Sequential learning to accelerate discovery of alkali-activated binders
Source abstract
Abstract
Abstract Alkali-activated binders (AAB) can provide a clean alternative to conventional cement in terms of CO2emissions. However, as yet there are no sufficiently accurate material models to effectively predict the AAB properties, thus making optimal mix design highly costly and reducing the attractiveness of such binders. This work adopts sequential learning (SL) in high-dimensional material spaces (consisting of composition and processing data) to find AABs that exhibit desired properties. The SL approach combines machine learning models and feedback from real experiments. For this purpose, 131 data points were collected from different publications. The data sources are described in detail, and the differences between the binders are discussed. The sought-after target property is the compressive strength of the binders after 28 days. The success is benchmarked in terms of the number of experiments required to find materials with the desired strength. The influence of some constraints was systematically analyzed, e.g., the possibility to parallelize the experiments, the influence of the chosen algorithm and the size of the training data set. The results show the advantage of SL, i.e., the amount of data required can potentially be reduced by at least one order of magnitude compared to traditional machine learning models, while at the same time exploiting highly complex information. This brings applications in laboratory practice within reach.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Concrete and Cement Materials Research · Concrete Corrosion and Durability · Recycled Aggregate Concrete Performance
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (72)
Show source-provided references
1. Hatfield, Ashley (2021) "National Minerals Information Center," [Online]. Available: https://www.usgs.gov/centers/nmic/cement-statistics-and-information. [Accessed 20 04 2021].
2. Andrew RM (2018) "Global CO2 emissions from cement production," Earth System Science Data, p. 195–217.10.5194/essd-10-195-2018 ↗
3. Dell R (2021) "Making the Concrete and Steel We Need Doesn’t Have to Bake the Planet," [Online]. Available: https://www.nytimes.com/2021/03/04/opinion/climate-change-infrastructure.html . [Accessed 29. March 2021].
4. World Business Council for Sustainable Development, Cement Sustainability Initiative. , "Technology Roadmap - Low-Carbon Transition in the Cement Industry," International Energy Agency, France, 2018.
5. Provis JL, van Deventer J (2014) Alkali Activated Materials, State-of-the-Art Report, RILEM TC 224-AAM. Springer, Dordrecht, Heidelberg, New York, London10.1007/978-94-007-7672-2 ↗
6. Shobeiri V, Bennett B, Xie T and Visintin P (2021) "A comprehensive assessment of the global warming potential of geopolymer concrete," J Cleaner Production, 297, p. https://doi.org/10.1016/j.jclepro.2021.126669.10.1016/j.jclepro.2021.126669 ↗
7. Firdous R, Stephan D, Djobo JNY (2018) Natural pozzolan based geopolymers: A review on mechanical, microstructural and durability characteristics. Constr Build Mater 190:1251–1263. https://doi.org/10.1016/j.conbuildmat.2018.09.191,27September10.1016/j.conbuildmat.2018.09.191,27September ↗
8. Firdous R and Stephan D (2019) "Effect of silica modulus on the geopolymerization activity of natural pozzolans," Construction and Building Materials, pp. 31–43, 219; https://doi.org/10.1016/j.conbuildmat.2019.05.161.10.1016/j.conbuildmat.2019.05.161 ↗
9. Damtoft JS, Lukasik J, Herfort D, Sorrentino D, Gartner EM (2008) Sustainable development and climate change initiatives. Cem Concr Res 38(2):115–127. https://doi.org/10.1016/j.cemconres.2007.09.00810.1016/j.cemconres.2007.09.008 ↗
10. Guess M (2019) "It’ll cost $45 billion, but Germany proposes to eliminate coal in 19 years. Ars Technica," [Online]. Available: https://arstechnica.com/tech-policy/2019/01/itll-cost-45-billion-but-germany-proposes-to-eliminate-coal-in-19-years/?comments=1&pos. [Accessed 27 03 2021].
