Workability Prediction of Self-Compacting Concrete Based on CatBoost Optimization Algorithm

doi:10.16552/j.cnki.issn1001-1625.2025.0079

Abstract

Abstract: Aiming at the problem that improper control of self-compacting concrete workability can easily lead to defects in concrete structure, this paper presents a method of predicting self-compacting concrete workability based on CatBoost optimization algorithm. First of all, a dataset consisting of 313 sets of data was established through experiment. Subsequently, a predictive model for self-compacting concrete workability was established based on the CatBoost algorithm, and the optimal model hyperparameters were determined using the Optuna automatic hyperparameter optimization framework. Finally, the input parameters of the model were analyzed. The results show that, compared with the dataset without the inclusion of paste test data, the dataset established in this study can effectively improve the prediction accuracy of self-compacting concrete workability. The paste test data exhibits strong correlation and importance in the prediction process. The developed model can accurately predict slump flow and V-shaped funnel time, indicating that the model has good generalization ability and can provide a reference for the design and application of self-compacting concrete.

Key words: self-compacting concrete, paste threshold theory, CatBoost algorithm, Optuna framework, sensitivity analysis, workability prediction

CLC Number:

TU528

LIANG Qimin, WANG Zhe, MEI Yingjie, SUN Ao, LI Pengfei. Workability Prediction of Self-Compacting Concrete Based on CatBoost Optimization Algorithm[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2025, 44(9): 3178-3187.

