proportions, gradations, packing density, and air content, provides the highest predictability at 28 days and at different ages. The assessment revealed that the majority of models give acceptable predictions because strength is mostly affected by water-to-cement ratio (w/c) in comparison to aggregates' properties and gradation.

Prediction of Concrete Compressive Strength by Evolutionary Artificial Neural Networks

Compressive strength of concrete has been predicted using evolutionary artificial neural networks (EANNs) as a combination of artificial neural network (ANN) and evolutionary search procedures, such as genetic algorithms (GA). In this paper for purpose of constructing models samples of cylindrical concrete parts with different characteristics have been used with 173 experimental data patterns. Water-cement ratio, maximum sand size, amount of gravel, cement, 3/4 sand, 3/8 sand, and coefficient of soft sand parameters were considered as inputs; and using the ANN models, the compressive strength of concrete is calculated. Moreover, using GA, the number of layers and nodes and weights are optimized in ANN models. In order to evaluate the accuracy of the model, the optimized ANN model is compared with the multiple linear regression (MLR) model. The results of simulation verify that the recommended ANN model enjoys more flexibility, capability, and accuracy in predicting the compressive strength of concrete.

An ERAFNN Prediction of Concrete Compressive Strength Based on Physical Properties of Electric Arc Furnace Oxidizing Slag

In this study, exact radial basis function neural network (ERBFNN) was used to predict the concrete compressive strength based on physical properties of electric arc furnace oxidizing slag. The mean absolute percentage error (MAPE) was used to evaluate the predicting performance. The results indicated the minimum MAPE of 0.08 % and 5.28 % could be achieved when training and predicting, respectively. According to the results, it revealed that ERBFNN was an efficiently tool for providing information.

Prediction of concrete compressive strength considering humidity and temperature in the construction of nuclear power plants

This study proposes a method for predicting compressive strength developments in the early ages of concretes used in the construction of nuclear power plants. Three representative mixes with strengths of 6000psi (41.4MPa), 4500psi (31.0MPa), and 4000psi (27.6MPa) were selected and tested under various curing conditions; the temperature ranged from 10 to 40°C, and the relative humidity from 40 to 100%. In order to consider not only the effect of the temperature but also that of humidity, an existing model, i.e. the humidity-adjusted maturity function, was adopted and the parameters used in the function were determined from the test results. A series of tests were also performed in the curing condition of a variable temperature and constant humidity, and a comparison between the measured and predicted strengths were made for the verification.

Prediction of concrete compressive strength due to long term sulfate attack using neural network

This work was divided into two phases. Phase one included the validation of neural network to predict mortar and concrete properties due to sulfate attack. These properties were expansion, weight loss, and compressive strength loss. Assessment of concrete compressive strength up to 200years due to sulfate attack was considered in phase two. The neural network model showed high validity on predicting compressive strength, expansion and weight loss due to sulfate attack. Design charts were constructed to predict concrete compressive strength loss. The inputs of these charts were cement content, water cement ratio, C3A content, and sulfate concentration. These charts can be used easily to predict the compressive strength loss after any certain age and sulfate concentration for different concrete compositions.

Prediction of Concrete Compressive Strength by Means of Combined Non-Destructive Testing

The problem of estimating mechanical properties of buildings materials is a major issue in civil engineering, especially when dealing with existing structures. In such occasion a valuable help is given by Non Destructive Testings, which allow materials characteristics to be evaluated in a non-invasive way. This paper analyses the efficacy of the SonReb method - a non destructive technique which statistically combines the results of Ultrasonic and Rebound Hammer Testing in predicting concrete compressive strength. The SonReb method has been applied on concrete specimens later on exposed to compressive strength test. The effectiveness of several empirical formulas expressing SonReb results is illustrated and discussed.

Prediction of concrete compressive strength by combined non-destructive methods

The compressive strength of concrete fc is one of most important mechanical parameter in the performance assessment of existing reinforced concrete (RC) structures. Many non-destructive testing methods have been developed to estimate the in situ value of fc. In particular, the combination of rebound hammer and ultrasonic pulse velocity (UPV) tests, known as SonReb method, is frequently used. In this paper an experimental research has being carried out involving both destructive and non-destructive testing methods for the evaluation of structural performance of existing RC structures. The aim of this study is to verify the reliability of predicting fc by some formulas available in literature involving measured parameters such rebound index and UPV.

Performance Comparison of SVM and ANN in Predicting Compressive Strength of Concrete

Concrete compressive strength prediction is very important in structure and building design, particularly in specifying the quality and measuring performance of concrete as well as determination of its mix proportion. The conventional method of determining the strength of concrete is complicated and time consuming hence artificial neural network (ANN) is widely proposed in lieu of this method. However, ANN is an unstable predictor due to the presence of local minima in its optimization objective. Hence, in this paper we have studied the performance of support vector machine (SVM), a stable and robust learning algorithm, in concrete strength prediction and compare the result to that of ANN. It is found that SVM displayed a slightly better performance compared to ANN and is highly stable.

