Prediction of Chloride Diffusion Coefficient in Concrete Based on Machine Learning and Virtual Sample Algorithm
Abstract
:1. Introduction
2. Machine Learning Methods
2.1. MLP
2.2. SVM
2.3. GMM-VSG
3. Data Collection and Performance Evaluation
3.1. Database
3.1.1. Data Sources
3.1.2. Input Variable Selection
3.1.3. Statistical Characteristics of Data
3.1.4. Variable Correlation
3.2. Data Preprocessing
3.3. Partition of Training and Testing Set
3.4. Performance Evaluation
4. Results and Discussion
4.1. MLP Model
4.2. SVM Model
4.3. Comparisons
5. Conclusions and Prospects
5.1. Conclusions
- The connection between macroscopic properties and microstructure of concrete can be assessed by machine learning methods.
- The MLP and SVM models built by the virtual database are capable of predicting the chloride diffusion coefficients in concrete. The R2 obtained from MLP and SVM models is 0.95 (by normalization) and 0.97 (by standardization), respectively. The OBJ value obtained by the MLP and SVM model is 0.68 (by normalization) and 0.30 (by standardization), respectively.
- The expended data set produced by GMM-VSG algorithm can help to improve the accuracy of MLP and SVM models, and the improvement is greater for SVM model, probably because the increase in the amount of data weakens the impact of noise values.
- The normalization preprocessing method is more suitable to the MLP model, while the standardization preprocessing method is adapted to the SVM model. This may be due to the fact that normalization preserves sample spacing better, and normalization prevents oversaturation of neurons. The normalization preprocessing method is also more robust than the standardization preprocessing method.
5.2. Prospects
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Lin, S.Z.; Li, H.M.; Zhang, H.Y. Experimental study on frost resistance durability and service life prediction of normal cement concrete. Adv. Mater. Res. 2011, 368–373, 2425–2429. [Google Scholar] [CrossRef]
- Chen, X.; Li, Z.Y.; Jin, W.L.; Zhang, Y.; Yao, C.J. Chloride ion ingress distribution within an alternate wetting-drying marine environment area. Sci. China Technol. Sci. 2012, 55, 970–996. [Google Scholar]
- Basheer, L.; Kropp, J.; Cleland, D.J. Assessment of the durability of concrete from its permeation properties: A review. Constr. Build. Mater. 2001, 15, 93–103. [Google Scholar] [CrossRef]
- Evangelista, L.; Brito, J.D. Durability performance of concrete made with fine recycled concrete aggregates. Cem. Concr. Compos. 2010, 32, 9–14. [Google Scholar] [CrossRef]
- Dunker, K.F.; Rabbat, B.G. Why America’s bridges are crumbling. Sci. Am. 1993, 266, 66–72. [Google Scholar] [CrossRef]
- Hong, N.F. Development and difficulty of forecast for corrosion and durability of concrete. Concrete 2006, 10, 10–12+16. (In Chinese) [Google Scholar]
- Zhang, M.H.; Li, H. Pore structure and chloride permeability of concrete containing nano-particles for pavement. Constr. Build. Mater. 2011, 25, 608–616. [Google Scholar] [CrossRef]
- Moon, H.Y.; Kim, H.S. Relationship between average pore diameter and chloride diffusivity in various concretes. Constr. Build. Mater. 2006, 20, 725–732. [Google Scholar] [CrossRef]