References

[1] OKAMURA H, OUCHI M. Self-compacting concrete[J]. Journal of Advanced Concrete Technology, 2003, 1(1): 5-15.
[2] HABIBI A, GHOMASHI J. Development of an optimum mix design method for self-compacting concrete based on experimental results[J]. Construction and Building Materials, 2018, 168: 113-123.
[3] SAFHI A M, DABIRI H, SOLIMAN A, et al. Prediction of self-consolidating concrete properties using XGBoost machine learning algorithm: part 1-workability[J]. Construction and Building Materials, 2023, 408: 133560.
[4] OKAMURA H, OZAWA K. Self-compacting high performance concrete[J]. Structural Engineering International, 1996, 6(4): 269-270.
[5] SONEBI M. Medium strength self-compacting concrete containing fly ash: modelling using factorial experimental plans[J]. Cement and Concrete Research, 2004, 34(7): 1199-1208.
[6] 吴琼. 基于净浆流变性的自密实混凝土配合比设计方法研究[D]. 北京: 清华大学, 2013.
WU Q. The development of mix design method for self-compacting concrete based on the rheological characteristics of paste[D]. Beijing: Tsinghua University, 2013 (in Chinese).
[7] 张涛. 基于净浆流变理论与材料堆积特性的含粉煤灰自密实混凝土配合比设计方法[D]. 重庆: 重庆交通大学, 2020.
ZHANG T. An enhanced mix design method of self-compacting concrete with fly ash content based on paste rheological threshold theory and material packing characteristics[D]. Chongqing: Chongqing Jiaotong University, 2020 (in Chinese).
[8] 冉军. 基于材料堆积特性的自密实混凝土流变性能配合比设计模型研究[D]. 重庆: 重庆交通大学, 2022.
RAN J. Research on rheological mix design model of self-compacting concrete based on material packing characteristics[D]. Chongqing: Chongqing Jiaotong University, 2022 (in Chinese).
[9] 胡以婵, 梁铭, 谢灿荣, 等. 基于Stacking模型融合的高性能混凝土强度预测方法[J]. 硅酸盐通报, 2023, 42(11): 3914-3926.
HU Y C, LIANG M, XIE C R, et al. Strength prediction method of high performance concrete based on stacking model fusion[J]. Bulletin of the Chinese Ceramic Society, 2023, 42(11): 3914-3926 (in Chinese).
[10] 李硕, 艾丽菲拉·艾尔肯, 罗文波, 等. 基于AutoML-SHAP的超高性能混凝土抗压强度可解释预测[J]. 硅酸盐通报, 2024, 43(10): 3634-3644.
LI S, AILIFEILA A, LUO W B, et al. Interpretable prediction of compressive strength of ultra-high performance concrete based on AutoML-SHAP[J]. Bulletin of the Chinese Ceramic Society, 2024, 43(10): 3634-3644 (in Chinese).
[11] SONEBI M, CEVIK A. Prediction of fresh and hardened properties of self-consolidating concrete using neurofuzzy approach[J]. Journal of Materials in Civil Engineering, 2009, 21(11): 672-679.
[12] SAHA P, DEBNATH P, THOMAS P. Prediction of fresh and hardened properties of self-compacting concrete using support vector regression approach[J]. Neural Computing and Applications, 2020, 32(12): 7995-8010.
[13] LONG W J, CHENG B Y, LUO S Y, et al. Interpretable auto-tune machine learning prediction of strength and flow properties for self-compacting concrete[J]. Construction and Building Materials, 2023, 393: 132101.
[14] CAKIROGLU C, BEKDAŞ G, KIM S, et al. Explainable ensemble learning models for the rheological properties of self-compacting concrete[J]. Sustainability, 2022, 14(21): 14640.
[15] CHENG B Y, MEI L, LONG W J, et al. Ai-guided proportioning and evaluating of self-compacting concrete based on rheological approach[J]. Construction and Building Materials, 2023, 399: 132522.
[16] 中国土木工程学会. 自密实混凝土设计与施工指南: CCES 02—2004[S]. 北京: 中国建筑工业出版社, 2004.
China Civil Engineering Society. Guide to design and construction of self-compacting concrete: CCES 02—2004[S]. Beijing: China Construction Industry Press, 2004 (in Chinese).
[17] 霍锐, 马芹永, 张鸿朋, 等. 钢-聚丙烯腈纤维掺量对自密实混凝土流动性与力学性能的影响[J]. 硅酸盐通报, 2024, 43(5): 1867-1877+1888.
HUO R, MA Q Y, ZHANG H P, et al. Effect of steel-polyacrylonitrile fiber content on fluidity and mechanical properties of self-compacting concrete[J]. Bulletin of the Chinese Ceramic Society, 2024, 43(5): 1867-1877+1888 (in Chinese).
[18] 聂鼎. 基于净浆流变理论的自密实混凝土配合比优化方法研究[D]. 北京: 清华大学, 2016.
NIE D. Research on optimizing mix proportion of self-compacting concrete based on paste rheological theory[D]. Beijing: Tsinghua University, 2016 (in Chinese).
[19] RICE J A. Mathematical statistics and data analysis[M]. Belmont, CA: Duxbury Press, 1994.
[20] CHEN T, HE T, BENESTY M, et al. Xgboost: extreme gradient boosting[J]. R Package Version 0.4-2, 2015, 1(4): 1-4.
[21] PROKHORENKOVA L, GUSEV G, VOROBEV A, et al. CatBoost: unbiased boosting with categorical features[EB/OL]. (2017-10-15)[2025-01-07]. 2017: 1706.09516. https://arxiv.org/abs/1706.09516v5.
[22] FENG J P, ZHANG H W, GAO K, et al. A machine learning and game theory-based approach for predicting creep behavior of recycled aggregate concrete[J]. Case Studies in Construction Materials, 2022, 17: e01653.
[23] AKIBA T, SANO S, YANASE T, et al. Optuna: a next-generation hyperparameter optimization framework[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage AK USA. ACM, 2019: 2623-2631.
[24] PEDREGOSA F, VAROQUAUX G, GRAMFORT A, et al. Scikit-learn: machine learning in python[J]. the Journal of Machine Learning Research, 2011, 12: 2825-2830.
[25] LUNDBERG S, LEE S. A unified approach to interpreting model predictions[C]//Conference and Workshop on Neural Information Processing Systems. California, USA, 2017.