Model For Prediction Of Compressive Strength Research Articles

In-depth insight into the driving factors of the compressive strength development of MKPC based on interpretable machine learning methods

Magnesium potassium phosphate cement (MKPC) is a kind of Mg-chemically bonded phosphate ceramic commonly used for rapidly repairing dilapidated structures. In this study, a compressive strength dataset of MKPC was constructed, and four advanced machine learning (ML) algorithms (XGBoost, RF, GBDT and ANN) were selected to establish a high-precision compressive strength prediction model of MKPC. The SHAP and PDP methods are also used for interpretability analysis of ML-MKPC models. The XGBoost model has good generalizability and reliability while achieving high prediction accuracy. The RF and GBDT models performed similarly to the XGBoost model on the training set but performed poorly on the testing set. The ANN model is poorly trained on both the training and testing sets, with a risk of underfitting. The R2 of the XGBoost model at the different compressive strength stages still reaches above 0.80, indicating that it not only captures the complex relationships of the overall dataset well but also effectively predicts the staged strength dataset. Feature importance analysis revealed that the curing age (T), water-to-binder ratio (W/B), mineral admixtures-to-binder ratio (MA/B) and phosphate-to-magnesium ratio (P/M) are the principal variables affecting the compressive strength of MKPC. The partial interpretation shows that the optimum value range is determined when W/B is 0.10–0.18, MA/B is 0–0.20, P/M is 0.40–1.0, and R/M is 0–0.12. The composition of mineral admixtures with high-Ca, high-Si and low-Al systems seems to be more conducive to participating in the hydration reaction of MKPC. The ML-MKPC compressive strength prediction model developed in this study can provide theoretical support for the subsequent composition design and performance optimization of MKPC.

Concrete Compressive Strength Prediction by Ensemble Machine Learning Approach

The prediction of concrete compressive strength is a crucial aspect of ensuring the structural integrity and durability of construction projects. In recent years, machine learning approaches have improved upon the limitations of empirical formulas and laboratory testing methods for predicting concrete compressive strength. This study utilizes ensemble machine-learning techniques, such as Bagging, XGBoost, and Stacking models, to enhance the accuracy of concrete compressive strength prediction models. A five-fold cross-validation technique was applied to mitigate the problems of underfitting or overfitting in the regression model. Furthermore, various statistical indices were employed to compare the forecasting performance of these ensemble techniques. The prediction performance of this research revealed that XGBoost achieved the highest R-squared value of 93%, followed by Stacking and Bagging regression models at 92%. Consequently, this research underscores the potential of ensemble techniques as valuable tools in the domain of civil engineering, paving the way for more reliable and efficient construction practices.

Prediction model for compressive strength of rice husk ash blended sandcrete blocks using a machine learning models

Prediction model for compressive strength of rice husk ash blended sandcrete blocks using a machine learning models

Machine learning guided iterative mix design of geopolymer concrete

Current mix design methods for Geopolymer concrete (GPC) require substantial efforts of trial-and-error experiments and is applicable only to those formulated by specific precursor materials. In this work, a machine learning (ML) guided mix design method for GPC is proposed, which can considerably reduce the experimental workload and be versatile with a broad range of precursor materials. First, a database for the slump and compressive strength of GPC was established, and ML prediction models were developed with these as objectives. Subsequently, the mix design for GPC was conducted using the particle swarm optimization (PSO) algorithm. The designed mixture proportions underwent experimental validation, and if targets were not met, results were then used for model iteration until desired performance targets were achieved. The results showcase the excellent performance of the established slump and compressive strength prediction models, with an accuracy and coefficient of determination of 94 % and 0.95, respectively. The developed ML guided model effectively generates concrete mix achieving compressive strengths of 20 MPa, 40 MPa, and 60 MPa, while concurrently satisfying slump requirements. The experimentally validated results for the three designated target strengths after iterations were 25.2 MPa, 43.8 MPa, and 66.6 MPa, respectively. These positive results emphasize the ability of the proposed optimized mix design method to meet both the strength and workability targets in GPC via a minimum number of trail experiments.

