高温季节大坝混凝土仓面喷雾措施优选研究  

Study on Optimization of Spraying Measures for Dam Concrete Warehouse Surface in High Temperature Season

作　　者：周俊杰 樊仕文 庄维 蔺满刚 魏海东 付雨晨 黄耀英[2] ZHOU Junjie;FAN Shiwen;ZHUANG Wei;LIN Mangang;WEI Haidong;FU Yuchen;HUANG Yaoying(Shanxi Datong Pumped Storage Co.,Ltd.,Shanxi 037400,China;College of Hydraulic and Environmental Engineering,China Three Gorges University,Yichang 443002,China)

机构地区：[1]山西大同抽水蓄能有限公司,山西大同037400 [2]三峡大学水利与环境学院,湖北宜昌443002

出　　处：《三峡大学学报（自然科学版）》2024年第5期13-19,共7页Journal of China Three Gorges University：Natural Sciences

基　　金：国网新源集团有限公司科技项目(SGXY-2022-118);国家自然科学基金项目(52239009)。

摘　　要：高温季节浇筑仓喷雾措施受多种复杂因素影响,本文提出基于正交设计-现场试验-Attention-LSTM神经网络模型-GWO算法优化多目标规划方法进行高温季节混凝土仓面喷雾措施优选.首先采用正交设计构建不同喷雾因素组合工况;然后针对每个喷雾因素组合工况进行喷雾试验,监测获得喷雾区域内温度时空分布规律,将喷雾因素组合工况和相应的实测温度时空分布组成学习样本;进而对Attention-LSTM神经网络模型进行训练和检验,获得预测效果好的Attention-LSTM神经网络模型;最后根据设计确定喷雾降温效果评价指标建立多目标规划模型,通过GWO算法的快速寻优确定最佳喷雾措施.结合实际喷雾试验展示了本文所提仓面喷雾措施优选方法的可行性.The spray measures of pouring warehouse in high temperature season are affected by many complex factors.This paper proposes a multi-objective programming method based on orthogonal design-field test-Attention-LSTM neural network model-GWO algorithm optimization to optimize the spray measures of concrete warehouse surface in high temperature season.Firstly,orthogonal design was used to construct the combination of different spray factors.Then,the spray test was carried out for each combination of spray factors,and the temporal and spatial distribution of temperature in the spray area was monitored.The combination of spray factors and the corresponding measured temperature temporal and spatial distribution were composed of learning samples.Then,the Attention-LSTM neural network model was trained and tested to obtain the Attention-LSTM neural network model parameters with good prediction results.Finally,a multi-objective programming model was established according to the design,which was used to determine the evaluation index of spray cooling effect.The optimal spray measures were determined by the rapid optimization of GWO algorithm.Combined with the actual spray test,the feasibility of the optimization method of warehouse surface spray measures proposed in this paper is demonstrated.

关 键 词：仓面喷雾 高温季节 正交设计 Attention-LSTM神经网络模型 GWO算法 

分 类 号：TV523[水利工程—水利水电工程]