摘要
为实现对某工程中承台大体积混凝土水化热的最佳降温效果,使用实测数据训练了基于BP神经网络的温度预测模型,并结合改进后的遗传算法建立了混凝土水化热管冷参数的数学优化模型。通过模型间的嵌套达到了对冷却水进水温度、冷却水流量和冷水管管径的最优求解。计算结果显示:3项管冷参数的优化均对混凝土水化热温度的降低有一定的效果,其中在一定范围内增大冷却水流量对核心区的降温效果最明显,当冷却水流量由2.0 m^(3)/h增加至2.5 m^(3)/h时,混凝土核心区温度峰值降低4.6℃,累计水化热降低36.4%,降温效果最显著。
In order to achieve the best cooling effect on the mass concrete hydration heat of a certain project,a temperature prediction model based on BP neural network was trained using measured data,and combined with an improved genetic algorithm,the mathematics of the cooling parameters of the concrete hydration heat pipe was established Optimize the model.Through the nesting of the models,the optimal solution of the cooling water inlet temperature,cooling water flow rate and the diameter of the cooling water pipe is achieved.The calculation results show that the optimization of the three tube cooling parameters has a certain effect on the reduction of the concrete hydration heat temperature.Increasing the cooling water flow within a certain range has the most obvious cooling effect on the core area.When the cooling water flow is reduced from 2.0 When m^(3)/h increases to 2.5 m^(3)/h,the peak temperature of the concrete core area decreases by 4.6℃,the cumulative heat of hydration decreases by 36.4%,and the cooling effect is the most significant.
作者
葛庆雷
郭德胜
陈明芳
张祖军
GE Qinglei;GUO Desheng;CHEN Mingfang;ZHANG Zujun(Southwest Branch of Zhejiang Communications Construction Group Co.,Ltd.,Guiyang , Guizhou 550000,China;Hunan Lianzhi Technology Co.,Ltd.,Changsha, Hunan 410000,China;School of Civil Engineering,Changsha University of Science and Technology,Changsha,Hunan 410000,China)
出处
《公路工程》
2022年第2期55-60,共6页
Highway Engineering
基金
国家自然科学基金项目(51478049)。
关键词
大体积混凝土
水化热反应
管冷
人工神经网络
遗传算法
massive concrete
hydration heat response
pipe cooling
artificial neural network
genetic algorithm
