Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2022, Vol. 56 Issue (8): 1523-1532    DOI: 10.3785/j.issn.1008-973X.2022.08.006
土木与交通工程     
基于PSO-SVR的土压平衡盾构施工进度优化
秦元1(),余宏淦2,陶建峰2,*(),孙浩2,刘成良2
1. 上海隧道工程有限公司，上海 200232
2. 上海交通大学 机械与动力工程学院，上海 200240
Advance rate optimization of earth pressure balance shield based on PSO-SVR
Yuan QIN1(),Hong-gan YU2,Jian-feng TAO2,*(),Hao SUN2,Cheng-liang LIU2
1. Shanghai Tunnel Engineering Co. Ltd, Shanghai 200232, China
2. School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai 200240, China
 全文: PDF(2477 KB)   HTML
摘要：

针对隧道掘进机(TBM)利用率预测研究匮乏的问题，建立数据驱动的利用率预测模型并进一步对施工进度展开优化. 结合新加坡某地铁隧道项目数据，研究地质类型、司机操作与载荷对TBM利用率的影响，提出基于支持向量回归(SVR)的利用率预测方法，并以施工进度最大为目标展开操作参数优化. 利用SVR建立掘进环利用率与地质类型、载荷、操作参数的映射模型；建立以施工进度最大为目标，以地质类型、载荷、操作参数为约束边界的优化方程；利用粒子群优化(PSO)寻找特定地质类型下最优的操作参数. 结果表明：SVR模型在验证集和测试集上的R2分别为0.729和0.625，均优于多元线性回归、决策树、k最近邻、随机森林、AdaBoost和XGBoost模型；PSO能准确地找出最优的操作参数.
关键词： 土压平衡盾构;  施工进度优化;  利用率预测;  PSO-SVR;  操作参数    
Abstract:

A data-driven utilization factor prediction model was established and the advance rate was further optimized, in view of the lack of research on the utilization factor prediction of tunnel boring machine (TBM). The effects of geological type, driver operation and loads on TBM utilization factor were studied on the basis of on-site data collected from a metro tunnel project in Singapore. A utilization factor prediction method based on support vector regression (SVR) was proposed, and the operational parameters optimization was further carried out with the goal of maximizing the advance rate. SVR was used to build the mapping model between the utilization factor of tunneling cycle and the geological type, loads and operational parameters. Then, optimization equations were established with the maximum advance rate as the objective and the geological type, loads and operational parameters as the constraints. Finally, particle swarm optimization (PSO) was applied to optimize the equations and find the optimal operational parameters under specific geological type. Results showed that R2 of SVR model on the validation set and the test set were 0.729 and 0.625 respectively, which were better than that of multiple linear regression, decision tree, k-nearest neighbors, random forest, AdaBoost and XGBoost models. Moreover, PSO can accurately find out the optimal operational parameters.
Key words: earth pressure balance shield    advance rate optimization    utilization factor prediction    PSO-SVR    operational parameter
收稿日期: 2021-08-27 出版日期: 2022-08-30
CLC:  TH 69  
基金资助: 教育部-中国移动联合基金建设项目(MCM20180703)；国家重点研发计划课题资助项目(2018YFB1702503)
通讯作者: 陶建峰     E-mail: qinyuan@stecmc.com;jftao@sjtu.edu.cn
作者简介: 秦元（1983—），男，高级工程师，从事盾构智能化研究. orcid.org/0000-0002-9446-0977. E-mail： qinyuan@stecmc.com
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
秦元
余宏淦
陶建峰
孙浩
刘成良

引用本文:

秦元,余宏淦,陶建峰,孙浩,刘成良. 基于PSO-SVR的土压平衡盾构施工进度优化[J]. 浙江大学学报(工学版), 2022, 56(8): 1523-1532.