11. Mathieson JG, Somerville MA, Deev A and Jahanshahi (2015) "Utilization of biomass as an alternative fuel in ironmaking.," in Iron Ore, vol. 42, Elsevier, p. 581–613.10.1016/B978-1-78242-156-6.00019-8 ↗
12. Manz OE (1997) Worldwide production of coal ash and utilization in concrete and other products. Fuel 76(8):691–69610.1016/S0016-2361(96)00215-3 ↗
13. Provis JL (2018) Alkali-activated materials. Cem Concr Res 114:40–48. https://doi.org/10.1016/j.cemconres.2017.02.00910.1016/j.cemconres.2017.02.009 ↗
14. Lookman T, Balachandran PV, Xue D, Yuan R (2019) Active learning in materials science with emphasis on adaptive sampling using uncertainties for targeted design. NPJ Comput Mater 21(5):2–17
15. Reyes K and Powell WB (2020) "Optimal Learning for Sequential Decisions in Laboratory Experimentation," SIAM Review (preprint), pp. 1–47.
16. Addy SA, Koshla NP (1986) Computer Aided Design of Portland Cement Concrete Mixes. Transp Res Rec 1062:63–69
17. Naderpour H, Rafiean AH and Fakharian P (2018) "Compressive strength prediction of environmentally friendly concrete using artificial neural networks," J Build Eng, pp. 213–219; https://doi.org/10.1016/j.jobe.2018.01.007.10.1016/j.jobe.2018.01.007 ↗
18. Chaabene WB, Flah M and Nehdi ML (2020) "Machine learning prediction of mechanical properties of concrete: Critical review," Construction and Building Materials, vol. 260, p. https://doi.org/10.1016/j.conbuildmat.2020.119889.10.1016/j.conbuildmat.2020.119889 ↗
19. Thomas RJ and Peethamparan S (2017) "Stepwise regression modeling for compressive strength of alkali-activated concrete," Construction and Building Materials, p. 141 (2017) 315–324; https://doi.org/10.1016/j.conbuildmat.2017.03.006.10.1016/j.conbuildmat.2017.03.006 ↗
20. Bagheri A, Nazari A, Sanjayan J (2019) “The use of machine learning in boron-based geopolymers: Function approximation of compressive strength by ANN and GP,” Measurement, p. Measurement 141(2019):241–249. https://doi.org/10.1016/j.measurement.2019.03.001,18April10.1016/j.measurement.2019.03.001,18April ↗
21. Gomaa E, Han T, ElGawady M, Huang J and Kumar A (2020) "Machine learning to predict properties of fresh and hardened alkali-activated concrete," Cement and Concrete Composites, p. 115 (2021) 103863; https://doi.org/10.1016/j.cemconcomp.2020.103863,10.1016/j.cemconcomp.2020.103863 ↗
22. Ahmad A, Farooq F, Niewiadomski P, Ostrowski K, Aslam AAF and Alyousef R (2021) "Prediction of Compressive Strength of Fly Ash Based Concrete Using Individual and Ensemble Algorithm," Materials (Basel). 14(4):794. https://doi.org/10.3390/ma1404079410.3390/ma14040794 ↗
23. Van Dao D, Ly H, Trinh S, Le T and Pham B (2019) "Artificial intelligence approaches for prediction of compressive strength of geopolymer concrete," Materials (Basel), vol. 12, p. https://doi.org/10.3390/ma12060983.10.3390/ma12060983 ↗
24. Kandiri A, Sartipi F, Kioumarsi M (2021) Predicting compressive strength of concrete containing recycled aggregate using modified ANN with different optimization algorithms. Appl Sci 11(485):1–19. https://doi.org/10.3390/app11020485,6.January10.3390/app11020485,6.January ↗
25. Deng F, He Y, Zhou S, Yu Y, Cheng H, Wu X (2018) Compressive strength prediction of recycled concrete based on deep learning. Constr Build Mater 175:562–569. https://doi.org/10.1016/j.conbuildmat.2018.04.16910.1016/j.conbuildmat.2018.04.169 ↗
