| 1,768 | 10 | 364 |
| 下载次数 | 被引频次 | 阅读次数 |
机器学习方法(ML)在解决数据高度非线性和大时滞问题上得到了关注，为目标属性的预测及新材料的设计开辟了数据驱动的新模式。水泥基材料在组分、制备与其结构、性能的关系上存在时滞性和周期性长的问题。而机器学习方法在解决水泥基材料上述问题时具有积极作用和非常大的潜力。为此，本文介绍了机器学习方法的特征及其应用过程，综述了机器学习方法在预测水泥基材料微观结构、组分、力学性能和耐久性等方面的研究进展，分析比较了几种用于水泥基材料研究模型的特点和优势，并对所面临的问题和研究方向进行了展望分析，以期促进机器学习方法的应用。
Abstract：The integration of Machine Learning (ML) in material science,particularly in cement-based materials,is revolutionizing the field.This approach addresses challenges like high non-linearity and significant time-lags in data,which are prevalent in the complex interplay of composition,process,structure,and performance in these materials.ML’s effectiveness in this context is not just in predictive accuracy,but also in its potential to significantly advance research and development.ML operates at the intersection of computer science and statistics,functioning as a‘black-box’that links data without requiring an understanding of the underlying mechanisms.This enables a comprehensive cycle from data collection to performance prediction and experimental validation.By transitioning from a traditional‘experience plus trial-and-error’method to a data-driven approach,ML facilitates a deeper understanding of the cause-effect relationships aiming at material properties.This is particularly transformative in cement-based materials research,where ML’s ability to predict various properties opens up new possibilities for enhancing the efficiency of their development and application.This comprehensive article delves into the intricate characteristics and varied application processes of ML in the realm of material science,with a specific focus on cement-based materials.It provides a thorough review of the recent progress,showcasing how ML techniques have become instrumental in predicting key aspects such as microstructure,components,mechanical properties,and durability of these materials.The article not only illustrates the predictive power of ML,but also sheds light on its role in enhancing the understanding of complex material behaviors and properties.Additionally,the article examines the unique characteristics and structural intricacies of various ML models used in this context.It meticulously discusses and compares different model algorithms,dissecting their methodologies and computational approaches.By doing so,it offers a clear picture of the strengths and limitations of each algorithm,providing valuable insights into their practical applicability in predicting and understanding the behavior of cement-based materials.Moreover,the article reviews the evaluation metrics used to assess the performance of these ML models,emphasizing the importance of accuracy,reliability,and efficiency in material science research.This evaluation not only highlights the current state of ML applications in this field,but also suggests areas for future improvements and developments.Despite the notable advancements in using ML for predicting the structure and performance of cement-based materials,significant challenges remain.Firstly,there is an issue of data quality and quantity imbalance due to the multi-temporal relationship between the cement process and the development of its properties,which necessitates the creation of a comprehensive database including diverse factors such as material components,structures,performance metrics,environmental influences,and chemical reaction parameters.Secondly,the highly complex and nonlinear nature of most predictive models for cement-based materials leads to insufficient model interpretability,that is,understanding of these models'decision-making processes is quite complicated.Enhancing interpretability is essential for a deeper comprehension of material performance under various conditions,which can be achieved through advanced post-processing tools and integrating ML with precise physical models.Lastly,the limited generalizability of ML models,due to the inherent complexity of cement-based materials,poses a challenge.Training data may not cover all categories,leading to diminished performance upon treating new data sets.To address this problem,it requires exploring meta-learning methods that can quickly adapt to new material combinations and combining these methods with other advanced ML techniques to improve the predictive power and adaptability of models.The evolution of Artificial Intelligence and increased computational power,as exemplified by advanced generative AI models like ChatGPT,offer boundless potential in enhancing the efficiency of material research and development.Their multi-dimensional data processing capability allows for comprehensive consideration of various factors affecting target performance,predicting cement properties under different conditions.As these models evolve in self-learning and optimization,integrating molecular dynamics simulations of cement-based materials,three-dimensional simulations of hardened paste structures,and reaction kinetics and thermodynamics data becomes feasible.This integration could lead to precise predictions of composition,structure,and performance,and even the reverse engineering of cement-based materials,significantly accelerating the development of new cement-based materials.Summary and prospects The integration with ML is a major advancement in the field of cement-based materials.ML is able to process and to learn from huge datasets to predict a range of material properties.This approach not only solves the complex problem of nonlinear regression of materials,but also marks a new era in materials research.However,to realize the full potential of machine learning in this area,challenges such as imbalances in data quality and quantity,insufficient model interpretability and limited model commonality need to be addressed.The research prospect of cement-based materials by integrating with artificial intelligence is promising.And,with the continuous improvement of AI capability and computational power,we can foresee that more complex models combined with more advanced algorithms will be able to predict and design cement-based materials more accurately.This progress will likely lead to the development of new materials with enhanced properties,contributing significantly to the field of construction and material science.This research area remains a hotspot,promisingly exciting developments and breakthroughs in the near future.