High-performance concrete compressive strength prediction using Genetic Weighted Pyramid Operation Tree (GWPOT)

This study uses the Genetic Weighted Pyramid Operation Tree (GWPOT) to build a model to solve the problem of predicting high-performance concrete compressive strength. GWPOT is a new improvement of the genetic operation tree that consists of the Genetic Algorithm, Weighted Operation Structure, and Pyramid Operation Tree. The developed model obtained better results in benchmark tests against several widely used artificial intelligence (AI) models, including the Artificial Neural Network (ANN), Support Vector Machine (SVM), and Evolutionary Support Vector Machine Inference Model (ESIM). Further, unlike competitor models that use “black-box” techniques, the proposed GWPOT model generates explicit formulas, which provide important advantages in practical application.

Applying Support Vector Machines in Rebound Hammer Test

There are several nondestructive testing techniques available to test the compressive strength of the concrete and the Rebound Hammer Test is among one of the fast and economical methods. Nevertheless, it is found that the prediction results from Rebound Hammer Test are not satisfying (over 20% mean absolute percentage error). In view of this, this research intends to develop a concrete compressive strength prediction model for the SilverSchmidt test hammer, using data collected from 838 lab tests. The Q-values yield from the concrete test hammer SilverSchmidt is set as the input variable and the concrete compressive strength is set as the output variable for the prediction model. For the non-linear relationships, artificial intelligence technique, Support Vector Machines (SVMs), are adopted to develop the prediction models. The results show that the mean absolute percentage errors for SVMs prediction model, 6.76%, improves a lot when comparing to SilverSchmidt predictions. It is recommended that the artificial intelligence prediction models can be applied in the SilverSchmidt tests to improve the prediction accuracy.

Prediction of concrete compressive strength: Research on hybrid models genetic based algorithms and ANFIS

The management of concrete quality is an important task of concrete industry. This paper researched on the structured and unstructured factors which affect the concrete quality. Compressive strength of concrete is one of the most essential qualities of concrete, conventional regression models to predict the concrete strength could not achieve an expected result due to the unstructured factors. For this reason, two hybrid models were proposed in this paper, one was the genetic based algorithm the other was the adaptive network-based fuzzy inference system (ANFIS). For the genetic based algorithm, genetic algorithm (GA) was applied to optimize the weights and thresholds of back-propagation artificial neural network (BP-ANN). For the ANFIS model, two building methods were explored. By adopting these predicting methods, considerable cost and time-consuming laboratory tests could be saved. The result showed that both of these two hybrid models have good performance in desirable accuracy and applicability in practical production, endowing them high potential to substitute the conventional regression models in real engineering practice.

Predictive modeling of concrete compressive strength based on cement strength class

In the current study, a method for concrete compressive strength prediction (based on cement strength class), incorporated in a software package developed by the authors for the estimation of concrete service life under harmful environments, is presented and validated. Prediction of concrete compressive strength, prior to real experimentation, can be a very useful tool for a first mix screening. Given the fact that lower limitations in strength have been set in standards, to attain a minimum of service life, a strength approach is a necessity. Furthermore, considering the number of theoretical attempts on strength predictions so far, it can be seen that although they lack widespread accepted validity, certain empirical expressions are still widely used. The method elaborated in this study, it offers a simple and accurate, compressive strength estimation, in very good agreement with experimental results. A modified version of the Feret\'s formula is used, since it contains only one adjustable parameter, predicted by knowing the cement strength class. The approach presented in this study can be applied on any cement type, including active additions (fly ash, silica fume) and age.

Two-level and hybrid ensembles of decision trees for high performance concrete compressive strength prediction

Accurate prediction of high performance concrete (HPC) compressive strength is very important issue. In the last decade, a variety of modeling approaches have been developed and applied to predict HPC compressive strength from a wide range of variables, with varying success. The selection, application and comparison of decent modeling methods remain therefore a crucial task, subject to ongoing researches and debates. This study proposes three different ensemble approaches: (i) single ensembles of decision trees (DT) (ii) two-level ensemble approach which employs same ensemble learning method twice in building ensemble models (iii) hybrid ensemble approach which is an integration of attribute-base ensemble method (random sub-spaces RS) and instance-base ensemble methods (bagging Bag, stochastic gradient boosting GB). A decision tree is used as the base learner of ensembles and its results are benchmarked to proposed ensemble models. The obtained results show that the proposed ensemble models could noticeably advance the prediction accuracy of the single DT model and for determining average determination of correlation, the best models for HPC compressive strength forecasting are GB–RS DT, RS–GB DT and GB–GB DT among the eleven proposed predictive models, respectively. The obtained results show that the proposed ensemble models could noticeably advance the prediction accuracy of the single DT model and for determining determination of correlation (R2max), the best models for HPC compressive strength forecasting are GB–RS DT (R2=0.9520), GB–GB DT (R2=0.9456) and Bag–Bag DT (R2=0.9368) among the eleven proposed predictive models, respectively.