26. Xu J, Zhao X, Yu Y, Xie T, Yang G and Xue J (2019) "arametric sensitivity analysis and modelling of mechanical properties of normal- and high-strength recycled aggregate concrete using grey theory, multiple nonlinear regression and artificial neural networks," Construction and Building Materials, vol. 211, pp. 479–491, https:// doi.org/https://doi.org/10.1016/j.conbuildmat.2019.03.234.10.1016/j.conbuildmat.2019.03.234 ↗
27. Mohammadi E, Behnood A (2018) Application of soft computing methods for predicting the elastic modulus of recycled aggregate concrete Input layer Output layer. J Clean Prod 176:1163–1176. https://doi.org/10.1016/j.jclepro.2017.11.18610.1016/j.jclepro.2017.11.186 ↗
28. Behnood A and Golafshani E (2020) "Machine learning study of the mechanical properties of concretes containing waste foundry sand," Constr. Build. Mater. vol. 243, p. https://doi.org/10.1016/j.conbuildmat.2020.118152 118152.10.1016/j.conbuildmat.2020.118152 ↗
29. Gholampour A, Mansouri I, Kisi O (2020) Evaluation of mechanical properties of concretes containing coarse recycled concrete aggregates using multivariate adaptive regression splines (MARS), M5 model tree (M5Tree) and least squares support vector regression (LSSVR) models. Neural Comput & Applic 32:295–308. https://doi.org/10.1007/s00521-018-3630-y10.1007/s00521-018-3630-y ↗
30. Yu Y, Li W, Li J and Nguyen TN (2018) "A novel optimised self-learning method for compressive strength prediction of high performance concrete," Construction and Building Materials, pp. 229–247; https://doi.org/10.1016/j.conbuildmat.2018.06.219,10.1016/j.conbuildmat.2018.06.219 ↗
31. Douma OB, Boukhatem B, Ghrici M and Tagnit-Hamou A (2017) "Prediction of properties of self-compacting concrete containing fly ash using artificial neural network," Neural Comput. Appl. vol. 28, pp. 707–718.10.1007/s00521-016-2368-7 ↗
32. Kao CY, Shen CH, Jan JC and Hung SL (2018) "A Computer-Aided Approach to Pozzolanic Concrete Mix Design," Advances in Civil Engineering, pp. 1–15; https://doi.org/10.1155/2018/439801710.1155/2018/4398017 ↗
33. Ziolkowski P, Niedostatkiewicz M (2019) Machine learning techniques in concrete mix design. Mater 12(1256):1–16. https://doi.org/10.3390/ma12081256,17April10.3390/ma12081256,17April ↗
34. Golafshani EM, Behnood A and Arashpour M (2020) "Predicting the compressive strength of normal and High-Performance Concretes using ANN and ANFIS hybridized with Grey Wolf Optimizer," Constr Build Mater. pp. 1–1410.1016/j.conbuildmat.2019.117266 ↗
35. Behnood A, Behnood V, Modiri M, Esat K (2017) Prediction of the compressive strength of normal and high-performance concretes using M5P model tree algorithm. Constr Build Mater 142:199–207. https://doi.org/10.1016/j.conbuildmat.2017.03.06110.1016/j.conbuildmat.2017.03.061 ↗
36. Han Q, Gui JXC, Lacidogna G (2019) A generalized method to predict the compressive strength of high-performance concrete by improved random forest algorithm. Constr Build Mater 226:734–742. https://doi.org/10.1016/j.conbuildmat.2019.07.31510.1016/j.conbuildmat.2019.07.315 ↗
37. Feng DC, Liu ZT, Wang XD, Chen Y, Chang JQ, Wei DF and Jiang ZM (2020) "Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach," Construction and Building Materials, pp. 1–11; https://doi.org/10.1016/j.conbuildmat.2019.11700010.1016/j.conbuildmat.2019.117000 ↗
38. Hameed M, AlOmar M, Baniya W and AlSaadi MA (2021) "Prediction of high-strength concrete: high-order response surface methodology modeling approach," Engineering with Computers, pp. https://doi.org/10.1007/s00366-021-01284-z10.1007/s00366-021-01284-z ↗
39. Lapeyre J, Han T, Wiles B, Ma H, Huang J, Sant G and Kumar A (2021) "Machine learning enables prompt prediction of hydration kinetics of multicomponent cementitious systems," scientific reports, pp. 1–16; https://doi.org/10.1038/s41598-021-83582-610.1038/s41598-021-83582-6 ↗
