Concrete compressive strength prediction using ultrasonic pulse velocity through artificial neural networks

Deep learning based concrete compressive strength prediction model with hybrid meta-heuristic approach

Experimental and machine learning approaches to investigate the application of sugarcane bagasse ash as a partial replacement of fine aggregate for concrete production

With the improvement of the ridge regression and EM algorithm, the application of it in the prediction and analysis of concrete compressive strength has attracted more and more attention, and with the introduction of various new methods and theories, the EM algorithm and ridge regression parameters the application of estimation in algorithms is becoming more and more extensive. The purpose of this paper is to study the prediction of concrete compressive strength based on improved EM algorithm and ridge regression algorithm. In this paper, the principles of EM algorithm and ridge regression algorithm are described, the improved EM algorithm and ridge regression algorithm are proposed, and the prediction effect of the algorithm is tested. The experimental results show that when the real value of concrete compressive strength is 37.9mpa, the predicted value before algorithm optimization is 37.42mpa and the predicted value after algorithm optimization is 39.65mpa; Before optimization, the maximum absolute error reached 1.36mpa, and after optimization, the maximum absolute error was 0.37mpa. It can be seen that the improved EM algorithm and ridge regression algorithm have higher accuracy in predicting concrete compressive strength, and the error between them and the real value is the smallest.

This paper aims at evaluating the compressive strength of structural sand lightweight aggregate concrete (LWC) by the non-destructive ultrasonic pulse velocity method. To this end, a comprehensive experimental study was carried out involving more than 70 different compositions tested between 3 and 180 days with mean compressive strengths ranging from about 30 to 80 MPa and density classes from D1.6 to D2.0. The influence of several factors on the relation between the ultrasonic pulse velocity (UPV) and compressive strength was examined, including the type and binder content, type and volume of aggregates. It is found that lightweight and normal weight concretes (NWC) are affected differently by mix design parameters. NWC is more affected by the volume of aggregate and the dispersion effect caused by concrete heterogeneity tends to be lower in LWC. The prediction of the concrete's compressive strength by means of the non-destructive UPV test is analysed. Based on the dependence of the UPV on the density and elasticity of concrete, a simplified expression is proposed to estimate the compressive strength, regardless of the type of concrete and its composition. High correlation coefficients were obtained taking into account several results for different types of aggregates and concrete compositions.

Soft computing techniques for the prediction of concrete compressive strength using Non-Destructive tests

This study aims to utilize fly ash and waste PET bottles for producing more sustainable self-compacting concrete (SCC) with better mechanical strength. Fly ash is utilized as a supplementary cement material and waste PET bottles as fiber reinforcement to improve its flexural strength and achieve the targeted compressive strength. The experimental works were conducted on eight variations using 80 specimens, divided into two main groups of partial cement replacement using 0% and 15% fly ash by weight. The two variants are added with PET fiber based on the volume fractions of 0%, 0.25%, 0.50%, and 0.75%. Fresh concrete was tested using the slump flow method (T50) and the Visual Stability Index (VSI) based on ASTM 1611. The hardened concrete tests are conducted after 56 days and include testing the concrete's compressive strength, flexural strength, and ultrasonic pulse velocity. Test results showed that the presence of PET fiber in the SCC mix decreased its flowability. However, when added up to 0.75%, the mixes still meet the flowability requirements of fresh-state SCC. PET fiber addition tends to reduce the compressive strength, whereas the reduction in compressive strength of SCC with PET fiber without fly ash is insignificant. However, in SCC that uses fly ash, the addition of PET fiber causes a significant decrease in its compressive strength. Adding PET fiber into SCC mixes can increase flexural strength, both for the two variants: SCC without fly ash and SCC with fly ash. It can be concluded that PET waste fiber with an aspect ratio of 40 can be added up to 0.5% for SCC without fly ash and up to 0.25% by volume fraction for SCC with fly ash addition. The ultrasonic pulse velocity test results have an excellent tendency to predict the concrete's compressive and flexural strengths. Therefore, the UPV test can be applied for the non-destructive test evaluation of PET fiber-reinforced SCC. Doi: 10.28991/CEJ-2023-09-02-014 Full Text: PDF

Combined Use of Non-Destructive Tests for Assessment of Strength of Concrete in Structure

