摘要
为有效预测混凝土浇筑仓内部最高温度,引入随机森林算法,建立了基于随机森林算法的混凝土浇筑仓内部最高温度预测模型,结合西部某典型混凝土坝工程,收集高、低温季节浇筑仓实测温度,以冷却水管进出口水温、通水流量、浇筑仓层厚和环境气温为输入,混凝土浇筑仓内部最高温度为输出,采用所建模型进行训练和优化。实例应用结果表明,与传统BP神经网络预测模型相比,基于随机森林算法的混凝土浇筑仓内部最高温度预测模型对混凝土内部最高温度预测具有更好的效果。
In order to effectively predict the internal maximum temperature of concrete pouring block,random forest algorithm was used to establish the prediction model of concrete pouring block.Combined with a typical concrete dam project in the west region,the measured temperature of the pouring block in high and low temperature seasons was collected firstly.The water temperature at the inlet and outlet of the cooling water pipe,the water flow rate,the thickness of the pouring block and the ambient temperature were taken as the input,and the maximum temperature of the concrete pouring block was taken as the output.Then,the random forest model was used for training and optimization.Comparison of the traditional BP neural network,the example analysis shows that the random forest prediction model has better effect on prediction of internal temperature of concrete.
作者
周俊杰
樊仕文
方晨
黄耀英
刘富昌
ZHOU Jun-jie;FAN Shi-wen;FANG Chen;HUANG Yao-ying;LIU Fu-chang(Shanxi Datong Pumped Storage Co.,Ltd.,Datong 037400,China;College of Hydraulic and Environment Engineering,China Three Gorges University,Yichang 443002,China)
出处
《水电能源科学》
北大核心
2024年第9期84-87,共4页
Water Resources and Power
基金
国网新源集团有限公司科技项目(SGXY-2022-118)。
关键词
混凝土浇筑仓
最高温度
随机森林
温度预测
温控防裂
concrete pouring block
the highest temperature
random forest
temperature prediction
temperature control and crack prevention