40. Moro C, Fil HE, Francioso V and Velay-Lizancos M (2020) "Influence of water-to-binder ratio on the optimum percentage of nano- TiO2 addition in terms of compressive strength of mortars: A laboratory and virtual experimental study based on ANN model," Constr Build Mater, pp. 1–11; https://doi.org/10.1016/j.conbuildmat.2020.12096010.1016/j.conbuildmat.2020.120960 ↗
41. Zhang J, Ma G, Huang Y, sun J, Aslani F and Nener B (2019) "Modelling uniaxial compressive strength of lightweight self-compacting concrete using random forest regression," Constr Build Mater, pp. 713–719; https://doi.org/10.1016/j.conbuildmat.2019.03.189 .10.1016/j.conbuildmat.2019.03.189 ↗
42. Mundher Z, Deo R, Hilal A, Abd A, Cornejo L, Salcedo-sanz S, Nehdi M (2018) Software Predicting compressive strength of lightweight foamed concrete using extreme learning machine model. Adv Eng Softw 115:112–125. https://doi.org/10.1016/j.advengsoft.2017.09.00410.1016/j.advengsoft.2017.09.004 ↗
43. Asteris G, Kolovos K (2019) Self-compacting concrete strength prediction using surrogate models. Neural Comput Appl 31:409–424. https://doi.org/10.1007/s00521-017-3007-710.1007/s00521-017-3007-7 ↗
44. Gupta T, Patel K, Siddique S, Sharma R and Chaudhary S (2019) "Prediction of mechanical properties of rubberised concrete exposed to elevated temperature using ANN," Measurement, vol. 147.10.1016/j.measurement.2019.106870 ↗
45. Bachir R, Mohammed A, Habib T (2018) Using artificial neural networks approach to estimate compressive strength for rubberized concrete. Period Polytech Civ Eng 62:858–865. https://doi.org/10.3311/PPci.1192810.3311/PPci.11928 ↗
46. Xie T, Visintin P, Zhao X and Gravina R (2020) "Mix design and mechanical properties of geopolymer and alkali activated concrete: Review of the state-of-the-art and the development of a new unified approach," Constr Build Mater, p. 256 (2020) 119380; https://doi.org/10.1016/j.conbuildmat.2020.11938010.1016/j.conbuildmat.2020.119380 ↗
47. Kim Y, Kim E, Antono E, Meredig B and Ling J (2020) "Machine-learned metrics for predicting the likelihood of success in materials discovery," npj Computational Materials, vol. 6, no. 13110.1038/s41524-020-00401-8 ↗
48. Bondar D, Lynsdale CJ, Milestone NB, Hassani N and Ramezanianpour AA (2011) "Effect of type, form and dosage of activators on strength of alkali-activated natural pozzolans," Cem. Concr. Compos. 33 (2)10.1016/j.cemconcomp.2010.10.021 ↗
49. Shi C, Krivenko PV and Roy D (2006) Alkali-Activated Cements and Concretes, Abingdon: Taylor & Francis Group.10.4324/9780203390672 ↗
50. Firdous R, Stephan D (2020) Impact of the mineralogical composition of natural pozzolan on properties of resultant geopolymers. J Sustainable Cem-Based Mater 176(1):1–16. https://doi.org/10.1080/21650373.2020.180902810.1080/21650373.2020.1809028 ↗
51. Lindley DV (1956) On a measure of the information provided by an experiment. Ann Math Stat 27:986–1005. https://doi.org/10.1214/aoms/117772806910.1214/aoms/1177728069 ↗
52. Quinlan RJ (1986) Induction of Decision Trees, Boston. Kluwer Academic Publishers, USA
53. Opitz D and Maclin D (1999) "Popular Ensemble Methods: An Empirical Study," Journal of Artificial Inteligence Research, p. https://doi.org/10.1613/jair.61410.1613/jair.614 ↗
54. Mathworks (2021) "fitrtree - Fit binary decision tree for regression," [Online]. Available: https://www.mathworks.com/help/stats/fitrtree.html. [Accessed 29 03 2021].