Structural safety, optimization of materials, and sustainable construction practice depend on the prediction of concrete compressive strength. Traditional methods of testing use the laboratory method, which is time‑consuming, expensive, and destructive. Recent progress in deep learning has made it possible to predict the compressive strength in an accurate, rapid, and non‑destructive way by modeling nonlinear complex relationships between the constituents of concrete, curing conditions, and mechanical performance. This review is a systematic review of the deep learning architectures that have been applied to predict the concrete compressive strength with state‑of‑the‑art, such as Convolutional Neural Networks (CNNs), Artiϐicial Neural Networks (ANNs), Deep Neural Networks (DNNs), Long Short‑Term Memory (LSTM) networks, Gated Recurrent Units (GRU), Transformer‑based models, and hybrid architectures (CNN‑LSTM, CNN‑GRU, and ensemble stacking). It has been shown in the literature that higher hybrid and ensemble models allow the high predictive performance to be achieved, with the value of R² often exceeding 0.95, with the best possible models having an R² of 0.99 when using controlled datasets. Both metaheuristic optimization algorithms (e.g., PSO, GA, ACO, TLBO) and Bayesian hyperparameter tuning would greatly increase the model generalization and robustness. Moreover, interpretable artificial intelligence tools, such as SHAP and sensitivity analysis, have enhanced interpretability, and cement content, curing age, and water‑cement ratio are confirmed to be the most significant predictors of strength. Applications have been spread over the spe‑ cialized materials like ultra‑high‑performance concrete (UHPC), geopolymer concrete, recycled aggregate concrete, self‑compacting concrete, and waste‑based sustainable concretes. However, the issues of data standardization, cross‑laboratory generalization, and model transparency persist in spite of impressive advances. The future research is to be directed at physics‑informed neural networks, the multi‑objective optimization that considers the metrics of environmental impact, real‑time edge deployment, and the standardized benchmark datasets. In general, methods using deep learning as its core technology can be discussed as a revolutionary development in intelligent concrete design and sustainable construction engineering

To study the antifreezing durability of internal coating silicon fume concrete with different external coatings, fast freeze-thaw (FT) cycle testing was performed for three types of external coatings applied to the internal coatings of silicon fume concrete. Using ultrasonic testing and compressive strength tests, we analysed the relationships between the ultrasonic pulse velocity and the mechanical and physical properties of concrete under freeze-thaw action. The results show that the compressive strength and pulse velocity of the studied concrete changed little before the first 100 FT cycles but varied significantly after being subjected to 100 FT cycles and diminished linearly with increasing FT cycles. The dynamic elastic parameters of the concrete were inferred using pulse velocity calculations, and the dependence on FT cycles was very similar to that of ultrasonic pulse velocity. The concrete strength was strongly and positively correlated with ultrasonic pulse velocity. The linear regression model of between ultrasonic pulse velocity, kinetic coefficient, and compressive strength of concrete was also established. The damage incurred to the external coating material (XT-HPA + XT-SS and XT-HPS) was small, and the good performance of the concrete with the added inorganic coating after freeze-thaw cycles indicates good frost resistance.

The integration of sustainable materials such as fly ash, blast-furnace slag, recycled aggregates, and seawater into concrete mixes offers significant potential for reducing the environmental impact of construction. However, traditional experimental methods for determining the compressive strength of these concrete mixes are time-consuming and resource-intensive. This study leverages Artificial Neural Networks (ANNs) and Machine Learning (ML) to develop a predictive model for the compressive strength of sustainable concrete, using a dataset of 768 concrete mix samples. Input variables include the concrete age as well as concrete composition, including cement, water, fine and coarse aggregates, seawater, fly ash, blast-furnace slag, and superplasticizer contents, while the output is the compressive strength. The developed model captures the non-linear relationships among these variables to predict compressive strength efficiently. The best ANN model achieved a test loss of 0.051, demonstrating its ability to accurately predict compressive strength and reduce reliance on traditional testing methods. Moreover, the model’s results were compared with those of alternative algorithms to ensure its validity. These findings highlight the potential of machine learning in advancing sustainable construction practices. A relevant future research direction is to analyze feature importance in machine learning models to identify key variables and guide more effective optimization and decision-making, in addition to extending their application to other material properties and advanced concrete mixes.

