Prediction Of Compressive Strength Research Articles

Study on Strength Prediction and Material Scheme Optimization for Modified Red Mud Based on Artificial Neural Networks

Focusing on the complex nonlinear problems of strength prediction and the material scheme design of modified red mud for use as a road material in engineering applications, a strength prediction neural network is established and utilized to optimize the material scheme, including the compound-solidifying agent ratio, water content, and curing age, based on experimental data accumulated during years of engineering practice and an artificial neural network. In this study, a backpropagation (BP) neural network is adopted, and 114 sets of experimental data are used to train the parameters of the unconfined compressive strength prediction model. Then, using the BP strength prediction model, the material scheme optimization process is carried out, with the strength and material costs as the objectives. The results show that the BP neural network model has a high prediction accuracy, the relative prediction error is basically within 10%, the root-mean-squared error is less than 0.04, and the correlation coefficient is more than 0.99. According to the strength requirements of modified red mud in different road projects and the constraints of each property, an optimal material scheme with a lower cost and higher 7 d target strength is obtained using a mix of polymer agent–fly-ash–cement–speed-cement in a ratio of 0.02%:1.96%:4.78%:0%, with a 33.93% water content of raw red mud, so that the target strength and material cost are 2.987 MPa and 17.099 CNY/T. This study creates an optimal material scheme, incorporating the compound-solidifying agent ratio, curing age, and water content of the modified red mud road material according to the strength requirements of different projects, thereby promoting the popularization of the utilization of red mud with better engineering practicability and economy.

Compressive strength prediction of high-performance concrete using MLP model

ABSTRACT The remarkable mechanical properties of high-performance concrete (HPC) make it indispensable in a wide range of engineering applications. Predicting HPC’s compressive strength with precision is essential to guaranteeing its structural stability and longevity. The present work introduces a novel method for predicting HPC compressive strength (CS) by utilizing the Multi-layer Perceptron (MLP) model in conjunction with three distinct optimizers, namely Northern Goshawk Optimization (NGO), Manta Ray Foraging Optimization (MRFO), and Atom Search Optimization (ASO). An effective method for capturing complex relationships between input and output variables is the multi-layer perceptron (MLP) in artificial neural networks. Developing a dependable, robust predictive model is the goal of training the MLP model on a variety of datasets of HPC mixtures and CS. As evidenced by the MLNG model’s R2 value of 0.994 and RMSE of 1.3572, the joint efforts of MRFO, NGO, and ASO during the optimization phase improve the MLP model and produce promising results. These results clarify how different optimizers play a crucial role in improving the accuracy of HPC compressive strength predictions made with the MLP model. This information helps make better decisions regarding design and construction methods for HPC.

Experimental Study on Mechanical Properties and Stability of Marine Dredged Mud with Improvement by Waste Steel Slag

As marine-dredged mud and waste steel slag in coastal port cities continue to soar, the traditional treatment method of land stockpiling has caused ecological problems. Thus, it is necessary to find a large-scale resource-comprehensive utilization method for dredged mud and waste steel slag. This study uses waste steel slag and composite solidifying agents (cement, lime, fly ash) to physically and chemically improve marine-dredged mud. The physical improvement effect of the particle size and dosage of waste steel slag was studied by the shear strength test under the effect of freeze–thaw cycle. Then, based on the Box–Behnken design of the response surface method, the interaction effects of the solidifying agent components on the unconfined compressive strength were studied. Then, the water stability under dry–wet cycles and a microscopic mechanism were analyzed by XRD and SEM tests. The results show that the waste steel slag with a dosage of 30% and a particle size of 1.18~2.36 mm has the best improvement. The interaction between cement and lime and lime and fly ash has a significant effect on the linear effect and surface effect of 7d unconfined compressive strength, and the strength increases first and then decreases with the increase in its dosage. For the 14d unconfined compressive strength, only the interaction between cement and lime is still significant. The unconfined compressive strength prediction model is established to optimize the mix ratio of the composite solidifying agent. In the water stability, the water stability coefficients of the 7d and 14d tests are 0.68 and 0.95, respectively, and the volume and mass loss rates are all below 1.5%, showing a good performance in dry–wet resistance and durability. Microscopic mechanism analysis shows that waste steel slag provides an ‘anchoring surface’ as a skeleton, which improves the pore structure of dredged mud, and the hydration products generated by the solidifying agent play a role in filling and cementation. The results of the study can provide an experimental and technical basis for the resource engineering of marine-dredged mud and waste steel slag, helping the construction of green low-carbon and resource-saving ports.