[1] CLAUDE S, GEOFFREY I W, Encyclopedia of machine learning[M]:Springer Science&Business Media, New York, US, 2011.
[2] JORDAN M I, MITCHELL T M. Machine learning:Trends,perspectives, and prospects[J]. Science, 2015, 349(6245):255–260.
[3] TOM M M, Machine learning[M]:McGraw-Hill Science/Engineering/Math,New York, US, 1997.
[4] KADRI E H, KENAI S, EZZIANE K, et al. Influence of metakaolin and silica fume on the heat of hydration and compressive strength development of mortar[J]. Appl Clay Sci, 2011, 53(4):704–708.
[5] LAWLER J S, ZAMPINI D, SHAH S P. Microfiber and macrofiber hybrid fiber-reinforced concrete[J]. J Mater Civ Eng, 2005, 17(5):595–604.
[6]王云帆,田原,周玉美,等.主动学习辅助材料研发进展[J].硅酸盐学报, 2023, 51(2):544–551.WANG Yunfan, TIAN Yuan, ZHOU Yumei, et al. J Chin Ceram Soc,2023, 51(2):544–551.
[7] On machine learning and the next wave of innovation[EB/OL].(2020)[2023–12–05]. http://www.linkedin.com/pulse/machine-learning-next-waveinnovation-eric-miley-mba.
[8] LIU Y, GUO B R, ZOU X X, et al. Machine learning assisted materials design and discovery for rechargeable batteries[J]. Energy Storage Mater, 2020, 31:434–450.
[9] LIU Y, ZHAO T L, JU W W, et al. Materials discovery and design using machine learning[J]. J Materiomics, 2017, 3(3):159–177.
[10] BUTLER K T, DAVIES D W, CARTWRIGHT H, et al. Machine learning for molecular and materials science[J]. Nature, 2018,559(7715):547–555.
[11] LIU Y, ZOU X X, MA S C, et al. Feature selection method reducing correlations among features by embedding domain knowledge[J]. Acta Mater, 2022, 238:118195.
[12] LIU Y, WANG S Y, YANG Z W, et al. Auto-MatRegressor:Liberating machine learning alchemists[J]. Sci Bull, 2023, 68(12):1259–1270.
[13] AZIMI S M, BRITZ D, ENGSTLER M, et al. Advanced steel microstructural classification by deep learning methods[J]. Sci Rep,2018, 8:2128.
[14] SAHA S K, PRADHAN S, BARAI S V. Use of machine learning based technique to X-ray microtomographic images of concrete for phase segmentation at meso-scale[J]. Constr Build Mater, 2020, 249:118744.
[15] LIU F Y, XU K, DING W Q, et al. Microstructural characteristics and their impact on mechanical properties of steel-PVA fiber reinforced concrete[J]. Cem Concr Compos, 2021, 123:104196.