Objective: Attempt is made to estimate the compressive strength of Reactive Powder Concrete (RPC) by developing Artificial Neural Network model. Methods/Analysis: RPC is a complex composite material made using a combination of constituent materials viz. fine quartz, silica fume and cement to form a cementitious matrix. Mix design for RPC is more complicated as it excludes the coarse aggregates and a low water/cement ratio is maintained. The mix proportioning methods applicable to normal concrete cannot be applied readily to RPC. The complexity in behavior of concrete with special constituents makes the mix design complicated. Artificial neural network (ANN) techniques can capture complex relations among input/output variables in a given system and this method has been implemented in Reactive Powder Concrete mix proportion design, for predicting its strength. Findings: This paper presents the development and application of ANN for predicting the strength of RPC using data related to 112 mix proportions and the results of the model are checked experimentally. Compressive strength of concrete is chosen as a function of the 'Water/Cement' ratio (w/c), 'Silica Fume/Cement' ratio (sf/c), and 'Quartz Sand/Cement' ratio (qs/c). The ANN model developed is tested for the 12 mix proportions randomly selected from experimental data for which the experimental compressive strengths were available. The compressive strength given by the model when compared with the experimental values showed an average error of only 3.99 %. Novelty/Improvement: The model developed has coefficient of determination R² equal to 0.9066 which indicates a significant enough correlation.

RETRACTED: The impact of GGBS and ferrous on the flow of electrical current and compressive strength of concrete

Compression strength is the most important factor in concrete structures, and there are two methods to evaluate that compressive strength, destructive and non-destructive method. This work presents a study on the deference and comparison between Destructive (compressive strength test) and a Non-Destructive Method (Rebound Hammer and Ultrasonic test) for testing the compressive strength of concrete.The investigations aimed to develop the method of assessment the strength of concrete of both non-destructive tests with greater accuracy. Destructive method, to determine the strength of the in-situ concrete, and also destructive testing (DT), includes methods where the material is broken down in order to determine its mechanical properties. From the obtained results it is observed that the Rebound Hammer readings increased with the compressive strength of concrete. And the Ultrasonic pulse velocity were greatly influenced by the cements and aggregate, extent of moist curing and presence of flaws and voids in concrete, more than their influence on the measured strengths. This demonstrates the limitation of using ultrasonic pulse velocity tests for estimating compressive strength of concrete. Combined use of Ultrasonic pulse velocity (UPV) and Rebound Hammer tests for assessment of concrete strength in structures with greater reliability.

Investigation of engineering properties of normal and high strength fly ash based geopolymer and alkali-activated slag concrete compared to ordinary Portland cement concrete

Among all the concrete characteristics, the compressive strength is known as the most important feature and therefore, it is often utilized to quantify the quality of concretes. The assessment of concretes strength in existing constructions is important during their service life. Thus, in addition to destructive evaluation of strength, many non-destructive techniques have been adopted to assess the concretes strength. Ultrasonic Pulse Velocity (UPV) is one of the non-destructive techniques that involve measuring the speed of ultrasonic wave through concretes to predict concretes strength. According to the Griffith’s theory, the strength of the material is greatly affected by the defects’ existence (for example, small cracks). These cracks can be current prior to the application of any loading or could create throughout the load application. The presence of micro-cracks in concretes that generate due to the applied load may affect the UPV test results in comparison with unloaded concretes. An experimental investigation is conducted in order to assess the impact of the applied of load (ultimate load) on the measured ultrasonic wave velocity and compared with the results of compressive strength from destructive test. A total of thirty (150 mm × 150mm) cube specimens have been utilized with water to cement ratios (0.45). These cubes have been examined in ultrasound without any loads and then they have been tested in the same way with applied ultimate load which is approximately equal to half of the required design load (failure load) and then inspected or tested in the normal way (destructive test). In order to assess the compressive strength of concretes using UPV, loading condition is considered in order to study the impact of the applying load. In addition, the water absorption has been examined for thirty concretes samples under the impact of the applied loads (20%, 40% and 60%). The results of non-destructive tests of concretes samples under the influence of imposed loads (60%) have showed to be less by compressive strength 5 MPa than the results obtained from testing of unloaded concretes samples through the resulting equation and comparing them to the equation from previous research. This is consistent with the water absorption test of the specimens under the imposed loads (60%) where Absorption values have been greater than concretes models without applied loads.