- Luo, R.; Cai, Y.B.; Wang, C.Y.; Huang, X. Study of chloride binding and diffusion in GGBS concrete. Cem. Concr. Res. 2003, 33, 1–7. [Google Scholar] [CrossRef]
- Jin, L.; Yu, G.; Wang, Z.; Fan, T. Developing a model for chloride transport through concrete considering the key factors. Case Stud. Constr. Mater. 2022, 17, e01168. [Google Scholar] [CrossRef]
- Qi, D.; Zheng, H.; Zhang, L.; Sun, G.; Yang, H.; Li, Y. Numerical simulation on diffusion reaction behavior of concrete under sulfate chloride coupled attack. Constr. Build. Mater. 2023, 405, 133237. [Google Scholar] [CrossRef]
- Shazali, M.A.; Rahman, M.K.; Al-Gadhib, A.H.; Balunch, M.H. Transport modeling of chlorides with binding in concrete. Arab. J. Sci. Eng. 2012, 37, 469–479. [Google Scholar] [CrossRef]
- Weiss, T.; Mareš, J.; Slavík, M.; Bruthans, J. A microdestructive method using dye-coated-probe to visualize capillary, diffusion and evaporation zones in porous materials. Sci. Total Environ. 2020, 704, 135339. [Google Scholar] [CrossRef] [PubMed]
- Wernert, V.; Nguyen, K.L.; Levitz, P.; Coasne, B.; Denoyel, R. Impact of surface diffusion on transport through porous materials. J. Chromatogr. A 2022, 1665, 462823. [Google Scholar] [CrossRef] [PubMed]
- Gao, Y.W.; Pastrana, A.P.C.; Manogharan, G.; van Duin, A.C.T. Molecular dynamics study of melting, diffusion, and sintering of cementite chromia core–shell particles. Comput. Mater. Sci. 2021, 199, 110721. [Google Scholar] [CrossRef]
- Cai, R.; Han, T.H.; Liao, W.Y.; Huang, J.; Li, D.W.; Kumar, A.; Ma, H.Y. Prediction of surface chloride concentration of marine concrete using ensemble machine learning. Cem. Concr. Res. 2020, 136, 106164. [Google Scholar] [CrossRef]
- Liu, Q.F.; Iqbal, M.F.; Yang, J.; Lu, X.Y.; Zhang, P.; Rauf, M. Prediction of chloride diffusivity in concrete using artificial neural network: Modelling and performance evaluation. Constr. Build. Mater. 2021, 268, 121082. [Google Scholar] [CrossRef]
- Tran, V.Q. Machine learning approach for investigating chloride diffusion coefficient of concrete containing supplementary cementitious materials. Constr. Build. Mater. 2022, 328, 127103. [Google Scholar] [CrossRef]
- Liu, Y.; Cao, Y.; Wang, L.; Chen, Z.S.; Qin, Y. Prediction of the durability of high-performance concrete using an integrated RF-LSSVM model. Constr. Build. Mater. 2022, 356, 129232. [Google Scholar] [CrossRef]
- Jin, L.; Dong, T.; Fan, T.; Duan, J.; Yu, H.L.; Jiao, P.F.; Zhang, W.B. Prediction of the chloride diffusivity of recycled aggregate concrete using artificial neural network. Mater. Today Commun. 2022, 32, 104137. [Google Scholar] [CrossRef]
- Polyzotis, N.; Roy, S.; Whang, S.E.; Zinkevich, M. Data management challenges in production machine learning. In Proceedings of the 2017 ACM International Conference on Management of Data, Chicago, IL, USA, 14–19 May 2017; pp. 1723–1726. [Google Scholar]
- Wang, J.C. A neural network initialization method based on machine learning. Comput. Res. Dev. 1997, 8, 41–46. (In Chinese) [Google Scholar]
- Niu, X.X.; Yang, C.L.; Wang, H.C.; Wang, Y.Y. Investigation of ANN and SVM based on limited samples for performance and emissions prediction of a CRDI-assisted marine diesel engine. Appl. Therm. Eng. 2017, 111, 1353–1364. [Google Scholar] [CrossRef]
