摘要
分析泵站大体积混凝土施工过程中温度演变规律,并构建其混凝土温度预测模型,对工程实际生产意义重大。文章结合邢台市尖冢灌区重建工程冢灌泵站实测数据,分析了其温度变化特征,并结合了卷积神经网络(CNN)、门控循环单元(GRU)和注意力机制(SE)构建了组合深度学习模型(CGS模型),利用非洲秃鹰算法(AVOA)、改进的灰狼算法(I-GWO)等仿生优化算法优化CGS模型超参数,如学习率、隐藏层单元的数量等,以提升模型的预测性能,最后利用指标体系展开性能评价。结论如下:组合模型的预测精度高于单一模型,其中经过注意力机制(SE)优化的CGS模型预测精度高于CG模型,经过仿生算法优化的CGS模型,在泵站底板各层混凝土温度预测中均表现出了较高的精度,可推广应用于实际大体积混凝土温控及养护领域。
It is of great significance to the actual production of the project to analyze the temperature evolution law during the construction of mass concrete in pumping station and construct its concrete temperature prediction model.In this paper,the temperature change characteristics of the mound irrigation pumping station of Xingtai Jianzuka Irrigation District Reconstruction Project were analyzed by combining the measured data,and constructed a combined deep learning model(CGS model)by combining convolutional neural network(CNN),gated recurrent unit(GRU)and attention mechanism(SE),and optimized the CGS model by using bionic optimization algorithms,such as African Vulture Algorithm(AVOA),Improved Gray Wolf Algorithm(I-GWO)and so on.The hyperparameters of the CGS model,such as the learning rate and the number of hidden layer units,were optimized to improve the prediction performance of the model,and finally the performance evaluation was carried out using Ithe index system.The conclusions are as follows:the prediction accuracy of the combined model is higher than that of the single model,in which the prediction accuracy of the CGS model optimized by the attention mechanism(SE)is higher than that of the CG model,and the CGS model optimized by bionic optimization algorithm shows high accuracy in predicting the temperature of the concrete in all layers of the pumping station bottom slab,which can be promoted and applied in the field of actual temperature control and maintenance of large volume concrete.
作者
陈警
王世亮
张帅
CHEN Jing;WANG Shiliang;ZHANG Shuai(Hebei Provincial Water Conservancy Engineering Bureau Group Co.,Ltd,Shijiazhuang 050021,China)
出处
《黑龙江水利科技》
2025年第4期52-57,共6页
Heilongjiang Hydraulic Science and Technology
基金
河北省省级水利科技计划项目(2024-39)
河北省水利工程局集团有限公司科技计划项目(2024-08-CX-002)。
关键词
混凝土
泵站
温度预测模型
仿生优化算法
concrete
pumping station
temperature prediction model
bionic optimization algorithm
