复合地层地质参数与水下大直径盾构掘进参数相关性预测研究  

Research on Correlation Prediction between Geological Parameters of Composite Strata and Underwater Large Diameter Shield Tunneling Parameters

作　　者：范文超 卢高明 张兵 周建军 FAN Wenchao;LU Gaoming;ZHANG Bing;ZHOU Jianjun(State Key Laboratory of Shield Machine and Boring Technology,Zhengzhou 450001,China;China Railway Tunnel Group Co.,Ltd.,Guangzhou 511458,China)

机构地区：[1]盾构及掘进技术国家重点实验室,郑州450001 [2]中铁隧道局集团有限公司,广州511458

出　　处：《铁道标准设计》2023年第8期129-135,共7页Railway Standard Design

基　　金：国家重点研发计划项目(2020YFF0218001);国家自然科学基金项目(42002281);河南省自然科学基金项目(202300410002,212300410325)。

摘　　要：为探索复合地层地质参数与水下大直径盾构掘进参数的内在联系,以汕头海湾隧道工程为背景,采用BP神经网络方法构建相关性预测模型,实现从地质参数到掘进参数的量化预测。研究结果表明:(1)通过建立基于掘进环数的地质参数及盾构稳定掘进时的掘进参数数据库,可有效表征地质及设备特性;(2)通过数据预处理、优选网络结构及训练函数,构建基于BP神经网络的相关性预测模型,总推进力、刀盘转速、掘进速度和刀盘扭矩平均误差小于6%,最大误差小于18%,预测效果较好,基本达到工程要求。基于BP神经网络相关性预测模型,输入地质参数即可定量预测掘进参数,结构简单、运算方便、精度较高。In order to explore the internal relationship between geological parameters of composite strata and underwater large diameter shield tunneling parameters,the correlation prediction model based on Shantou Bay Tunnel project is constructed with BP neural network method to realize the quantitative prediction from geological parameters to shield tunneling parameters.The research results show that:(1)The geological and equipment characteristics can be effectively represented by establishing the databases of geological parameters based on the number of driving rings and tunneling parameters during shield stable driving.(2)Through data preprocessing,optimization of network structure and training function,a correlation prediction model based on BP neural network is constructed.The average error of total thrust,cutter head speed,tunneling speed and cutter head torque is less than 6%,and the maximum error is less than 18%,which basically meet the engineering accuracy requirements.It is concluded that the shield tunneling parameters can be predicted quantitatively by inputting geological parameters based on the BP neural network correlation prediction model featured by simple structure,convenient operation and high precision.

关 键 词：复合地层 水下大直径盾构 盾构掘进参数 BP神经网络 相关性预测模型 

分 类 号：U25[交通运输工程—道路与铁道工程] U455.43[建筑科学—桥梁与隧道工程] U459.5