[16] LORENZONI R, CUROSU I, PACIORNIK S, et al. Semantic segmentation of the micro-structure of strain-hardening cement-based composites(SHCC)by applying deep learning on micro-computed tomography scans[J]. Cem Concr Compos, 2020, 108:103551.
[17] WEI L Y, LEVOY M. Fast texture synthesis using tree-structured vector quantization[C]//Proceedings of the 27th annual conference on Computer graphics and interactive techniques. ACM, 2000:479–488.
[18]范嘉薇.基于神经网络的水泥微结构快速建模及性能分析[D].济南:济南大学, 2021.FAN Jiawei. Rapid modeling and performance analysis of cement microstructure based on neural network[D]. Jinan:University of Jinan,2021.
[19]张玉玲.基于数据驱动的水泥熟料游离氧化钙软测量方法研究[D].秦皇岛:燕山大学, 2021.ZHANG Yuling. Research on soft-sensing method of free calcium oxide in cement clinker based on data drive[D]. Qinhuangdao:Yanshan University, 2021.
[20] ALBERT C, ISGOR O B, ANGST U. Exploring machine learning to predict the pore solution composition of hardened cementitious systems[J]. Cem Concr Res, 2022, 162:107001.
[21] BOLSTAD W M. Introduction to Bayesian Statistics[M]. John Wiley&Sons, 2016:428–595.
[22] KE X Y, DUAN Y. A Bayesian machine learning approach for inverse prediction of high-performance concrete ingredients with targeted performance[J]. Constr Build Mater, 2021, 270:121424.
[23]裘国华,申屠南瑛,施正伦.基于GM-RBF神经网络组合模型的水泥强度预测方法[J].科技导报, 2014, 32(3):56–61.QIU Guohua, SHENTU Nanying, SHI Zhenglun. Sci Tech Rev, 2014,32(03):56–61.
[24] ASTERIS P G, KOOPIALIPOOR M, ARMAGHANI D J, et al.Prediction of cement-based mortars compressive strength using machine learning techniques[J]. Neural Comput Appl, 2021, 33(19):13089–13121.
[25] HABEEB G A, FAYYADH M M, Rice husk ash concrete:the effect of RHA average particle size on mechanical properties and drying shrinkage[J]. Aust J Basic Appl Sci, 2009, 3(3):1616–1622.
[26] JAIN A, PATEL H, NAGALAPATTI L, et al. Overview and importance of data quality for machine learning tasks[C]//Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery&data mining. 2020:3561–3562.
[27] GUDIVADA V, APON A, DING J. Data quality considerations for big data and machine learning:Going beyond data cleaning and transformations[J]. Int J Adv Softw, 2017, 10(1):1–20.
[28]刘悦,马舒畅,杨正伟,等.面向材料领域机器学习的数据质量治理[J].硅酸盐学报, 2023, 51(2):427–437.LIU Yue, MA Shuchang, YANG Zhengwei, et al. J Chin Ceram Soc,2023, 51(2):427–437.
[29] EBERHART R, KENNEDY J. A new optimizer using particle swarm theory[C]//MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science. Nagoya, Japan. IEEE, 1995:39–43.
[30] KENNEDY J, EBERHART R. Particle swarm optimization[C]//Proceedings of ICNN'95-International Conference on Neural Networks. Perth, WA,Australia. IEEE, 1995:1942–1948.
[31] STORN R, PRICE K. Differential evolution–A simple and efficient heuristic for global optimization over continuous spaces[J]. J Glob Optim, 1997, 11(4):341–359.
[32] TEODOROVIC D, DELL’ORCO M, Bee colony optimization–a cooperative learning approach to complex transportation problems[J].Adv Or Ai, 2005, 51:60.