Multi-objective optimization design of recycled aggregate concrete mixture proportions based on machine learning and NSGA-II algorithm

This paper employs Support Vector Regression (SVR), Random Forest Regression (RF), Gradient Boosting (GB), and Extreme Gradient Boosting (XGB) algorithms to establish the compressive strength prediction models for Recycled Aggregate Concrete (RAC) and analyze the influence of ten inputs on RAC compressive strength. Combined with the best prediction model, the Non-dominated Sorting Genetic Algorithm Ⅱ (NSGA-II) is applied for multi-objective optimization of mixture proportions in RAC addressing cost, carbon emissions, and compressive strength as key objectives. The results demonstrate that the RAC compressive strength prediction model using the GB algorithm exhibits the highest accuracy, with the R2 value of 0.99 for the training set and 0.85 for the testing set. Feature importance and SHAP analysis reveal that cement and water contents are the primary factors affecting the compressive strength of RAC. Among the three mineral admixtures including Silica Fume (SF), Fly Ash (FA), and Ground Granulated Blast Furnace Slag (GGBFS), SF exhibits superior improvement in the compressive strength of RAC compared to FA and GGBFS. Partial dependency plot analysis indicates that concrete strength remains unaffected by recycled coarse aggregate content within the range of 0 to 300 kg/m3. Pareto fronts of a tri-objective mixture optimization problem for RAC are successfully obtained through the GB model combined with the NSGA-II algorithm, which can guide the optimization of RAC preparation.

Statistical Reliability Analysis of Ultrasonic Velocity Method for Predicting Residual Strength of High-Strength Concrete under High-Temperature Conditions.

Herein, we conducted a comprehensive statistical assessment of the ultrasonic pulse velocity (UPV) method's effectiveness in predicting concrete strength under diverse conditions, specifically early age, middle age, and high-temperature exposure. The concrete mixtures, with water-to-cement (W/C) ratios of 0.33 and 0.28, were classified as granite aggregate or coal-ash aggregate mixes. Compressive strength and UPV measurements were performed under these conditions, and subsequent statistical analyses treated the identified factors as distinct groups. The results revealed a substantial difference in compressive strength between specimens at early age (average of 13.01) and those at middle age (average of 41.96) and after high-temperature exposure (average of 48.08). Conversely, UPV analysis showed an insignificant difference between the early-age specimens and those after high-temperature exposure. The analysis of the W/C ratio and coarse aggregate demonstrated significant differences (p-value < 0.05) in compressive strength between specimens in middle age and those exposed to high temperatures, excluding the early-age specimens. However, UPV analysis revealed insignificant differences, with p-values of 0.67 and 0.38 between specimens at an early age and post-high-temperature exposure, respectively. Regression analysis identified suitable functions for each scenario, emphasizing the importance of a strength prediction model for concrete after high-temperature exposure, particularly considering the W/C ratio. Since concrete showed statistically different compressive strength, UPV, and strength prediction models in three conditions (early age, middle age, and high temperature), different strength prediction models must be used for the purpose of accurately predicting the strength of concrete.

Study on Coordinated Deformation Failure Mechanism and Strength Prediction Model of Rock-lining Concrete

Shotcrete has been widely used in the excavation of caverns and tunnels as a key load-bearing element in the support system. However, the interfacial stresses and bond strengths within the composite structures formed by the interaction of geologic bodies and tectonic formations are not well understood, which can lead to safety accidents and excessive wastage of support materials in engineering. To address this gap, rock-concrete composite specimens were created using six distinct types of surrounding rocks. The strength law of composite specimens was analyzed, and the uniaxial compressive strength prediction model of composite specimens was derived based on Mohr−Coulomb strength yield criterion. The results show that the uniaxial compressive strength of the composite specimen changes with the change of rock type, which is between the strength of rock and concrete. The uniaxial peak strength and elastic modulus of composite specimens are linearly positively correlated with the ratio (Y) of concrete-rock elastic modulus, which is easily affected by concrete properties. However, the peak strain is linearly and negatively correlated with Y, which is greatly influenced by rock. The rock-concrete interface of the composite specimen plays an important role in the failure of the specimen, which changes the stress transfer state of the composite specimen, produces uncoordinated deformation, and finally causes the failure of the specimen. The theoretical and test value error of the derived composite specimen uniaxial compressive strength prediction model are –1.19%~12.20%, and the theoretical calculation model can be used to predict the uniaxial compressive strength of rock-concrete composite specimens.

