Predicting the Compressive Strength and the Effective Porosity of Pervious Concrete Using Machine Learning Methods

Le, Ba-Anh; Hùng, Vũ Văn; Seo, Soo-Yeon; Tran, Bao-Viet; Nguyen‐Sy, Tuan; Le, Minh-Cuong; Vu, Thai-Son

doi:10.1007/s12205-022-1918-z

Cited by 25 publications

(7 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, the problem is also compared with predictions from the classic ANN model and the extreme gradient boosting method (XGB) for reference. More details about the XGB method can be found in works of Le et al [27,28]. It is worth noting that XGB is considered one of the best regression machine learning methods for tabular data, as reported in recent publications [29,30].…”

Section: Resultsmentioning

confidence: 99%

Deep artificial neural network-powered phase field model for predicting damage characteristic in brittle composite under varying configurations

Nguyen,

Le,

Tran

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

This work introduces a novel artificial neural network-powered phase field model, offering rapid and precise predictions of fracture propagation in brittle materials. To improve the capabilities of the ANN model, we incorporate a loop of conditions into its core to regulate the Absolute Percentage Error for each observation point, that filters and consistently selects the most accurate outcome. This algorithm enables our model to better adapt to the highly sensitive validation data arising from varying configurations. The effectiveness of the approach is illustrated through three examples involving changes in the microgeometry and material properties of steel fiber-reinforced high-strength concrete structures. Indeed, the predicted outcomes from the improved ANN phase field model in terms of stress-strain relationship, and crack propagation path demonstrates an outperformance compared with that based on the Extreme Gradient Boosting method (XGB), a leading regression machine learning technique for tabular data. Additionally, the introduced model exhibits a remarkable speed advantage, being 180 times faster than traditional phase field simulations, and provides results at nearly any fiber location, demonstrating superiority over the phase field model. This study marks a significant advancement in the application of artificial intelligence for accurately predicting crack propagation paths in composite materials, particularly in cases involving the relative positioning of the fiber and initial crack location.

show abstract

Section: Resultsmentioning

confidence: 99%

Deep artificial neural network-powered phase field model for predicting damage characteristic in brittle composite under varying configurations

Nguyen,

Le,

Tran

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…HYBRID MODEL In recent years, several analytical models [15] and artificial intelligence approaches [16] have been developed to predict the fundamental properties of porous concrete. In [17] a symbolic regression approach based on the genetic programming framework was utilized to assess the compressive strength of the porous concrete, achieving favorable results compared to purely analytical solutions or black-box machine learning models such as ANN and XGB [16]. However, symbolic regression models often suffer from drawbacks such as lack of physical significance and violations of fundamental physical laws.…”

Section: B Application To Pervious Concretementioning

confidence: 99%

“…It is worth mentioning again that the Pareto front consists of optimal points that are not dominated by any other solution in both accuracy and complexity. Table III introduces the two best-fitting equations obtained on the corresponding Pareto fronts, and the correlation coefficients with the two constructed datasets, and the independent validation dataset in [16]. It is apparent that the GP formula is slightly better than the hybrid model on the constructed dataset and displays no difference on the independent dataset.…”

Section: B Application To Pervious Concretementioning

confidence: 99%

A Genetic Programming-Assisted Analytical Formula for Predicting the Permeability of Pervious Concrete

Le,

Vu,

Nguyen

et al. 2024

Eng. Technol. Appl. Sci. Res.

Self Cite

View full text Add to dashboard Cite

This study proposes a new approach to construct predictive formulas for the permeability of Pervious Concrete (PC), which depends on PC mixture and porosity. To achieve this, a dataset of 195 samples collected from different sources was used. In the dataset the permeability is dependent on porosity, aggregate-to-cement ratio (AC), maximum nominal sizes (MS) of coarse aggregate, and water-to-cement or binder ratios (WC). From the dataset and through applying simple regression techniques, several analytical functions based on the Kozeny-Carman model were constructed and evaluated for their effectiveness in implementing independent datasets and similar analytical functions. Furthermore, for the first time, the Genetic Programming-based Symbolic Regression method was adopted to construct hybrid models combined with the Kozeny-Carman analytical model. The equation of the hybrid model ensures both basic physical conditions and efficiency while being simple enough for engineering-level applications.

show abstract

“…The permeability was also affected by the pore characteristics, such as pore size and pore composition. [26][27][28][29] Therefore, there is a theoretical method to enhance the permeability and strength of pervious concrete by optimizing the pore structure. By this method, the pervious concrete gains a reduced porosity (improve strength) and optimized pore structure (maintain/improve permeability).…”

Section: Introductionmentioning

confidence: 99%

“…The previous research had found that porosity was not the only factor influencing permeability of pervious concrete. The permeability was also affected by the pore characteristics, such as pore size and pore composition 26–29 . Therefore, there is a theoretical method to enhance the permeability and strength of pervious concrete by optimizing the pore structure.…”

Section: Introductionmentioning

confidence: 99%

Enhancing both strength and permeability of pervious concrete by optimizing pore structure: An experimental study

Guo

et al. 2023

Structural Concrete

View full text Add to dashboard Cite

The large porosity enables pervious concrete to obtain water permeability, but seriously decreases its strength. Hence, how to coordinate the contradiction between strength and permeability is a critical challenge for pervious concrete. This research was mainly aimed at mix design of pervious concrete for enhancing both strength and permeability by optimizing the pore structure. Six groups of pervious concrete with 15% target porosity were prepared by 10–15 mm coarse aggregate and different fine aggregate (natural sand, basalt aggregate, and steel slag). Moreover, a group of pervious concrete with 20% target porosity as control samples was prepared by 4.75–9.5 mm aggregate. The compressive strength and permeability coefficient of these samples were tested. Based on CT scanning and image analysis, the influence of pore structure optimization on permeability and compressive strength was analyzed. The results show that though the porosity is reduced by 5% after pore structure optimization, the porosity of large pores and medium pores changes little, the porosity of small pores decreased significantly, and finally the average pore size is still increased. This can explain why the experimental groups gain a similar/better permeability as the control group. The experimental groups of cement or steel slag have an increased compressive strength compared to the control group, which is due to the decrease of porosity and “small pores” content, as well as more abundant hydration products.

show abstract

Predicting the Compressive Strength and the Effective Porosity of Pervious Concrete Using Machine Learning Methods

Cited by 25 publications

References 65 publications

Deep artificial neural network-powered phase field model for predicting damage characteristic in brittle composite under varying configurations

Deep artificial neural network-powered phase field model for predicting damage characteristic in brittle composite under varying configurations

A Genetic Programming-Assisted Analytical Formula for Predicting the Permeability of Pervious Concrete

Enhancing both strength and permeability of pervious concrete by optimizing pore structure: An experimental study

Contact Info

Product

Resources

About