Compressive Strength Prediction of Fly Ash-Based Concrete Using Single and Hybrid Machine Learning Models

The compressive strength of concrete is a crucial parameter in structural design, yet its determination in a laboratory setting is both time-consuming and expensive. The prediction of compressive strength in fly ash-based concrete can be accelerated through the use of machine learning algorithms with artificial intelligence, which can effectively address the problems associated with this process. This paper presents the most innovative model algorithms established based on artificial intelligence technology. These include three single models—a fully connected neural network model (FCNN), a convolutional neural network model (CNN), and a transformer model (TF)—and three hybrid models—FCNN + CNN, TF + FCNN, and TF + CNN. A total of 471 datasets were employed in the experiments, comprising 7 input features: cement (C), fly ash (FA), water (W), superplasticizer (SP), coarse aggregate (CA), fine aggregate (S), and age (D). Six models were subsequently applied to predict the compressive strength (CS) of fly ash-based concrete. Furthermore, the loss function curves, assessment indexes, linear correlation coefficient, and the related literature indexes of each model were employed for comparison. This analysis revealed that the FCNN + CNN model exhibited the highest prediction accuracy, with the following metrics: R2 = 0.95, MSE = 14.18, MAE = 2.32, SMAPE = 0.1, and R = 0.973. Additionally, SHAP was utilized to elucidate the significance of the model parameter features. The findings revealed that C and D exerted the most substantial influence on the model prediction outcomes, followed by W and FA. Nevertheless, CA, S, and SP demonstrated comparatively minimal influence. Finally, a GUI interface for predicting compressive strength was developed based on six models and nonlinear functional relationships, and a criterion for minimum strength was derived by comparison and used to optimize a reasonable mixing ratio, thus achieving a fast data-driven interaction that was concise and reliable.

Ensemble machine learning models for compressive strength and elastic modulus of recycled brick aggregate concrete

This study delivers an in-depth analysis of predicting the compressive strength and elastic modulus of recycled brick aggregate concrete (RBAC). It assembles an extensive database of 633 compression test results and develops five ensemble learning models—random forest (RF), gradient boosting regression trees (GBRT), extreme gradient boosting (XGB), light gradient boosting machine (LGBM) and stacking (St). These models are evaluated against existing empirical formulas with respect to their effectiveness. The findings reveal that machine learning (ML) models outperform existing formulas: for compressive strength prediction, traditional models achieve a maximum determination coefficient R² of 0.38, whereas ML models attain an R² range of 0.91–0.94. For elastic modulus, the highest R² values are 0.44 for traditional models and 0.97 for ML models. Notably, the LGBM and St models excel in predicting compressive strength and elastic modulus, respectively. The study also identifies critical parameters influencing RBAC’s compressive behavior and highlights their impact trends. Remarkably, while the mass-weighted water absorption of coarse aggregates and the replacement ratio of recycled brick aggregates (RBAs) have less impact on compressive strength compared to the effective water-to-cement ratio, they become more influential for elastic modulus. Additionally, the decrease in elastic modulus due to higher RBA replacement ratios or increased mass-weighted water absorption of coarse aggregates exceeds the corresponding reduction in compressive strength. This research not only deepens the understanding of RBAC’s mechanical properties but also provides valuable predictive tools for civil engineering applications.