[33] DORIGO M, MANIEZZO V, COLORNI A. Ant system:Optimization by a colony of cooperating agents[J]. IEEE Trans Syst Man Cybern B Cybern, 1996, 26(1):29–41.
[34] HAMIDIAN P, ALIDOUST P, GOLAFSHANI E M, et al. Introduction of a novel evolutionary neural network for evaluating the compressive strength of concretes:A case of rice husk ash concrete[J]. J Build Eng,2022, 61:105293.
[35] 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[J]. Constr Build Mater, 2020, 232:117266.
[36] ASTERIS P G, SKENTOU A D, BARDHAN A, et al. Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models[J]. Cem Concr Res, 2021, 145:106449.
[37] LUNDBERG S M, NAIR B, VAVILALA M S, et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery[J]. Nat Biomed Eng, 2018, 2(10):749–760.
[38] LAURITSEN S M, KRISTENSEN M, OLSEN M V, et al. Explainable artificial intelligence model to predict acute critical illness from electronic health records[J]. Nat Commun, 2020, 11(1):3852.
[39] LYNGDOH G A, ZAKI M, ANOOP KRISHNAN N M, et al.Prediction of concrete strengths enabled by missing data imputation and interpretable machine learning[J]. Cem Concr Compos, 2022, 128:104414.
[40] LIANG M F, GAN Y D, CHANG Z, et al. Microstructure-informed deep convolutional neural network for predicting short-term creep modulus of cement paste[J]. Cem Concr Res, 2022, 152:106681.
[41] DUAN Z H, KOU S C, POON C S. Using artificial neural networks for predicting the elastic modulus of recycled aggregate concrete[J]. Constr Build Mater, 2013, 44:524–532.
[42] LIANG M F, CHANG Z, WAN Z, et al. Interpretable Ensemble-Machine-Learning models for predicting creep behavior of concrete[J]. Cem Concr Compos, 2022, 125:104295.
[43] KHAN I, XU T F, CASTEL A, et al. Risk of early age cracking in geopolymer concrete due to restrained shrinkage[J]. Constr Build Mater,2019, 229:116840.
[44] CASTEL A, FOSTER S J, NG T, et al. Creep and drying shrinkage of a blended slag and low calcium fly ash geopolymer concrete[J]. Mater Struct, 2016, 49(5):1619–1628.
[45] PACHECO-TORGAL F, DING Y N, JALALI S. Properties and durability of concrete containing polymeric wastes(tyre rubber and polyethylene terephthalate bottles):An overview[J]. Constr Build Mater,2012, 30:714–724.
[46] SHEN L H, LI Q H, GE W, et al. The mechanical property and frost resistance of roller compacted concrete by mixing silica fume and limestone powder:Experimental study[J]. Constr Build Mater, 2020,239:117882.
[47] FENG J P, ZHANG H W, GAO K, et al. Efficient creep prediction of recycled aggregate concrete via machine learning algorithms[J]. Constr Build Mater, 2022, 360:129497.
[48] TO?I?N, DE LA FUENTE A, MARINKOVI?S. Creep of recycled aggregate concrete:Experimental database and creep prediction model according to the fib model code 2010[J]. Constr Build Mater, 2019, 195:590–599.
[49] PAUL S C, PANDA B, HUANG Y H, et al. An empirical model design for evaluation and estimation of carbonation depth in concrete[J].Measurement, 2018, 124:205–210.
[50] KELLOUCHE Y, BOUKHATEM B, GHRICI M, et al. Exploring the major factors affecting fly-ash concrete carbonation using artificial neural network[J]. Neural Comput Appl, 2019, 31(2):969–988.
[51] CUNNINGHAM P, CARNEY J, JACOB S. Stability problems with artificial neural networks and the ensemble solution[J]. Artif Intell Med,2000, 20(3):217–225.
[52] LIU K H, ALAM M S, ZHU J, et al. Prediction of carbonation depth for recycled aggregate concrete using ANN hybridized with swarm intelligence algorithms[J]. Constr Build Mater, 2021, 301:124382.