55. Mathworks (2021) "fitrensemble - Fit ensemble of learners for regression," [Online]. Available: https://www.mathworks.com/help/stats/fitrensemble.html?searchHighlight=fitrensemble&s_tid=srchtitle. [Accessed 29 March 2021].
56. Ling J, Hutchinson M, Antono E, Paradiso S, Meredig B (2017) High-dimensional materials and process optimization using data-driven experimental design with well-calibrated uncertainty estimates. Integrating Materials and Manufacturing Innovation 6:207–217. https://doi.org/10.1007/s40192-017-0098-z,05July10.1007/s40192-017-0098-z,05July ↗
57. Krige DG (1951) A statistical approach to some mine valuations and allied problems at the Witwatersrand, University of Witwatersrand: Master thesis.
58. Mathworks (2021) "Kernel (Covariance) Function Options," [Online]. Available: https://www.mathworks.com/help/stats/kernel-covariance-function-options.html.
59. Firdous R and Stephan D (2019) "Influence of heat treatment and mechanical activation on reactivity of natural pozzolan for geopolymer synthesis," 2 – 15 August, Eindhoven, The Netherlands
60. Kani EN and Allahverdi A (2009) "Effect of Chemical Composition on Basic Engineering Properties of Inorganic Polymeric Binder Based on Natural Pozzolan," Ceram. – Silik. 53 (3), p. 195–204.
61. Tänzer R (2018) Alkalisch aktivierte Systeme auf Basis von Hüttensandmehlen (Dissertation). Technische Universität Berlin, Berlin
62. Firdous R and Stephan D (2019) "Factors effecting the properties of alkali activated natural pozzolan based geopolymers," Aachen Germany, pp. 100–106.
63. CEN/TC 51, DIN EN 196–6:2010–05 Methods of testing cement - Part 6: Determination of fineness; German version EN 196–6:2010, 2010.
64. ASTM C311 , Standard Test Methods for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland-Cement Concrete, Vol 05.05 ed., Annual Book of ASTM Standards : ASTM International, 2001.
65. DIN 66126, Determination of specific surface area of disperse solids by the gas permeability technique - Blaine method, 2015.
66. EN 196–6, Methods of testing cement – Part 6: Determination of fineness, 2010.
67. DIN EN ISO 11885, Water quality - Determination of selected elements by inductively coupled plasma optical emission spectrometry (ICP-OES) (ISO 11885:2007); German version EN ISO 11885:2009, 2009.
68. EN 196–2, Methods of testing cement - Part 2: Chemical analysis of cement, 2005.
69. van der Maaten L, Hinton G (2008) Visualizing Data Using t-SNE. J Mach Learn Res 9:2579–2605
70. Mathworks (2021) "k-means clustering," [Online]. Available: https://www.mathworks.com/help/stats/kmeans.html. [Accessed 4 April 2021].10.1109/MSP.2021.3117015 ↗
71. Mathworks (2021) "t-Distributed Stochastic Neighbor Embedding," [Online]. Available: https://www.mathworks.com/help/stats/tsne.html. [Accessed 4. April 2021].
72. Ortega PA, Wang JX, Rowland M, Genewein T, Kurth-Nelson Z, Pascanu R, Heess N, Veness J, Pritzel A, Sprechmann P, Jayakumar SM, McGrath T, Miller K, Azar M, Osband I and Rabinowitz (2019) "Meta-learning of Sequential Strategies," CoRR.