- Ping, Y. Support Vector Machine Based Clustering and Text Classification Research. Ph.D. Thesis, Beijing University of Posts and Telecommunications, Beijing, China, 2012. (In Chinese). [Google Scholar]
- Akande, K.O.; Owolabi, T.O.; Twaha, S.; Olatunji, S.O. Performance comparison of SVM and ANN in predicting compressive strength of concrete. IOSR J. Comput. Eng. 2014, 16, 88–94. [Google Scholar] [CrossRef]
- Yang, J.; Yu, X.; Xie, Z.; Zhang, J. A novel virtual sample generation method based on Gaussian distribution. Knowl. Based Syst. 2011, 24, 740–748. [Google Scholar] [CrossRef]
- Gong, H.; Chen, Z.; Zhu, Q.; He, Y. A Monte Carlo and PSO based virtual sample generation method for enhancing the energy prediction and energy optimization on small data problem: An empirical study of petrochemical industries. Appl. Energy 2017, 197, 405–415. [Google Scholar] [CrossRef]
- Dong, Q.Y.; Bai, S.W.; Wang, Z.; Zhao, X.Y.; Yang, S.S.; Ren, N.Q. Virtual sample generation empowers machine learning-based effluent prediction in constructed wetlands. J. Environ. Manag. 2023, 346, 118961. [Google Scholar] [CrossRef] [PubMed]
- Sang, K.H.; Yin, X.Y.; Zhang, F.C. Machine learning seismic reservoir prediction method based on virtual sample generation. Pet. Sci. 2021, 18, 1662–1674. [Google Scholar] [CrossRef]
- Lin, L.S.; Lin, Y.S.; Li, D.C.; Liu, Y.H. Improved learning performance for small datasets in high dimensions by new dual-net model for non-linear interpolation virtual sample generation. Decis. Support Syst. 2023, 172, 113996. [Google Scholar] [CrossRef]
- Zhang, X.Q. Study on the Mechanism of Chloride Ion Diffusion in Double-Doped Fly Ash and Slag Concrete. Ph.D. Thesis, Nanjing University of Technology, Nanjing, China, 2015. (In Chinese). [Google Scholar]
- Zhang, Y.R.; Wu, S.Y.; Ma, X.Q.; Fang, L.C.; Zhang, J.Z. Effects of additives on water permeability and chloride diffusivity of concrete under marine tidal environment. Constr. Build. Mater. 2022, 320, 126217. [Google Scholar] [CrossRef]
- Zhang, J.Z.; Fang, R.H.; Lv, M.; Zhang, Y.R.; Cao, Y.H.; Gao, Y.H. Time dependent microstructure evolution of fly ash concrete in the natural tidal environment. J. Nat. Disasters 2019, 28, 9–16. (In Chinese) [Google Scholar]
- Zhang, J.Z.; Jin, T.; He, Y.C.; Yu, W.L.; Gao, Y.H.; Zhang, Y.R. Time dependent correlation of permeability of fly ash concrete under natural tidal environment. Eur. J. Environ. Civ. Eng. 2022, 26, 8477–8501. [Google Scholar] [CrossRef]
- Zhang, J.Z.; Zhou, X.Y.; Zhao, J.; Wang, M.; Gao, Y.H.; Zhang, Y.R. Similarity of chloride diffusivity of concrete exposed to different environments. ACI Mater. J. 2020, 117, 27–37. [Google Scholar]
- Zhang, J.Z.; Wang, M.; Zhou, X.Y.; Yu, W.L.; Gao, Y.H.; Zhang, Y.R. Exploring the emerging evolution trends of probabilistic service life prediction of reinforced concrete structures in the chloride environment by scientometric analysis. Adv. Civ. Eng. 2021, 2021, 8883142. [Google Scholar] [CrossRef]
- Gao, Y.H.; Shao, X.J.; Zhang, Y.R.; Fang, R.H.; Zhang, J.Z. Permeability dependency of fly ash concrete in natural tidal environment. J. Hydroelectr. Eng. 2021, 40, 214–222. (In Chinese) [Google Scholar]
- Gao, Y.H.; Guo, B.L.; Wang, M.; Zhang, Y.R.; Zhang, J.Z. Stable time and mechanism of concrete permeability in natural tidal environment. J. Hydroelectr. Eng. 2022, 41, 50–62. (In Chinese) [Google Scholar]