[53] MOGHADDAS S A, NEKOEI M, MOHAMMADI GOLAFSHANI E,et al. Modeling carbonation depth of recycled aggregate concrete using novel automatic regression technique[J]. J Clean Prod, 2022, 371:133522.
[54] STROBL C, BOULESTEIX A L, KNEIB T, et al. Conditional variable importance for random forests[J]. BMC Bioinformatics, 2008, 9:307.
[55] NUNEZ I, NEHDI M L. Machine learning prediction of carbonation depth in recycled aggregate concrete incorporating SCMs[J]. Constr Build Mater, 2021, 287:123027.
[56] BO?A A R,?ZTüRK M, TOP?U?B. Using ANN and ANFIS to predict the mechanical and chloride permeability properties of concrete containing GGBFS and CNI[J]. Compos Part B Eng, 2013, 45(1):688–696.
[57] LIU Q, IQBAL M F, YANG J, et al. Prediction of chloride diffusivity in concrete using artificial neural network:Modelling and performance evaluation[J]. Constr Build Mater, 2021, 268:121082.
[58] INTHATA S, KOWTANAPANICH W, CHEERAROT R. Prediction of chloride permeability of concretes containing ground pozzolans by artificial neural networks[J]. Mater Struct, 2013, 46(10):1707–1721.
[59] SONG H W, KWON S J. Evaluation of chloride penetration in high performance concrete using neural network algorithm and micro pore structure[J]. Cem Concr Res, 2009, 39(9):814–824.
[60] VAN Q T. Machine learning approach for investigating chloride diffusion coefficient of concrete containing supplementary cementitious materials[J]. Constr Build Mater, 2022, 328:127103.
[61] ABDULALIM A A, IQBAL M, ZAHID M, et al. Prediction of rapid chloride penetration resistance of metakaolin based high strength concrete using light GBM and XGBoost models by incorporating SHAP analysis[J]. Constr Build Mater, 2022, 345:128296.
[62] LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA.ACM, 2017:4768–4777.
[63] LIU K H, ZOU C Y, ZHANG X C, et al. Innovative prediction models for the frost durability of recycled aggregate concrete using soft computing methods[J]. J Build Eng, 2021, 34:101822.
[64]吴贤国,刘茜,王雷,等.基于LSSVM与GA的混凝土抗冻性多目标配合比优化研究[J].沈阳建筑大学学报(自然科学版), 2021, 37(3):393–401.WU Xianguo, LIU Xi, WANG Lei, et al. J Shenyang Jianzhu Univ Nat Sci, 2021, 37(3):393–401.
[65] Zhang J, Huang Y, Wang Y, et al. Multi-objective optimization of concrete mixture proportions using machine learning and metaheuristic algorithms[J]. Constr Build Mater, 2020, 253:119208.
[66] ZHANG J F, HUANG Y M, MA G W, et al. Mixture optimization for environmental, economical and mechanical objectives in silica fume concrete:A novel frame-work based on machine learning and a new meta-heuristic algorithm[J]. Resour Conserv Recycl, 2021, 167:105395.
[67] CEMALGIL S, GüL E, ONAT O, et al. A novel prediction model for durability properties of concrete modified with steel fiber and Silica Fume by using Hybridized GRELM[J]. Constr Build Mater, 2022, 341:127856.
[68] ASTERIS P G, KOLOVOS K G, DOUVIKA M G, et al. Prediction of self-compacting concrete strength using artificial neural networks[J].Eur J Environ Civ Eng, 2016, 20(s1):s102–s122.
[69] RUTKOWSKA G, WICHOWSKI P, FRANUS M, et al. Modification of ordinary concrete using fly ash from combustion of municipal sewage sludge[J]. Materials, 2020, 13(2):487.
[70] SHEWA G A, UGWUOWO F I. A new hybrid estimator for linear regression model analysis:Computations and simulations[J]. Sci Afr,2023, 19:e01441.
