水工高性能混凝土配合比多目标智能优化设计与分析方法  

Multi-objective intelligent optimization design and analysis method for mix proportion of hydraulic high performance concrete

作　　者：任秋兵 李文伟 李明超[1] 杨琳[1] 张梦溪[1] 沈扬 REN Qiubing;LI Wenwei;LI Mingchao;YANG Lin;ZHANG Mengxi;SHEN Yang(State Key Laboratory of Hydraulic Engineering Simulation and Safety,Tianjin University,Tianjin 300354,China;China Three Gorges Corporation,Beijing 100038,China)

机构地区：[1]天津大学水利工程仿真与安全国家重点实验室,天津300354 [2]中国长江三峡集团有限公司,北京100038

出　　处：《水利学报》2022年第1期98-108,共11页Journal of Hydraulic Engineering

基　　金：国家自然科学基金重点项目(U2040222);国家自然科学基金面上项目(51879185);国家重点研发计划项目(2018YFC0406905)。

摘　　要：高性能混凝土(HPC)在水利工程中的多元化应用对其配合比设计提出了更高要求。为实现高效、准确的配合比设计,本文结合人工智能算法和元启发式搜索技术,开发了一种综合考虑多目标需求(力学性能、经济性和环保性)的HPC配合比优化设计新方法。其主要包括:(1)建立了以配合比参数为设计变量,以抗压强度、单方生产成本和碳排放量为优化目标的HPC配合比多目标联合优化数学模型;(2)提出了基于随机森林算法和自适应进化多目标粒子群优化算法(AEPSO)的模型求解方法,其中随机森林算法用于表征原料配合比与抗压强度之间的复杂非线性关系,而AEPSO用于迭代搜索最优Pareto前沿,以期获得不同偏向的最佳配比方案。利用本文方法对某工程用HPC的配合比进行多目标优化设计,并通过与传统方法进行对比验证了该方法的有效性与优越性。The diversified application of high performance concrete(HPC) in hydraulic engineering puts forward higher requirements for its mix proportion design. In order to achieve the efficient and accurate concrete mix, this paper combines the artificial intelligence algorithm and meta-heuristic search technique to develop a novel method for HPC mix optimization design that comprehensively considers multi-objective requirements(mechanical properties,economy and environmental protection). The developed method mainly includes that:(1) Establishing a HPC mix proportion multi-objective optimization mathematical model that takes mix parameters as design variables and takes compressive strength,unilateral production cost and carbon dioxide emissions as optimization goals;(2) Proposing a model solving method based on the random forest and adaptive evolutionary particle swarm optimization(AEPSO) algorithms, where random forest is used to characterize the complex nonlinear relationship between the mix proportion of raw materials and the compressive strength,while AEPSO for the optimal Pareto front,in order to obtain the optimal mix proportion of different biases. The proposed method has been used to optimize the mix proportion of HPC for a certain project,and the effectiveness and superiority of the proposed method are verified by comparing with the traditional method.

关 键 词：高性能混凝土 配合比设计 抗压强度预测 多目标优化 随机森林算法 

分 类 号：TV431[水利工程—水工结构工程]