• 中国科学引文数据库(CSCD)来源期刊
  • 中文核心期刊中文科技核心期刊
  • Scopus RCCSE中国核心学术期刊
  • 美国EBSCO数据库 俄罗斯《文摘杂志》
  • 《日本科学技术振兴机构数据库(中国)》
二维码

隧道建设(中英文) ›› 2023, Vol. 43 ›› Issue (4): 611-617.DOI: 10.3973/j.issn.2096-4498.2023.04.007

• 研究与探索 • 上一篇    下一篇

基于深度学习的隧道混凝土温度滚动预测方法研究

范中晶1, 郑晨路2   

  1. 1. 中铁隧道局集团有限公司, 广东 广州 511458 2. 上海大学土木工程系, 上海 200444
  • 出版日期:2023-04-20 发布日期:2023-05-23
  • 作者简介:范中晶(1988—),男,河南封丘人,2018年毕业于长沙理工大学,土木工程专业,本科,工程师,现从事隧道工程科研与施工管理工作。Email: 389173477@qq.com。

Rolling Prediction Method of Tunnel Concrete Temperature Based on Deep Learning

FAN Zhongjing1, ZHENG Chenlu2   

  1. (1. China Railway Tunnel Group Co., Ltd., Guangzhou 511458, Guangdong, China; 2. Department of Civil Engineering, Shanghai University, Shanghai 200444, China)
  • Online:2023-04-20 Published:2023-05-23

摘要: 为实时掌握隧道大体积混凝土内部温度变化,及时制定合理的控温措施,防止隧道大体积混凝土在浇筑过程中由于大范围的水化热反应而产生温度裂缝,基于长短期记忆神经网络LSTMlong short-term memory)和门控循环单元神经网络GRUgate recurrent unit2种深度学习算法的多参数、非线性拟合能力,提出隧道大体积混凝土内部温度变化的滚动预测方法,并依托衢州市智慧新城三江中路连通隧道工程建设中采用的双光栅传感器现场混凝土温度实测数据,采用平均绝对误差MAEmean absolute error)和决定系数(R2)对2类模型的预测结果精度进行检验评价。结果表明: 2种网络模型均能捕捉隧道大体积混凝土内部温度发展规律,准确预测隧道大体积混凝土内部温度变化曲线,且GRU的精度优于LSTM。其中,GRUMAE1.34 ℃,比LSTM减小1.07 ℃,同时GRU的R20.98,也优于LSTM的R20.9)。

关键词: 大体积混凝土, 深度学习, 混凝土浇筑, 温度预测, 明挖隧道

Abstract: The internal temperature change in mass tunnel concrete can be investigated in real time by formulating timely and reasonable temperature control measures that can prevent the temperature cracks caused by the largescale hydration heat reaction during the pouring process of mass tunnel concrete. A rolling prediction method is proposed for the internal temperature change of mass tunnel concrete based on the multiparameter and nonlinear fitting ability of two deep learning algorithms, namely, the long shortterm memory (LSTM) and the gate recurrent unit (GRU). Relying on the onsite concrete temperature measurement data of the doublegrating sensor used to construct the Sanjiang Middle road connecting tunnel project in Quzhou Smart New City, the mean absolute error (MAE) and the WT5《TNR#I》〗RWT5《TNR》〗squared (WT5《TNR#I》〗RWT5《TNR》〗2) values are used herein to test and evaluate the prediction result accuracies of the two models. The results show that the two network models can capture the internal temperature development law of mass tunnel concrete and accurately predict its internal temperature change curve. Furthermore, the GRU accuracy is better than that of LSTM. The MAE of GRU is 1.34 , which is 1.07 lower than that of LSTM, and the WT5《TNR#I》〗RWT5《TNR》〗2 of GRU is 0.98, which is better than that of LSTM (i.e., 0.9).

Key words: mass concrete, deep learning, concrete pouring, temperature prediction, opencut tunnel