- Zhang, J.Z.; Wu, J.; Zhang, Y.R.; Wang, J.D. Time-varying relationship between pore structures and chloride diffusivity of concrete under the simulated tidal environment. Eur. J. Environ. Civ. Eng. 2022, 26, 501–518. [Google Scholar] [CrossRef]
- Shen, L.J.; Qian, Q. A virtual sample generation algorithm supporting machine learning with a small-sample dataset: A case study for rubber materials. Comput. Mater. Sci. 2022, 211, 111475. [Google Scholar] [CrossRef]
- Rosenblatt, F. The perceptron: A probabilistic model for information storage and organization in the brain. Psychol. Rev. 1958, 65, 386–408. [Google Scholar] [CrossRef]
- Cortes, C.; Vapnik, V. Support-Vector Networks. Mach. Learn. 1995, 20, 273–297. [Google Scholar] [CrossRef]
- Wakjira, T.G.; Rahmzadeh, A.; Alam, M.S.; Tremblay, R. Explainable machine learning based efficient prediction tool for lateral cyclic response of post-tensioned base rocking steel bridge piers. Structures 2022, 44, 947–964. [Google Scholar] [CrossRef]
- Wakjira, T.G.; Abushanab, A.; Ebead, U.; Alnahhal, W. FAI: Fast, accurate, and intelligent approach and prediction tool for flexural capacity of FRP-RC beams based on super-learner machine learning model. Mater. Today Commun. 2022, 33, 104461. [Google Scholar] [CrossRef]
- Zhang, Y.R.; Yu, W.L.; Ma, X.Q.; Luo, T.Y.; Wang, J.J. Prediction of chloride concentration in fly ash concrete based on deep learning. J. Beijing Univ. Technol. 2023, 49, 205–212. (In Chinese) [Google Scholar]
- Adıgüzel, E.; Subaşı, N.; Mumc, T.V.; Ersoy, A. The effect of the marble dust to the efficiency of photovoltaic panels efficiency by SVM. Energy Rep. 2023, 9, 66–76. [Google Scholar] [CrossRef]
- Santos, C.E.d.S.; Sampaio, R.C.; Coelho, L.d.S.; Bestarsd, G.A.; Llanos, C.H. Multi-objective adaptive differential evolution for SVM/SVR hyperparameters selection. Pattern Recognit. 2021, 10, 107649. [Google Scholar] [CrossRef]
- Wang, L.L.; Yu, W.L.; Zhang, Y.R. Prediction of chloride concentration in fly ash concrete based on support vector regression. Zhejiang Architecture, (In Chinese, accepted.).
- Liu, X.T. BP neural network input layer data normalization study. Mech. Eng. Autom. 2010, 3, 122–126. (In Chinese) [Google Scholar]
- Vu, H.L.; Ng, K.T.W.; Richter, A.; An, C. Analysis of input set characteristics and variances on k-fold cross validation for a Recurrent Neural Network model on waste disposal rate estimation. J. Environ. Manag. 2022, 311, 114869. [Google Scholar] [CrossRef] [PubMed]
- Nguyen, X.C.; Nguyen, T.T.H.; La, D.D.; Kumar, G.; Rene, E.R.; Nguyen, D.D.; Chang, S.W.; Chung, W.J.; Nguyen, X.H.; Nguyen, V.K. Development of machine learning–based models to forecast solid waste generation in residential areas: A case study from Vietnam. Resour. Conserv. Recycl. 2021, 167, 105381. [Google Scholar] [CrossRef]
- Garre, A.; Ruiz, M.C.; Hontoria, E. Application of Machine Learning to support production planning of a food industry in the context of waste generation under uncertainty. Oper. Res. Perspect. 2020, 7, 100147. [Google Scholar] [CrossRef]
- Golafshani, E.M.; Behnood, A.; Arashpour, M. Predicting the compressive strength of normal and High-Performance Concretes using ANN and ANFIS hybridized with Grey Wolf Optimizer. Constr. Build. Mater. 2020, 232, 117266. [Google Scholar] [CrossRef]