Predictive models for properties of hybrid blended modified sustainable concrete incorporating nano-silica, basalt fibers, and recycled aggregates: Application of advanced artificial intelligence techniques

The main objective of this work is to improve the compressive strength of concrete, specially in sustainable construction is to develop more precise predictive modeling techniques. The compressive strength prediction of basalt fiber reinforced concrete filled with nano-silica and recycled aggregates can be done using a hybrid deep learning model suggesting the use of the combination of Convolutional Neural Networks and Long Short-Term Memory networks. The CNN captures microstructural features from SEM images, while the LSTM models temporal dependencies from sequential curing data samples. To enhance the prediction accuracy, PCA was performed on feature dimensionality reduction and GA optimized hyperparameters both for the model as well as the concrete mix design for improved strength with cost effectiveness. With an R² value of 0.92–0.95, the performance results of the presented model came out better than the baseline models, as well as reducing the MAE by 20 %. Besides, there existed a 5–8 % better compressive strength in GA-optimized mix designs. Robustness comes into play with the model that shows steady strength predictions, regardless of conditions of curing under multiple conditions and at different material composition levels. Furthermore, the reutilization of recycled aggregates and nano-silica gives a real environmental benefit as less waste is produced but the material performance is maximized. This kind of outcome indicates how the proposed model can be practically applied in optimizing concrete design in terms of strength and sustainability features, thus providing an accessible instrument for decision-making in the construction field. It is an effective tool to improve the performance of concrete while minimizing environmental and material wastes.

A novel composite machine learning model for the prediction of compressive strength of blended concrete

A novel composite machine learning model for the prediction of compressive strength of blended concrete

Enhancing Unconfined Compressive Strength prediction in geotechnical engineering: a novel approach using LSZO and LSWG optimization

ABSTRACT The Unconfined Compressive Strength (UCS) measures a rock’s ability to withstand axial loads without lateral support, crucial for engineering applications. Accurate prediction of UCS is critical for the stability and safety of various civil engineering projects, such as foundation design, mining, and tunneling. The ability to predict UCS reliably ensures better project planning and execution, ultimately enhancing the safety and efficiency of engineering applications. The paper aims to develop an innovative method for predicting the UCS of rocks using a combination of the Zebra Optimisation Algorithm (ZOA) and the Wild Geese Algorithm (WGA) integrated with the Least Square Support Vector Regression (LSSVR) predictive model. The novelty of the paper lies in the integration of ZOA and WGA optimizers with the LSSVR model to address the limitations of conventional methods, which often face challenges like slow achieving convergence and being stuck in local minima. The combination of these advanced algorithms results in significantly improved accuracy and convergence speed in UCS predictions, demonstrating a robust and reliable approach that advances the field of geotechnical engineering. The results show that combining ZOA and WGA optimizers with LSSVR enhances UCS prediction accuracy and speed convergence. The LSZO models achieve a high level of precision, with a low RMSE value of 2.206 and an R2 value of 0.993. The safety and effectiveness of civil engineering projects are improved by this model, which provides a strong and trustworthy method for UCS prediction.

Enhancing compressive behaviour of seawater sea sand concrete filled hybrid carbon-glass FRP tubes exposed to seawater: Effect of thickness

