Application of deep learning in structural health management of concrete structures

Uwanuakwa, Ikenna D.; Idoko, John Bush; Mbadike, Elvis Michael; Reşatoğlu, Rifat; Alaneme, George Uwadiegwu

doi:10.1680/jbren.21.00063

Cited by 21 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The obtained computation results showed that there is statistical significance between the variables at a 95% confidence interval with a P value of 0.047, 0.023 and 0.013 for compressive, flexural, and splitting tensile strength respectively which is less than the critical value of 0.05. This analysis offers valuable insights for optimizing mix designs and enhancing the overall performance and durability of laterized concrete in various construction applications 82 .…”

Section: Results Discussion and Analysismentioning

confidence: 99%

Effects of aggregate sizes on the performance of laterized concrete

Ukpata,

Ewa,

Success

et al. 2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

Due to the high costs of traditional concrete materials in Nigeria, such as river sand, there is an increasing demand to explore alternative materials like laterite for fine aggregates. Although laterite is abundant in Nigeria, its full potential in the construction industry remains untapped. Previous studies have shown that partially replacing river sand with laterite produces concrete with competitive strength properties. This research aims to validate and extend these findings, evaluating the impact of different aggregate sizes (12 mm, 20 mm, and 40 mm) on the strength of concrete with 10% and 25% laterite replacements for fine aggregate. Results revealed that as the laterite percentage increased, compressive, flexural, and split tensile strengths decreased. While 0% and 10% laterite replacements met the required strength, the mix with 25% laterite fell short. Increasing maximum coarse aggregate size led to higher strengths, with 40 mm sizes exhibiting the highest, and 12 mm the lowest. Compressive strengths ranged from 22.1 to 37.6 N/mm2, flexural strengths from 4.07 to 5.99 N/mm2 and split-tensile strengths from 2.93 to 4.30 N/mm2. This research highlights the need for meticulous mix design adjustments when using laterite, balancing workability with strength objectives. The developed regression models offer a valuable tool for predicting concrete properties based on mix parameters, providing insights for optimizing laterized concrete designs across diverse construction applications and supporting sustainable building practices.

show abstract

Section: Results Discussion and Analysismentioning

confidence: 99%

Effects of aggregate sizes on the performance of laterized concrete

Ukpata,

Ewa,

Success

et al. 2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…The design of experiments constitutes a systematic assessment of the factor levels or component variable effects of the mixture in a simultaneous manner on the target response function, which is achieved using response surface methodology [ 36 ]. The deployment of this essential tool in laboratory experiments research helps to yield the minimization of cost and time resources by the generation of a maximum quantity of information for limited laboratory test trials.…”

Section: Materials and Methodologymentioning

confidence: 99%

Scheffe’s Simplex Optimization of Flexural Strength of Quarry Dust and Sawdust Ash Pervious Concrete for Sustainable Pavement Construction

et al. 2023

Self Cite

View full text Add to dashboard Cite

Pervious concrete provides a tailored surface course with high permeability properties which permit the easy flow of water through a larger interconnected porous structure to prevent flooding hazards. This paper reports the modeling of the flexural properties of quarry dust (QD) and sawdust ash (SDA) blended green pervious concrete for sustainable road pavement construction using Scheffe’s (5,2) optimization approach. The simplex mixture design method was adapted to formulate the mixture proportion to eliminate the set-backs encountered in empirical or trials and the error design approach, which consume more time and resources to design with experimental runs required to evaluate the response function. For the laboratory evaluation exercise, a maximum flexural strength of 3.703 N/mm2 was obtained with a mix proportion of 0.435:0.95:0.1:1.55:0.05 for water, cement, QD, coarse aggregate and SDA, respectively. Moreover, the minimal flexural strength response of 2.504 N/mm2 was obtained with a mix ratio of 0.6:0.75:0.3:4.1:0.25 for water, cement, QD, coarse aggregate and SDA, respectively. The test of the appropriateness of the developed model was statistically verified using the Student’ t-test and an analysis of variance (ANOVA), and was confirmed to be acceptable based on computational outcomes at the 95% confidence interval. Furthermore, the scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) were used to evaluate the morphological and mineralogical behavior of green prior concrete samples with various additive mixture compositions. The addition of QD and SDA, on the other hand, aided the creation of porous microstructures in the concrete matrix due to fabric changes in the concrete mixture, potentially aided by the formation of cementitious compounds such as calcium aluminate hydrate and calcium silicate hydrate.

show abstract

“…Multi-class classification has also been used to classify cracks based on their characteristics. Uwanuakwa et al [21] used DL to classify the type/cause of cracks as vertical, diagonal, shrinkage, efflorescence, alkali-silica reaction or corrosion cracks; And Gao and Mosalam [22] used multi-class classification to classify the level and type of damage in concrete structures.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Concrete Crack Detection and Measurement

Nyathi,

Bai,

Wilson

2024

Metrology

View full text Add to dashboard Cite

Concrete structures inevitably experience cracking, which is a common form of damage. If cracks are left undetected and allowed to worsen, catastrophic failures, with costly implications for human life and the economy, can occur. Traditional image processing techniques for crack detection and measurement have several limitations, which include complex parameter selection and restriction to measuring cracks in pixels, rather than more practical units of millimetres. This paper presents a three-stage approach that utilises deep learning and image processing for crack classification, segmentation and measurement. In the first two stages, custom CNN and U-Net models were employed for crack classification and segmentation. The final stage involved measuring crack width in millimetres by using a novel laser calibration method. The classification and segmentation models achieved 99.22% and 96.54% accuracy, respectively, while the mean absolute error observed for crack width measurement was 0.16 mm. The results demonstrate the adequacy of the developed crack detection and measurement method, and shows the developed deep learning and laser calibration method promotes safer, quicker inspections that are less prone to human error. The method’s ability to measure cracks in millimetres provides a more insightful assessment of structural damage, which is, in comparison to traditional pixel-based measurement methods, a significant improvement for practical field applications.

show abstract

Application of deep learning in structural health management of concrete structures

Cited by 21 publications

References 13 publications

Effects of aggregate sizes on the performance of laterized concrete

Effects of aggregate sizes on the performance of laterized concrete

Scheffe’s Simplex Optimization of Flexural Strength of Quarry Dust and Sawdust Ash Pervious Concrete for Sustainable Pavement Construction

Deep Learning for Concrete Crack Detection and Measurement

Contact Info

Product

Resources

About