Statistical Parameter | Porosity (%) | Contribution Porosity (%) | D (10−12m2/s) | |||
---|---|---|---|---|---|---|
<20 nm | 20–50 nm | 50–200 nm | >200 nm | |||
Maximum | 10.85 | 3.10 | 2.87 | 5.26 | 4.01 | 10.04 |
Minimum | 3.19 | 0.91 | 0.23 | 0.20 | 0.21 | 0.41 |
Mean | 5.47 | 1.61 | 1.61 | 1.41 | 0.83 | 2.99 |
Median | 5.23 | 1.63 | 1.57 | 1.14 | 0.65 | 2.27 |
Standard Error | 1.41 | 0.42 | 0.36 | 0.92 | 0.65 | 2.22 |
Kurtosis | 1.77 | 0.43 | 2.29 | 3.25 | 6.75 | 0.33 |
Skewness | 1.11 | 0.44 | −0.02 | 1.64 | 2.36 | 1.04 |
Statistical Parameter | Porosity (%) | Contribution Porosity (%) | D (10−12m2/s) | |||
---|---|---|---|---|---|---|
<20 nm | 20–50 nm | 50–200 nm | >200 nm | |||
Maximum | 18.49 | 4.72 | 8.21 | 6.97 | 5.18 | 14.55 |
Minimum | 3.19 | 0.38 | 0.23 | 0.08 | 0.00 | 0.41 |
Mean | 7.02 | 1.91 | 2.42 | 1.79 | 0.90 | 4.28 |
Median | 5.74 | 1.77 | 1.70 | 1.35 | 0.65 | 3.20 |
Standard Error | 3.49 | 0.73 | 1.61 | 1.42 | 0.76 | 3.42 |
Kurtosis | 1.07 | 2.14 | 0.96 | 1.94 | 6.94 | −0.05 |
Skewness | 1.34 | 1.28 | 1.39 | 1.52 | 2.27 | 0.94 |
Statistical Parameter | Porosity (%) | Contribution Porosity (%) | D (10−12m2/s) | |||
---|---|---|---|---|---|---|
<20 nm | 20–50 nm | 50–200 nm | >200 nm | |||
Maximum | 18.49 | 4.72 | 8.21 | 6.97 | 5.18 | 14.55 |
Minimum | 2.89 | 0.38 | 0.23 | 0.08 | 0.00 | 0.41 |
Mean | 7.45 | 1.95 | 2.58 | 1.96 | 0.96 | 4.67 |
Median | 6.07 | 1.77 | 1.75 | 1.52 | 0.71 | 3.75 |
Standard Error | 3.58 | 0.69 | 1.61 | 1.49 | 0.76 | 3.52 |
Kurtosis | 0.33 | 1.86 | 0.20 | 1.54 | 5.65 | −0.51 |
Skewness | 1.12 | 1.22 | 1.15 | 1.41 | 2.03 | 0.74 |
Evaluation Indicator | First Group | Second Group | Third Group | |||
---|---|---|---|---|---|---|
Normalization | Standardization | Normalization | Standardization | Normalization | Standardization | |
MAE | 1.07 | 1.03 | 1.11 | 1.10 | 0.62 | 0.66 |
MSE | 1.87 | 1.75 | 1.67 | 1.86 | 0.71 | 0.73 |
R2 | 0.62 | 0.65 | 0.71 | 0.77 | 0.95 | 0.95 |
OBJ | 1.81 | 1.69 | 1.54 | 1.67 | 0.68 | 0.71 |
Evaluation Indicator | First Group | Second Group | Third Group | |||
---|---|---|---|---|---|---|
Normalization | Standardization | Normalization | Standardization | Normalization | Standardization | |
MAE | 0.70 | 1.16 | 1.03 | 1.37 | 0.59 | 0.29 |
MSE | 1.07 | 1.93 | 3.00 | 3.69 | 0.81 | 0.30 |
R2 | 0.76 | 0.57 | 0.68 | 0.60 | 0.93 | 0.97 |
OBJ | 1.00 | 1.97 | 2.41 | 3.61 | 0.73 | 0.30 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhou, F.-Y.; Tao, N.-J.; Zhang, Y.-R.; Yuan, W.-B. Prediction of Chloride Diffusion Coefficient in Concrete Based on Machine Learning and Virtual Sample Algorithm. Sustainability 2023, 15, 16896. https://doi.org/10.3390/su152416896
Zhou F-Y, Tao N-J, Zhang Y-R, Yuan W-B. Prediction of Chloride Diffusion Coefficient in Concrete Based on Machine Learning and Virtual Sample Algorithm. Sustainability. 2023; 15(24):16896. https://doi.org/10.3390/su152416896
Chicago/Turabian StyleZhou, Fei-Yu, Ning-Jing Tao, Yu-Rong Zhang, and Wei-Bin Yuan. 2023. "Prediction of Chloride Diffusion Coefficient in Concrete Based on Machine Learning and Virtual Sample Algorithm" Sustainability 15, no. 24: 16896. https://doi.org/10.3390/su152416896