This study presents a novel investigation into the impact of tube thickness on the residual compressive strength of sea water sea sand concrete (SWSSC) filled filament wound hybrid fibre-reinforced polymer (HFRP) tubes, positioning them as innovative alternatives to traditional concrete-filled steel tubes in corrosive marine environments. The research explores tube thicknesses of 2 mm, 3 mm, and 4 mm, employing a unique hybrid configuration of 50 % carbon fibre reinforced polymer (CFRP) internal layers and 50 % glass fibre reinforced polymer (GFRP) external layers. This study examines the effect of these configurations on compressive mechanical properties, microstructural characteristics, and damage progression under alkaline and seawater conditions. Additionally, the research evaluates cross-ply and hoop orientations as variables in filament winding fibre orientation. A comprehensive set of 108 hybrid tubes were filled with an alkaline solution to simulate the SWSSC environment and exposed to seawater at varying temperatures and durations. Accelerated aging experiments were conducted in the laboratory to assess the long-term compressive performance of SWSSC-filled HFRP tubes exposed to seawater. Long-term compressive strength predictions were made using the Arrhenius theory, based on short-term accelerated aging data. The accelerated test data revealed maximum compressive strength reductions of 33 %, 15 %, and 17 % for 2 mm, 3 mm, and 4 mm cross-ply HFRP tubes, respectively, after 150 days of exposure at 60 °C. For hoop HFRP tubes, the corresponding reductions were 19 %, 18 %, and 17 %. Thicker tubes (3 mm and 4 mm) with sufficient internal CFRP layers protect external GFRP layers from alkaline damage, akin to pure CFRP tubes. Thinner tubes (2 mm) exhibit similar performance to GFRP tubes as alkaline damage spreads from their less substantial internal CFRP layers to the outer GFRP layers. For cross-ply tubes, the recommended strength reduction factors are 0.6 for 2 mm, 0.8 for 3 mm, and 0.8 for 4 mm thicknesses. For hoop tubes, the suggested factors are 0.7 for 2 mm, 0.7 for 3 mm, and 0.8 for 4 mm thicknesses. This study shows the innovative potential of tailored hybrid HFRP tubes in enhancing the durability and performance of marine infrastructure.

Comparing Machine Learning Regression Models for Early-Age Compressive Strength Prediction of Recycled Aggregate Concrete

A branch of artificial intelligence called machine learning is well-positioned as a prediction method that can take into consideration several influencing factors and complex inter-factor connections. Without being specifically trained to do so, these machine learning models have the ability to generalize, predict, and learn from data. Regression theory is a key topic in statistical modelling and machine learning. The main goal of this study is to compare the performance of several popular machine learning regression models for predicting the early-age compressive strength of concretes made from recycled concrete aggregates from a structure that demolished following the Sivrice-Elazig earthquake on January 24, 2020. Early-age concrete compressive strength is even more crucial due to factors like the fact that there are thousands of newly built structures in the aftermath of the earthquake, the quick manufacturing of these structures, and the completion of the project in the lowest amount of time. Determining the early-age concrete strength with high accuracy and in a useful manner is crucial for this reason. Seven different classical machine learning algorithms were employed in this study to achieve all of these goals. Early-age concrete compressive strength values were considered for 1 and 3 days. The relationship between the experimental results and the predicted outcomes of the employed algorithms was investigated, and a thorough comparison of these intelligent regression algorithms was conducted. Within the scope of sustainable development and circular economy goals, it is thought that this article will make significant contributions to the literature in terms of utilizing these waste materials and determining the early-age compressive strengths of the concretes produced with high accuracy.

Machine learning based prediction of compressive and flexural strength of recycled plastic waste aggregate concrete