[71] FAN C C, ZHENG Y X, WANG S Q, et al. Prediction of bond strength of reinforced concrete structures based on feature selection and GWO-SVR model[J]. Constr Build Mater, 2023, 400:132602.
[72] MOTLAGH S A T, NAGHIZADEHROKNI M. An extended multi-model regression approach for compressive strength prediction and optimization of a concrete mixture[J]. Constr Build Mater, 2022,327:126828.
[73] ERDAL H I. Two-level and hybrid ensembles of decision trees for high performance concrete compressive strength prediction[J]. Eng Appl Artif Intell, 2013, 26(7):1689–1697.
[74] BELLO S A, OYEDELE L, OLAITAN O K, et al. A deep learning approach to concrete water-cement ratio prediction[J]. Results Mater,2022, 15:100300.
[75] MIN C D, XIONG S, SHI Y, et al. Early-age compressive strength prediction of cemented phosphogypsum backfill using lab experiments and ensemble learning models[J]. Case Stud Constr Mater, 2023, 18:e02107.
[76] WANG S Q, XIA P, CHEN K Y, et al. Prediction and optimization model of sustainable concrete properties using machine learning, deep learning and swarm intelligence:A review[J]. J Build Eng, 2023, 80:108065.
[77] PARUTHI S, HUSAIN A, ALAM P, et al. A review on material mix proportion and strength influence parameters of geopolymer concrete:Application of ANN model for GPC strength prediction[J]. Constr Build Mater, 2022, 356:129253.
[78] MOHTASHAM MOEIN M, SARADAR A, RAHMATI K, et al.Predictive models for concrete properties using machine learning and deep learning approaches:A review[J]. J Build Eng, 2023, 63:105444.
[79] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015,521:436–444.
[80] HEATON J. Ian Goodfellow, Yoshua Bengio, and Aaron Courville:Deep learning[J]. Genet Program Evolvable Mach, 2018, 19(1/2):305–307.
[81] EMMERT-STREIB F, YANG Z, FENG H, et al. An introductory review of deep learning for prediction models with big data[J]. Front Artif Intell, 2020, 3:4.
[82] ALZUBAIDI L, ZHANG J L, HUMAIDI A J, et al. Review of deep learning:Concepts, CNN architectures, challenges, applications, future directions[J]. J Big Data, 2021, 8(1):53.
[83] ZHOU J, CUI G Q, HU S D, et al. Graph neural networks:A review of methods and applications[J]. AI Open, 2020, 1:57–81.
[84] ISLAM S, ELMEKKI H, ELSEBAI A, et al. A comprehensive survey on applications of transformers for deep learning tasks[J]. Expert Syst Appl, 2024, 241:122666.
[85] YIN W P, KANN K, YU M, et al. Comparative study of CNN and RNN for natural language processing[EB/OL]. 2017:arXiv:1702. 01923.http://arxiv. org/abs/1702. 01923. pdf
[86] LIU Y, YANG Z W, YU Z Y, et al. Generative artificial intelligence and its applications in materials science:Current situation and future perspectives[J]. J Materiomics, 2023, 9(4):798–816.
[87] YU J, WENG Y W, YU J T, et al. Generative AI for performance-based design of engineered cementitious composite[J]. Compos Part B Eng,2023, 266:110993.
基本信息:
DOI：10.14062/j.issn.0454-5648.20230925
中图分类号:TU528
引用信息:
[1]张文生,曹傅荔,郅晓,等.机器学习方法用于水泥基材料的研究进展[J].硅酸盐学报,2024,52(11):3617-3630.DOI:10.14062/j.issn.0454-5648.20230925.
基金信息:
中国建材集团原创技术策源地“揭榜挂帅”任务-“水泥基材料数字化研发”项目(2021YCJS01)
2024-04-30
2024-04-30
2024-04-30