Strength Prediction of Non-Sintered Hwangto-Substituted Concrete Using the Ultrasonic Velocity Method.

This paper presents and investigates the properties of concrete in which a portion of the cement is substituted with non-sintered Hwangto (NSH), a readily available building material in Asia. Given the inactive nature of NSH, this study aimed to determine the optimal cement replacement ratio and quantitative strength of the material. The unit weight, compressive strength, ultrasonic pulse velocity (UPV), and stress-strain of the NSH concrete (NSHC) were evaluated. Additionally, we developed a predictive model for determining compressive strength based on the regression analysis of compressive strength and UPV. The water-to-binder ratio was set to 0.41, 0.33, and 0.28, and the NSH replacement rates in the cement were set to 0%, 15%, 30%, and 45% for evaluating various strength ranges. The mechanical property measurements indicated reductions of 5.35% in unit weight, 35.62% in compressive strength, and 6.34% in UPV as the NSH was replaced. Notably, the smallest deviation from plain concrete was observed at a replacement rate of 15%. The scanning electron microscopy analysis results showed that the plain concrete exhibited a crystalloid structure; however, as the NSH replacement rate increased, the amorphous structure and pores increased while unreacted NSH particles were also observed. The X-ray diffraction analysis results demonstrate that the peak intensities for kaolinite and mullite increased as the NSH replacement rate increased, while those of C-S-H gel and CaO showed low peak intensities. Furthermore, the regression analysis concluded that an exponential function was suitable. Consequently, a compressive strength prediction model was developed, and in the error test, the NSHC model demonstrated an average error of <10%, with fewer errors at the lower compressive strength boundaries.

Prediction model of compressive strength of foamed concrete considering pore size distribution

To more accurately evaluate the relationship between pores and compressive strength of foamed concrete, the effects of density, water-cement ratio, and foaming agent type on porosity, pore size distribution, and compressive strength were studied. It was found that as the density decreased and the water-cement ratio increased, the number of larger-sized pores increased and the pore size distribution became wider and more non-uniform. Further, the pore structure of lower-density foamed concrete was more susceptible to the influence of density and water-cement ratio more likely to produce larger size pores. Particularly, foaming agent type had a significant effect on the pore size distribution. The pore size variation of foamed concrete was primarily attributed to the initial size of fresh foam and the complex merging and escape phenomena occurring in the foam in the slurry, the process of preparation and forming. It was also found that smaller porosity and more concentrated and uniform pore size distribution were more beneficial for strength. Additionally, to quantify the combined effect of porosity and pore size distribution on compressive strength, this paper innovatively proposes the parameters of the “pore synthesis impact factor”. Finally, a new empirical model relating the porosity, size distribution, and compressive strength is presented.

Evaluation on Early Strength Development of Concrete Mixed with Non-Sintered Hwangto Using Ultrasonic Pulse Velocity.

Currently, in order to reduce the greenhouse gases of global warming, research on alternative cement materials is being actively conducted in the construction industry to reduce cement use, and it is judged to be important to evaluate the timing of form removal for the initial age. Therefore, in this study, we evaluated the initial mechanical properties of concrete in which cement was partially replaced with non-sintered hwangto (NHT). Specimens without NHT (namely, normal mortar (NM) and normal concrete (NC)) and specimens with NHT (namely, non-sintered hwangto mortar (HTM) and non-sintered hwangto concrete (HTC)) were prepared. NHT was substituted for 15% and 30% of cement. Two water-to-binder (W/B) ratios, 41% and 33%, were used to analyze the variation in the mechanical properties according to the cement and NHT content per unit volume of concrete. The compressive strength and ultrasonic pulse velocity (UPV) were measured. Experimental results indicated that compressive strength decreased with an increase in NHT content. The mortar with NHT substitution rates of 15% and 30% exhibited higher UPV than NM at a W/B ratio of 41%, in contrast to the behavior observed for concrete. The UPVs of most specimens were similar regardless of the NHT substitution rate. The correlation between the compressive strength and UPV of HTC was analyzed, and therefrom, exponential equations with a high correlation coefficient (R2) were proposed for strength prediction; the resulting predictions were compared with the results of previous compressive strength prediction models.