Prediction of concrete compressive strength in buildings that would be reinforced by fuzzy logic

In this study, core samples were taken from column or reinforced wall in order to make reinforcement or restoration. The effect of the element height and volume of voids of these samples to the compressive strength were investigated. C25 type ready to use concrete was used and a column without steel was produced. During the production process, 15 cm reference cube samples were taken from the fresh concrete. After 28 days, 8 cores from reference cubes and 128 samples from column were taken and their compressive strengths were determined. As a result, the average compressive strength of the reference core samples was 36.95 Mpa and according to the height compressive strengths of other samples was ranging between 37.3 and 43.0 Mpa. The ratio of the increasing of the compressive strength changes between 0.95 and 16.37% according to the reference sample. Based on the experimental results, fuzzy Logic method was used for modelling of the experimental results. The paper demonstrates that fuzzy logic can be successfully used in modelling the compressive strength of concrete for different height and volume of voids. This model enables us to easily and reliably estimate the compressive strength of concrete. Key words: Concrete, reinforcement, compressive strength, prediction model, fuzzy logic.

Prediction of Early Age Normal Concrete Compressive Strength Based on Dynamic Shear Modulus Measurements

In this research, the relationship between the compressive strength, fc, and the dynamic shear modulus, Gd, of normal concrete at an early age was studied. To investigate the correlation between fc and Gd at an early age, different types of mixtures, including mortar and concrete, were prepared, and the corresponding fc and Gd values were measured every 12 h after initial mixing up to 72 h after casting. The influences of hydration age, water-to-cement (w/c) ratio, curing temperature, aggregate volume content, and maximum aggregate size on the Gd−fc relationship of concrete were studied. The Gd−fc relationship was then mathematically modeled by using multivariable power laws. The developed model is reasonably accurate to predict the early age compressive strength of concrete with variations in hydration age, aggregate content, and sizes. This type of model can be used directly in the field for the estimation of concrete strength when nondestructive testing (NDT) techniques are employed.

Comparative study on fuzzy inference systems for prediction of concrete compressive strength

The aim of this comparative study is to evaluate the effects of different methods, used for ''aggregation'' and ''defuzzification'', on the output. To reach this goal, six fuzzy inference systems (FIS-1, FIS-2, FIS-3, FIS-4, FIS-5 and FIS-6) have been designed (with the same rule bases) with different methods used for aggregation, and overlapping between the fuzzy sets. The idea of this system is based on the UCI dataset. To design the systems, some of the input fields of UCI dataset have been replaced with other important fields that made system more applicable and suitable. All of these designed systems have the same input fields such as: Water/cement ratio, slump, maximum size of aggregate, coarse aggregate, fine aggregate and age (day). The output field of all systems measures the compressive strength of concrete. These three differences in 401 laboratory samples have caused the average error of predicted compressive strength, that is, 6.43% FIS-1, 6.64% FIS-2, 6.48% FIS-3, 5.56% FIS-4, 4.73% FIS-5 and 5.07% FIS-6. The experimental results reveal that the methods of sum and Centroid (used in the FIS-5) show the best results (among other methods) for the aggregation and the defuzzification, respectively. Key words: Aggregation step, defuzzification step, concrete compressive strength (CCS), fuzzy inference system (FIS), W/C ratio, slump, maximum size of aggregate, coarse aggregate, fine aggregate, age.

Concrete Compressive Strength Prediction Using Rebound Method with Artificial Neural Network

Concrete is a mixture of the cementing material, aggregate and water in a certain proportion and is the most main materials of the civil engineering materials. It is difficult to make modeling for a highly complex material. The concrete rebound value with wide randomness is a dependent variable, while the compressive strength value with narrow randomness is an independent variable. This paper aimed to show possible applicability of artificial neural networks (ANN) to predict the compressive strength. Back propagation neural networks (BPNN) model is constructed trained and tested using the available test data of 108 different concrete specimens. The data of input parameters used in BPNN model were covered the ratio of water to cement, fine aggregate ratio, coarse aggregates, mean value of test area of rebound method measurement. The mean absolute percentage error was less then 10.19% for compressive strength. The results showed that ANNs was good at as a feasible tool for predicting compressive strength.