In the last 50 years, the use of plastics has increased significantly due to advances in technology, population growth and increasing needs. However, this trend has led to the accumulation of large amounts of plastic waste. Numerous studies have been conducted on the use of recycled plastic materials in concrete production as a means of recycling and environmental protection. This study aims to predict the 28-day cylinder compressive and flexural strength of concretes using Polyethylene Terephthalate (PET), Phenolic Formaldehyde (PF), Polypropylene (PP), Polycarbonate (PC), Polyvinyl Chloride (PCV), High-Impact Polystyrene (HIPS) and High-Density Polyethylene (HDPE) plastic waste as aggregates. In this scope, a comprehensive database was created using experimental results obtained from the literature. Then, this study used six ML models, including Support Vector Regressor (SVR), Random Forest Regressor (RFR), Gradient Boosting Regressor (GBR), XGBoost Regressor (XGR), LightGBM Regressor (LGR), Bagged Decision Trees Regressor (BDTR), to predict the compressive and flexural strength of recycled plastic waste aggregate concrete (RPWAC). From the experimental database, 253 data points with strength between 6.00–70.05 MPa were identified to estimate the compressive strength (CS) and 175 data points with strength between 1.21–7.96 MPa were identified for flexural strength (FS). Cross-validation was used for six ML models and hyperparameter fine-tuning was performed. The results showed that six different ML models predicted the compressive and flexural strengths of RPWAC with high accuracy. The XGR model predicted CS and FS better than the other five different models with R2 values of 0.931 and 0.999, respectively. The three main properties effecting the CS of RPWAC were plastic waste content, water/cement ratio and concrete specific gravity, while the three main properties affecting the FS of RPWAC were cement specific gravity, plastic waste content and cement CS.

Intelligent Models for Prediction of Compressive Strength of Geopolymer Pervious Concrete Hybridized with Agro-Industrial and Construction-Demolition Wastes

Intelligent Models for Prediction of Compressive Strength of Geopolymer Pervious Concrete Hybridized with Agro-Industrial and Construction-Demolition Wastes

Comparing durability and compressive strength predictions of hyperoptimized random forests and artificial neural networks on a small dataset of concrete containing nano SiO2 and RHA

Through the increasing use of supplementary cementitious materials, the properties of concrete have taken on increased significance in a design code. Using reliable prediction models based on a small data set for the mechanical properties and durability of concrete can reduce the number of trial batches and experiments needed to produce useful design data in the laboratory, reducing time as well as resources. In this study, we investigate how the properties of water penetration, chlorine resistance, and compressive strength can be predicted by polynomial regression (PR), random forest (RF) regression, and artificial neural networks (ANNs) based on the input values of density, workability, and the constituent amount of rice husk ash, cement, and nano SiO 2 . We vary the training data used and test the coefficient of determination ( R 2 score) on the remaining data as a test set to measure predictive capability. We show that RFs and ANNs outperform PR in all settings and have unambiguously extrapolating properties when hyperparameter optimization is designed for this purpose. Remarkably, we obtain R 2 scores on the test data of 0.858 − 0.990 for RFs and 0.825 − 0.985 for ANNs.

Investigation and prediction of impact of aggregate size and shape on porosity and compressive strength of pervious concrete

ABSTRACT Pervious concrete is proven to environmentally benefit urban pavement construction. Optimising compaction energy evidently reduces uncertainty in the mass production of pervious concrete with uniform characteristics. The distribution of compaction energy in wet mix and rearrangement of binder-coated aggregate particles are affected by the size and shape of aggregates. This study analyses the impact of aggregate size and shape on porosity and compressive strength prediction from mix-design parameters. Aggregate-to-cement ratio (3, 4 and 5), compaction (0, 30 and 60 blows from standard Proctor rammer), aggregate sizes (5–12 mm, 12–18 mm and 18–25 mm) and aggregate shape (0, 200 and 1000 revolutions of milling in Los Angeles Abrasion Value) were used to cast a total number of 486 samples. Constituents were mixed in an electrically powered drum of capacity 96 L, for 20 min. Porosity was computed from constituent ratios (theoretical porosity) and apparent weight (measured porosity). The compressive strength of samples was recorded using a Universal Testing Machine. The measured porosity and theoretical porosity relationship was significantly affected by the shape of the aggregate and not by the size. Modified Ryshkewitch’s model predicted compressive strength from porosity, aggregate-to-cement ratio, compaction and aggregate size and shape with an accuracy of 0.92 (R2). Two of six arbitrary constants depended on aggregate shape and size.