Design-focused Interpretable Machine Learning Models for Compressive Capacity Prediction of Gusset Plate Connections

Gusset plates can experience large compressive loads when used to connect bracing systems to beam-column joints in steel frames and truss bridges. The strength of a gusset plate is therefore vital to allow the bracing members to reach their ultimate capacity without a brittle failure of the gusset plate connection. Due to the large complexity of the stress distribution mechanism, simplified approaches commonly adopted for design can lead to inaccurate and highly conservative results. The present study aims at developing a simple yet efficient gusset plate compressive strength prediction model that does neither rely on crude assumptions nor require complicated calculations. A meticulously compiled database consisting of 68 experimental and 184 numerical datasets was deployed to aid in developing a high-performance machine learning (ML) model selected amongst 10 machine learning techniques studied. The superiority of the CatBoost and XGBoost models was confirmed with an accuracy of 95% and 96%, respectively significantly outperforming current design practices and widely used empirical formulas. SHapley Additive exPlanations (SHAP) and Taylor diagram were used to explain the feature importance and model performance. The developed interpretable ML models can instill trust in a model that explains its decisions, as opposed to other black box approaches. It was found that CatBoost was the most efficient in predicting the compressive strength of gusset plates with lowest mean absolute error of 50.76 kN and Nash-Sutcliffe model efficiency coefficient of 0.984, compared to the most accurate predictions observed among the existing methods (modified Thornton method) with mean absolute error and Nash-Sutcliffe model efficiency coefficient of 227.22 kN and 0.872, respectively. Finally, a reliability analysis was performed according to the LRFD approach to identify the level of safety of predictions obtained by the CatBoost and XGBoost models. The CatBoost model exhibited remarkably low dispersion with a safety factor of 0.753 thereby providing an adequate level of safety for the design of gusset plates.

Damage evolution and strength prediction model of soda residue modified cemented tailings backfill under uniaxial compression

Recycling of solid waste materials reduces filling costs and reduces environmental pollution. This paper analyzed the availability of soda residue modified cemented tailings backfill (SRMCTB), and carried out uniaxial compression experiments to investigate the mechanical properties, energy evolution and damage characteristics of SRMCTB. The results show that the addition of soda residue effectively improves the compressive properties of SRMCTB. The uniaxial compressive strength and elastic modulus of SRMCTB increase and then decrease, and the elastic energy storage limit and total energy storage limit of SRMCTB increase and then decrease with increasing soda residue content. The damage constitutive models considering the compaction stage of SRMCTB were established and the damage evolution curves were plotted. Combined with the damage curves, the damage of SRMCTB can be divided into damage-free stage, damage stable development stage and damage rapid growth stage. The addition of soda residue has an effect on the stable development stage of the damage. When the soda residue content increased, the rate of variation of damage development stages first increases and then decreases. The proportioning parameters of SRMCTB were optimized in combination with the response surface methodology. It shows that when the soda residue content is at 5.45 wt%, the mass concentration is 76 wt% and the tailings gradation is 2:5, SRMCTB has the highest uniaxial compressive strength. A prediction model for the uniaxial compressive strength of SRMCTB was proposed and validated. The average error between the experimental and theoretical values is 8.30%, indicating that the model has a strong applicability and degree of extension.

Study on calcium dissolution behavior of manufactured sand concrete

The accelerated calcium dissolution tests of concrete with different replacement rates of manufactured sand (ρms ) and different contents of stone powder (ρsp ) were carried out using the ammonium chloride solution. The results show that the dissolution resistance of manufactured sand concrete (MSC) is higher than that of ordinary concrete, although the ρms and ρsp influence its physical and mechanical properties. When the ρms is low, the mass loss ratio and dissolution depth of MSC after dissolution both increase with the increasing ρsp . They also increase with the increasing of ρms , but the ρsp no longer influences them when the ρms is high. The ρsp corresponding to the maximum ultrasonic velocity of MSC is different for the various ρms . The addition of manufactured sand inhibits the reduction of the compressive strength after dissolution, while stone powder has no significant impact on this. Finally, the fitting model of dissolution depth and prediction model of compressive strength of MSC after dissolution are established.

