1 School of Civil and Hydraulic Engineerng, Huazhong University of Science and Technology, Wuhan 430074, China 2 School of Civil and Environmental Engineering, Nanyang University of Technology, Singapore 639798 3 Wuhan Huazhong University of Science and Technology Test Technology Co., Ltd., Wuhan 430074, China
Abstract: Aiming at the prediction and optimization of the durability of high-performance concrete in the alpine complex environment, this work developed a hybrid model that integrates random forest (RF) with nondominated sorting genetic algorithm-II with elite strategy (NSGAⅡ) to achieve high-precision concrete performance prediction and multi-objective mix ratio optimization. Taking the chloride ion permeability coefficient and 28 d compressive strength, which are important indicators of concrete durability, as the research objective, a concrete data set was established based on the orthogonal experimental design and engineering actual test samples. The random forest model was used to predict the chloride ion permeability coefficient and strength of concrete. A non-linear mapping function of chloride ion permeability coefficient and strength and mix ratio was obtained, which acted as the fitness function of the optimization target, while the concrete cost was introduced as the fitness function of another optimization target. According to the specifications and requirements of the project, the constraints of raw materials and mix ratios were established, and the NSGAⅡ algorithm was applied for multi-objective optimization of concrete mix ratio. Results show that the accuracy of using random forest to predict concrete performance is fairly high, and that the use of NSGAⅡ algorithm for multi-objective mix ratio optimization is quite effective. The optimized mix ratio scheme was tested and verified, and it was found that the error between the model optimization results and the actual test results was very small. It suggests that the concrete mix ratio meets specifications and engineering project requirements of durability, strength and work performance, reflecting the intelligence and precision of the model, which can provide guidance for the optimization of concrete mix ratio in engineering practice.
1 Long G C, Yang Z X, Bai C N, et al. Journal of the Chinese Ceramic Society, 2019, 47(7), 855 (in Chinese). 龙广成, 杨振雄, 白朝能, 等.硅酸盐学报, 2019, 47(7), 855. 2 He D L, Shi Y, Luo T L, et al. Journal of Adhesion Science and Techno-logy, 2020, 34(14), 1572. 3 Shaheen F, Pradhan B. American Society of Civil Engineers, 2020, 32(9),04020264. 4 Holt E, Ferreira M, Kousa H, et al. Journal of the Chinese Ceramic Society, 2015, 43(10), 1421 (in Chinese). Holt E, Ferreira M, Kousa H, 等. 硅酸盐学报, 2015, 43(10), 1421. 5 Jiang Z W, Xu C P, Zhang S.Journal of Building Materials, 2019, 22(4), 499 (in Chinese). 蒋正武, 徐春普, 张盛. 建筑材料学报, 2019, 22(4), 499. 6 Liang K X, Wang Q C, Zhang K, et al. Concrete, 2019 (5), 38 (in Chinese). 梁柯鑫, 王起才, 张凯, 等.混凝土, 2019 (5), 38. 7 Wang L X, Shan X T.Yangtze River, 2014, 45(13), 72 (in Chinese). 王丽学, 单晓婷.人民长江, 2014, 45(13), 72. 8 Fang W C. Research on the method of field rapid testing of concrete impermeability.Master's Thesis, Harbin Institute of Technology, China, 2018 (in Chinese). 方维澄. 混凝土抗渗性现场快速检测方法的研究. 硕士学位论文, 哈尔滨工业大学, 2018. 9 Liu Z H. Experimental study on early strength of precast concrete.Master's Thesis, Shandong Jianzhu University, China, 2015 (in Chinese). 刘振华. 预制混凝土早期强度试验研究. 硕士学位论文, 山东建筑大学, 2015. 10 Zheng S S, Zhao P, Shang X Y.China Sciencepaper, 2013, 8(5), 413 (in Chinese). 郑山锁, 赵鹏, 商效瑀. 中国科技论文, 2013, 8(5), 413. 11 Han B, Ji K, Hu Y F, et al. Journal of Mining & Safety Engineering, 2020, 38(3),584 (in Chinese). 韩斌, 吉坤, 胡亚飞, 等. 采矿与安全工程学报, 2020, 38(3),584. 12 Behforouz B, Memarzadeh P, Eftekhar M, et al. Computers and Concrete, 2020, 25(2), 119. 13 Zhao W L.Concrete, 2011(2), 46(in Chinese). 赵万里. 混凝土, 2011(2), 46. 14 Yan C L, Zhang A L, Hu H B, et al. Cement Engineering, 2016 (5), 19 (in Chinese). 闫春岭, 张爱玲, 胡海波, 等. 水泥工程, 2016 (5), 19. 15 Fan L Q, Lyu G F.Electronic Design Engineering, 2016, 24(8), 80 (in Chinese). 范立强, 吕国芳. 电子设计工程, 2016, 24(8), 80. 16 Li S, Du Y J.Concrete, 2016 (3), 44 (in Chinese). 李松, 杜应吉. 混凝土, 2016 (3), 44. 17 Huang M, Zhao Y R, Yuan J J, et al. Journal of Henan University of Engineering(Natural Science Edition), 2019, 31(4), 22 (in Chinese). 皇民, 赵玉如, 苑俊杰, 等. 河南工程学院学报(自然科学版), 2019, 31(4), 22. 18 Zhang J F, Huang Y M, Wang Y H, et al. Construction and Building Materials, 2020, 253, 119208. 19 Zhu Y, Xu W Y, Luo G L, et al. Artificial Intelligence in Medicine, 2020, 103, 101811. 20 Golrang A, Golrang A M, Yayilgan S Y, et al. Electronics, 2020, 9(4),577. 21 Pomalis C S M, Leborgne R C, Herrera-Orozco A R, et al. Electric Power Systems Research, 2019, 176, 105. 22 Breiman L. Machine Learning, 2001, 45(1), 5. 23 Yao H, Li X D, Pang H J, et al. Atmospheric Research, 2020, 244, 105093. 24 Zhu Y, Xu W Y, Luo G L, et al. Artificial Intelligence in Medicine, 2020, 103, 101811. 25 Lu S L, Li Q P, Bai L, et al. Energy Conversion and Management, 2019, 197, 111864. 26 Guo K R. Risk prediction and evaluation of P2P network loan platform based on random forest model.Master's Thesis, Tianjin University of Finance and Economics, China, 2019 (in Chinese). 郭凯瑞. 基于随机森林模型的P2P网贷平台风险预测与评价. 硕士学位论文, 天津财经大学, 2019. 27 Deb K, Pratap A, Agarwal S, et al. IEEE Transactions on Evolutionary Computation, 2002, 6(2), 182197. 28 Lee U, Park S, Lee I. Energy, 2020, 196, 117090. 29 Tabatabaei M, Salehpour J A, Hosseini S A. International Soil and Water Conservation Research, 2019, 7(2), 119. 30 Gunjan, Sharma A K, Verma K. Wireless Networks, 2020, 26(5), 3637. 31 Lai C G, Li L, Zhou Y T, et al. Journal of Chongqing University of Technology(Natural Science), 2022, 36(4), 19 (in Chinese). 赖晨光,李亮,周毓婷,等.重庆理工大学学报(自然科学), 2022, 36(4), 19.