Utilizing construction and demolition waste in concrete as a sustainable cement substitute: A comprehensive study on behavior under short-term dynamic and static loads via laboratory and numerical analysis

This study explores the utilization of recycled concrete powder (RCP) as a sustainable alternative to conventional cement in concrete production, focusing on its impact on mechanical properties and impact resistance. Two series of concrete mixtures were prepared: one without and one with 0.5 % polypropylene fibers by RCP-replacing cement at varying proportions (0 %–50 %). Methodologically, the research involved both laboratory tests and predictive modeling using artificial neural networks (ANN) and fuzzy logic (FL). Laboratory evaluations assessed compressive, flexural, and tensile strengths under static loads, as well as impact resistance through repeated drop weight impact (RDWI) tests. Key findings reveal that RCP-incorporation of 15 % leads to negligible reductions in compressive, flexural, and tensile strengths, while higher replacements significantly decrease performance. The addition of polypropylene fibers notably improved mechanical properties across all RCP replacement levels, enhancing compressive strength by up to 23.65 % and flexural strength by 27.49 % at the upper limits of RCP substitution. Accurate predictions of 28-day compressive strength were achieved via the ANN model with an average prediction error less than 1 %. The study also validates that the Weibull distribution effectively describes the impact resistance of concrete mixtures, confirming the advantages of combining RCP with fiber reinforcement to achieve sustainable and high-performance concrete solutions.

An explainable machine learning approach to predict the compressive strength of graphene oxide-based concrete

Graphene oxide (GO) has shown promise in improving concrete strength. Despite its frequent use in cement composites, its effect on concrete properties is less explored. The influence of GO on concrete remains unclear due to various interactions among constituents such as GO type and percentage, Superplasticiser (SP) type and percentage, dispersion technique, and curing age. Therefore, making prediction of compressive strength using simple mathematical formation is challenging. This study represents the first-ever attempt at developing a comprehensively validated machine learning model to predict GO-concrete's compressive strength characteristics by considering all key variable parameters. A comprehensive laboratory experiment program was conducted to collect the data required for training the machine learning (ML) models. Once the ML models were trained, they were used to predict the relationship of each of those input variables towards the compressive strength properties. Four machine learning models—(a) multiple linear regression (MLR), (b) k-nearest neighbour (KNN), (c) random forest (RF) and (d) extreme gradient boost (XGB)—were utilised to model the relationship between input parameters and compressive strength. Also, explainable machine learning (XML) methods were employed to elucidate the impact of mixed constituents on the compressive strength of GO concrete. The results from test predictions showcase that the XGB has attained a coefficient of determination (R2) of 0.981, coefficient of correlation (R) of 0.99, mean square error (MSE) of 0.9, mean normalised bias (MNB) of 0.004, scatter index (SI) of 0.2 with mean absolute error (MAE) of 0.8 MPa for the predictions, outperforming the remaining models. XML highlighted the true impact of GO and the remaining variables, emphasising that the presence of GO significantly improves compressive strength. The optimum GO content and optimum superplasticiser content observed from the experimental work agreed with results obtained from the XML analysis, showing the consistency of the explanation with the experimental results. This implies the superiority of using XML methods in concrete mix design applications in industry.

Prediction of the Unconfined Compressive Strength of a One-Part Geopolymer-Stabilized Soil Using Deep Learning Methods with Combined Real and Synthetic Data

Prediction of the Unconfined Compressive Strength of a One-Part Geopolymer-Stabilized Soil Using Deep Learning Methods with Combined Real and Synthetic Data

Deep learning neural networks for monitoring early-age concrete strength through a surface-bonded PZT sensor configuration