Hybrid random aggregation model and Bayesian optimization‐based convolutional neural network for estimating the concrete compressive strength

AbstractNumerous experimental studies have shown the type and gradation of coarse aggregates effect on the mechanical properties of concrete. The type and gradation of coarse aggregates have not been taken into account in the available machine learning prediction models. In this study, a two‐dimensional concrete microscopic image was generated by using a random aggregate model (RAM), and the coarse aggregate and other concrete ingredients were represented innovatively using polygons and trichromatic chromaticity values in the RAM images. The RAM image set was created by applying this method to represent 1110 sets of different concrete mixes. Then based on the Bayesian optimization algorithm and the image set, a compressive strength prediction model considering the effect of coarse aggregate types and gradations was developed utilizing a convolutional neural network (CNN) model. Meanwhile, an artificial neural network (ANN) compressive strength prediction model was developed using 1110 sets of mix ratio data. The results show that the proposed RAM image generation method has the capability to represent different concrete mix ratios collected in this study. The prediction performance of the CNN compressive strength model considering aggregate types and gradations is better than that of the ANN model. The method can provide a new perspective for predicting other concrete mechanical properties and technically support performance‐based intelligent concrete mix design.

Constitutive relationships and acoustic emission behavior of plant-fiber/magnesium oxychloride cement composites under uniaxial compressive load

Plant-fiber/magnesium oxychloride cement composites (PF-MOC) are bio-composite building materials that are produced by mixing magnesium oxychloride cement (MOC) with plant fibers. The application of PF-MOC in construction engineering can not only promote the reuse of bio-waste but also compensate for the shortage of wood. In this study, 33 prismatic specimens of PF-MOC were prepared using the extrusion molding method for the uniaxial compression test. The compression characteristics of PF-MOC under uniaxial compression load were studied by combining the test with acoustic emission (AE). The results illustrated that the apparent density of the PF-MOC ranged from 771 kg/m3 to 1806 kg/m3, and the compressive strength ranged from 4.7 MPa to 44 MPa. PF-MOC uniaxial compressive stress–strain relationship model was established. The compressive strength of PF-MOC with different types and content of plant fiber is closely related to the apparent density, and thus a compressive strength prediction model was developed. The PFC-MOC with bamboo scrap exhibited the highest AE energy near the peak load, indicating greater brittleness. Conversely, the PF-MOC with wheat straw demonstrated a well-distributed energy release throughout the compression process, suggesting lower brittleness in the straw specimens. The utilization of the b-value proved effective in characterizing the crack development and size transformation process of PF-MOC. Furthermore, SEM revealed that straw fibers exhibited the weakest bonding with the MOC matrix, while wood sawdust fibers and bamboo scrap showed superior interactions with the hydration products of MOC.

Characterization of thermal damage and compressive strength during drying at elevated temperatures using ultrasonic pulse velocity

This study attempts to visualize the thermal damage on a multiscale level and characterize the thermal damage and compressive strength of concrete using ultrasonic pulse velocity (UPV). First, based on the experimental results of mercury intrusion porosimetry (MIP), nitrogen adsorption, and scanning electron microscopy (SEM), the multiscale thermal damage of concrete subjected to elevated temperatures was visualized. Then, the UPV of the concrete during the entire drying process was detected. Consequently, the thermal damage can be characterized by the UPV. Finally, after considering the effect of both capillary pressure and thermal damage, a compressive strength prediction model, which could reflect the strength evolution during drying at elevated temperatures, was proposed. The model fits well with experimental results and provides an important basis for practical engineering.

Optimal machine learning-based method for gauging compressive strength of nanosilica-reinforced concrete

