A multi-layer approach to classify the risk of corrosion in concrete specimens that contain different additives
Source abstract
Abstract
There are many different ways to estimate whether a concrete structure will be suffering corrosion degradation as experimental processes and numerical estimations relied on theoretical equations. Nevertheless, these estimations require higher certainty and satisfactory estimation accuracy. Deep Learning is causing profound changes in society. It has inspired unprecedented advances in historically challenging problems. This paper proposes a mathematician approach. A Deep Learning model capable of classifying the risk of corrosion on concrete specimens relied on a standard methodology. It employs the Electrical resistivity measurement for achieving this goal. The deep neural network architecture presented in this study displays a merging of activation functions. Also, a regularized method to avoid overfitting problems was implemented. The model comprises different standard tests as input data and classification of the corrosion risk as output. The outcomes depict an accuracy of about 98%, given the nature of the stochastic learning algorithm. Different concrete samples were analyzed in this research in order to provide data variety to the algorithm and accomplish a robust and stable model.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Concrete Corrosion and Durability · Infrastructure Maintenance and Monitoring · Non-Destructive Testing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (50)
Show source-provided references
1. Concrete durability enhancement from nopal (Opuntia ficus-indica) additions10.1016/j.conbuildmat.2020.118170 ↗
2. El comportamiento del pulso ultrasónico en un concreto de alto desempeño adicionado con un polímero orgánico comparado con un concreto sin adiciones
3. effects of corrosion inhibiting admixtures and supplementary cementitious materials combinations on the strength and certain durability properties of hpc10.1139/cjce-2016-0237 ↗
4. What makes data science different? a discussion involving statistics2. 0 and computational sciences10.1007/s41060-017-0090-x ↗
5. Deep transfer learning for image-based structural damage recognition10.1111/mice.12363 ↗
6. A novel machine learning-based algorithm to detect damage in high-rise building structures10.1002/tal.1400 ↗
7. A new type of neurons for machine learning10.1002/cnm.2920 ↗
8. An approach to dynamical classification of daily traffic patterns10.1111/mice.12226 ↗
9. A novel unsupervised deep learning model for global and local health condition assessment of structures10.1016/j.engstruct.2017.10.070 ↗
10. Deep learning for accelerated seismic reliability analysis of transportation networks10.1111/mice.12359 ↗
11. Computational design optimization of concrete mixtures: a review10.1016/j.cemconres.2018.04.007 ↗
12. Neural network, machine learning, and evolutionary approaches for concrete material characterization
13. O.Troconis de Rincón y Miembros. de la red durar.Red Temática XV. B. Durabilidad de la Armadura. Manual De Inspección, Evaluación y Diagnóstico de Corrosión en Estructuras de Hormigón Armado, CYTED Maracaibo. Venezuela, 2007.
14. Long-term monitoring of reinforcement corrosion in concrete using ground penetrating radar10.1016/j.corsci.2016.11.003 ↗
15. Recurrent neural network for approximate earthquake time and location prediction using multiple seismicity indicators10.1111/j.1467-8667.2009.00595.x ↗
16. A modified firefly algorithm-artificial neural network expert system for predicting compressive and tensile strength of high-performance concrete10.1016/j.conbuildmat.2018.05.201 ↗
17. Prajit Ramachandran, Barret Zoph, and Quoc V Le. Searching for activation functions. arXiv:〈http://arXiv.org/abs/arXiv:1710.05941〉, 2017.
18. Compressive strength prediction of environmentally friendly concrete using artificial neural networks10.1016/j.jobe.2018.01.007 ↗
19. Prediction of electrical conductivity of fiber-reinforced cement-based composites by deep neural networks10.3390/ma12233868 ↗
20. Compressive strength prediction of recycled concrete based on deep learning10.1016/j.conbuildmat.2018.04.169 ↗
21. Deep learning-based automatic volumetric damage quantification using depth camera10.1016/j.autcon.2018.12.006 ↗
22. Autonomous concrete crack detection using deep fully convolutional neural network10.1016/j.autcon.2018.11.028 ↗
23. American Concrete Institute. Committee 211.Standard practice for selecting proportions for normal, heavyweight, and mass concrete:(aci 211.1–91). American Concrete Institute, 1991.
24. Modeling of strength of high-performance concrete using artificial neural networks10.1016/S0008-8846(98)00165-3 ↗
25. Prediction of the compressive strength of normal and high-performance concretes using m5p model tree algorithm10.1016/j.conbuildmat.2017.03.061 ↗
26. Prediction of compressive strength of scc and hpc with high volume fly ash using ann10.1016/j.conbuildmat.2008.01.014 ↗
27. Enhanced artificial intelligence for ensemble approach to predicting high performance concrete compressive strength10.1016/j.conbuildmat.2013.08.078 ↗
28. ASTM International Committee C09 on Concrete and Concrete Aggregates. Standard test method for compressive strength of cylindrical concrete specimens. ASTM International, 2014.
29. C78 ASTM et al.Standard test method for flexural strength of concrete (using simple beam with third-point loading). In American Society for Testing and Materials, 100, 19428–2959, 2010.
30. Prediction of compressive strength using ultrasonic pulse velocity for clsm with waste lcd glass concrete
31. Advancing concrete strength prediction using non-destructive testing: Development and verification of a generalizable model10.1016/j.conbuildmat.2015.10.131 ↗
32. A neural network approach for compressive strength prediction in cement-based materials through the study of pressure-stimulated electrical signals10.1016/j.conbuildmat.2011.11.036 ↗
33. In-situ strength assessment of concrete: detailed guidelines
34. How investigators can assess concrete strength with on-site non-destructive tests and lessons to draw from a benchmark
35. corrosion and protection of metals in contact with concrete
36. Learning overparameterized neural networks via stochastic gradient descent on structured data
37. Difan Zou, Yuan Cao, Dongruo Zhou, and Quanquan Gu. Stochastic gradient descent optimizes over-parameterized deep relu networks. arXiv: 〈http://arXiv.org/abs/arXiv:1811.08888〉, 2018.10.1007/s10994-019-05839-6 ↗
38. Ilya Loshchilov and Frank Hutter. Sgdr: Stochastic gradient descent with warm restarts. arXiv: 〈http://arXiv.org/abs/arXiv:1608.03983〉, 2016.
39. Using stochastic gradient decent algorithm for incremental matrix factorization in recommendation system
40. Goodfellow · 2016
41. Background information of deep learning for structural engineering10.1007/s11831-017-9237-0 ↗
42. Igual · 2017 · 1
43. Michael A. Nielsen.Neural networks and deep learning, volume 2018. Determination press San Francisco, CA, USA:, 2015.
44. Aggarwal · 2018 · 978
45. Modelling the strength of lightweight foamed concrete using support vector machine (svm)
46. New formulation of compressive strength of preformed-foam cellular concrete: an evolutionary approach10.1061/(ASCE)MT.1943-5533.0001602 ↗
47. JA Guzmán Torres. Concreto de alta durabilidad y alto desempeño en infraestructura. PhD thesis, Tesis de Maestría, Fac. Ing. Civil Universidad Michoacana de San Nicolás de., 2014.
48. Predicting compressive strength of lightweight foamed concrete using extreme learning machine model10.1016/j.advengsoft.2017.09.004 ↗
49. Deep neural architectures for mapping scalp to intracranial eeg10.1142/S0129065718500090 ↗
50. Bonamente · 2017