Monitoring the immediate evolution of concrete’s strength is essential to ensure structural integrity and construction efficiency, requiring Forecasting to avoid unforeseen and severe failures during construction. The present investigation introduces an Equivalent structural parametric (ESP) study using a surface-bonded piezo sensor and electromechanical impedance (EMI) methodology to monitor and forecast the strength of concrete by implementing machine learning and deep learning methods. The concrete hydration processes are simulated using COMSOL™ 5.5. The concrete cube’s hydration is represented by altering Young’s modulus and damping ratio to show the rate of curing. Concrete strength development is examined in terms of conductance resonant frequency (CRF) and Conductance resonant peak (CRP). Continuous conductance signature monitoring and data analysis show that CRF and CRP increase with compressive strength. The system’s mechanical impedance is measured, and EMI signatures vs. frequency plots within a specific frequency range are compared to healthy impedance graphs. A mass-spring-damper system with identical properties is identified, and relevant structural parameters are computed. Modern ML algorithms include linear regression (LR), interaction LR, fine, medium, and coarse Gaussian SVM, etc. reliably predict strength with an error rate of less than 2%. Convolutional neural networks (CNN) have advanced image-based recognition, but their usage in EMI-based structural strength assessment is still being studied. A unique strategy using 2D CNN, 2D CNN– Long short-term memory (LSTM), and 2D CNN Bidirectional-LSTM to forecast concrete structure compressive strength shows the potential of deep learning. The proposed 2D CNN-Bi-LSTM model excels in compressive strength prediction, obtaining an R2 value of 0.99 and stabilizing loss error over a few epochs.

A Prediction Model for the Unconfined Compressive Strength of Pervious Concrete Based on Mix Design and Compaction Energy Variables Using the Response Surface Methodology

Pervious concrete is desirable for water drainage in building systems, but achieving both high strength and good permeability can be challenging. Also, the importance of compaction energy is significant in determining the efficiency of pervious concrete. However, research on the development of unconfined compressive strength (UCS) prediction models for pervious concrete materials that incorporate compaction energy parameters remains unexplored. Therefore, this study aimed to balance strength and permeability while optimizing the compaction energy required for concrete production. A Central Composite Design (CCD) was used to design experiments within the response surface methodology (RSM) and evaluate the UCS, the porosity and permeability of pervious concrete specimens produced with varying cement content (280.00–340.00 kg/m3), the water-to-cement ratio (0.27–0.33), the aggregate-to-cement ratio (4:1–4.5:1), and compaction energy (represented by VeBe compaction time, 13–82 s). A regression model with goodness of fit (R2adjusted > 0.87) was calibrated to estimate the UCS of pervious concrete as a function of mix design parameters and VeBe compaction time (Tvc). This model can potentially guide field practices by recommending compaction strategies and mix designs for pervious concrete, achieving a desirable balance between mechanical strength and hydraulic permeability for building construction applications.

Explainable Ensemble Learning and Multilayer Perceptron Modeling for Compressive Strength Prediction of Ultra-High-Performance Concrete.

The performance of ultra-high-performance concrete (UHPC) allows for the design and creation of thinner elements with superior overall durability. The compressive strength of UHPC is a value that can be reached after a certain period of time through a series of tests and cures. However, this value can be estimated by machine-learning methods. In this study, multilayer perceptron (MLP) and Stacking Regressor, an ensemble machine-learning models, is used to predict the compressive strength of high-performance concrete. Then, the ML model's performance is explained with a feature importance analysis and Shapley additive explanations (SHAPs), and the developed models are interpreted. The effect of using different random splits for the training and test sets has been investigated. It was observed that the stacking regressor, which combined the outputs of Extreme Gradient Boosting (XGBoost), Category Boosting (CatBoost), Light Gradient Boosting Machine (LightGBM), and Extra Trees regressors using random forest as the final estimator, performed significantly better than the MLP regressor. It was shown that the compressive strength was predicted by the stacking regressor with an average R2 score of 0.971 on the test set. On the other hand, the average R2 score of the MLP model was 0.909. The results of the SHAP analysis showed that the age of concrete and the amounts of silica fume, fiber, superplasticizer, cement, aggregate, and water have the greatest impact on the model predictions.