As nanotechnology developed, new materials emerged that might be employed to improve the mechanical properties of materials like concrete. Intensity factors are proportional to the kind of nanomaterial and its concentration. There is not yet a reliable prediction model for the uniaxial compressive strength (UCS) of nanomaterial-reinforced concrete. Access to such models is crucial for developing and evaluating nanomaterial-reinforced concrete structures. This study investigated the potential of eleven well-known machine learning (ML) algorithms to determine the most accurate and suitable ones to estimate the UCS of nanosilica-reinforced concrete. For this purpose, 460 data points were collected from experimental tests, including five input parameters and one target (UCS). The considered input parameters are the percentage of nanosilica in concrete cement (NS), sample diameter (D), sample length (L), porosity (n), and P-wave velocity (Vp). 80% of the data points were used for training and 20% for testing. The models were then verified using statistical analysis, and their behavior was compared to that in practice. Statistical analysis showed that all the models have achieved good accuracy compared to the experimental results, so the model with the lowest accuracy was the decision tree regressor (DTR) model, with a correlation coefficient of 0.68. However, comparing the behavior of the models with the practice mode by changing the value of one of the input parameters while keeping the values of other parameters constant showed that only the support vector regression (SVR) and null space SVR (NuSVR) models behave correctly. These two models were proposed as the most suitable for estimating the UCS of nanosilica-reinforced concrete. To further aid in the estimation of the UCS of nanosilica-reinforced concrete for engineering challenges, a graphical user interface (GUI) for the ML-based models was developed.

Grey relativity correlations between the pore structures and compressive strength of aeolian sand concrete undergoing carbonation and freeze-thaw cycles

In this study, the compressive strength of aeolian sand concrete (ASC) was predicted using pore structure parameters and pore structure types under the conditions of a complex environment. The relationships between the pore structures and the compressive strength of the ASC undergoing carbonation and freeze-thaw coupling actions were examined. Full ASC with a 100% replacement rate of aeolian sand (AS) was used in this study’s experiments. The microstructure characteristics of the ASC were analyzed using nuclear magnetic resonance (NMR) and scanning electron microscopy (SEM). At the same time, the correlation degrees of the pore structures and compressive strength were examined using a grey relativity analysis method. A GM (1, 3) prediction model of the ASC compressive strength under different carbonization times was established. The model showed that the effects of the carbonization and freeze-thaw cycles (FTCs) on the pore structures and the compressive strength of the ASC were different. According to the grey relativity analysis results, the bound fluid saturation and harmless pore ratios had the greatest impacts on the compressive strength of the ASC. The model established based on those factors had the ability to accurately predict the compressive strength of the ASC under the conditions of a carbonation-salt freezing and thawing environment.

Experimental Investigation on the Mix Proportion of Fiber-Reinforced Ultrahigh-Strength Grouting Material

Concrete tends to develop cracks once damaged, and wide cracks specifically would affect the whole structure negatively. When used for repairing wide cracks, traditional cement-based grouting materials could no longer meet the demand for tensile strength. The paper aims to develop a high-performance grouting material for crack repair and further investigate its fluidity, and compressive strength, flexural strength, and splitting tensile strength after hardening. Based on the design of ultrahigh-performance concrete (UHPC), this grouting material consists of cement, sand, water, superplasticizer, expansion agent, silica fume, fly ash, and flat copper-plated steel fiber. The initial fluidity and 30-min fluidity of the grouting material decreased with the addition of fly ash, silica fume, and flat copper-plated steel fiber, among which the silica fume had the greatest influence on the fluidity, followed by fly ash and steel fiber. The optimal content of flat copper-plated steel fiber, fly ash, and silica fume was 2%, 10%, and 15%, respectively. Furthermore, the prediction model of compressive strength of grouting material was established. It indicated that cement grade, water-cement ratio, curing age, silica fume, fly ash, and flat copper-plated steel fiber and can be used for the mix proportion design of grouting materials.

Mechanical properties and microscopic characteristics of steel fiber coal gangue concrete

Incorporation of coal gangue in the concrete mixes can realize utilization of the solid waste and reduce extraction/use of natural aggregates. To improve the mechanical properties of coal gangue concrete, this paper studies use of steel fibre together with coal gangue coarse aggregate, coal gangue fine aggregate/sand in various concrete mixes. The effect of volume dosages of steel fibre and different levels of replacing nature coarse aggregate and river sand with coal gangue aggregates on concrete compressive strength was first investigated. Then, a design of experiment using orthogonal test was adopted to study concrete mixes with 3 factors, namely, coal gangue coarse aggregate, coal gangue sand and steel fibre, and each at 3 levels. Through multidimensional statistical data analysis of the test results, the primary and secondary factors and the optimal composition of the steel fibre reinforced coal gangue concrete were identified, and a grey prediction model for compressive strength of the concrete mixes established. The microstructural characteristics and failure mechanism of steel fiber reinforced coal gangue concrete was also studied and discussed.