Yuan QIN,Hong-gan YU,Jian-feng TAO,Hao SUN,Cheng-liang LIU. Advance rate optimization of earth pressure balance shield based on PSO-SVR. Journal of ZheJiang University (Engineering Science), 2022, 56(8): 1523-1532.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2022.08.006    或    https://www.zjujournals.com/eng/CN/Y2022/V56/I8/1523

图 1  支持向量回归示意图
图 2  基于PSO-SVR的施工进度优化方法流程图
图 3  施工区间示意图
技术参数 设计值
D/mm 6720
P/kW 8×160
Tmax/(kN·m) 8322
n/ (r?min?1) 0~4
Fmax/ kN 49220
vmax/( mm?min?1) 80
表 1  EPB主要技术规格
钻孔编号 上隧道 下隧道
R 地质类型 R 地质类型
1 500 句容组(V) 500 句容组(IV)
2 388 句容组(V) 388 句容组(V)
3 302 句容组(IV) 302 句容组(复合地层)
4 265 句容组(V) 265 句容组(复合地层)
5 229 句容组(V) 229 句容组(复合地层)
6 202 句容组(IV) 202 句容组(IV)
7 192 句容组(V) 192 句容组(IV)
8 137 坎宁堡卵石层 137 坎宁堡卵石层
9 123 坎宁堡卵石层 123 ?
10 119 坎宁堡卵石层 119 坎宁堡卵石层
11 117 坎宁堡卵石层 117 坎宁堡卵石层
12 98 河流黏土 98 坎宁堡卵石层
13 96 河流黏土 96 坎宁堡卵石层
14 63 海洋黏土 64 古冲积层(复合地层)
15 44 海洋黏土 44 河流黏土(复合地层)
表 2  钻孔对应的环号与地质类型
图 4  上隧道第90环推力随系统状态的变化
图 5  上隧道EPB掘进状态下关键参数的分布
图 6  上、下隧道各环利用率
图 7  上、下隧道不同地质类型下利用率的统计
图 8  操作参数对利用率的影响
图 9  不同地质类型下扭矩和推力随贯入度的变化
数据集 数据来源 输入特征 回归变量 样本量
训练集 上隧道 T、F、n、v、geo U 67
测试集 下隧道 T、F、n、v、geo U 42
表 3  训练集与测试集信息
图 10  SVR模型的网格搜索法寻优
图 11  SVR模型在验证集与测试集上的表现
图 12  各算法模型在测试集上的表现
算法 模型关键参数 验证集 测试集
MSE R2 MSE R2
MLR ? 0.005 9 0.602 0.007 2 0.396
RR alpha=100.0 0.005 5 0.630 0.005 6 0.529
DT 'max_depth': 2, 'max_features': 8 0.004 5 0.695 0.028 3 ?1.374
KNN 'n_neighbors': 15 0.007 1 0.520 0.009 0 0.245
SVR kernel='linear',C=0.002 0.003 8 0.729 0.004 5 0.625
RF 'max_depth': 2, 'max_features': 8, 'n_estimators': 25 0.004 4 0.705 0.008 9 0.247
AdaBoost 'learning_rate': 0.015, 'loss': 'linear', 'n_estimators': 7 0.006 8 0.540 0.012 5 ?0.046
XGBoost 'max_depth': 1, 'n_estimators': 10 0.006 5 0.558 0.009 4 0.214
表 4  各算法模型在验证集和测试集上的表现
图 13  PSO寻优时目标函数值的变化
图 14  PSO在不同地质类型下的寻优结果
地质类型 n*/(r?min?1) v*/(mm?min?1) AR*/(mm?min?1)
句容组(V) 3.50 31.60 8.89
句容组(IV) 3.50 33.43 9.34
坎宁堡卵石层 3.50 31.81 8.78
河流黏土 1.24 30.45 7.49
海洋黏土 3.50 27.68 6.45
表 5  不同地质类型下PSO的寻优结果
1 LIU Q, HUANG X, GONG Q, et al Application and development of hard rock TBM and its prospect in China[J]. Tunnelling and Underground Space Technology, 2016, 57: 33- 46
doi: 10.1016/j.tust.2016.01.034
2 YU H, JAO J, HUANG S, et al A field parameters-based method for real-time wear estimation of disc cutter on TBM cutterhead[J]. Automation in Construction, 2021, 124: 103603
doi: 10.1016/j.autcon.2021.103603
3 王厚同. 基于机器学习的TBM性能预测方法 [D]. 济南: 山东大学, 2019.
WANG Hou-tong. Method for predicting TBM performance based on machine learning [D]. Jinan: Shandong University, 2019.
4 ROSTAMI J Performance prediction of hard rock tunnel boring machines (TBMs) in difficult ground[J]. Tunnelling and Underground Space Technology, 2016, 57: 173- 182
doi: 10.1016/j.tust.2016.01.009
5 ZHOU J, BEJARBANEH B Y, ARMAGHANI D J, et al Forecasting of TBM advance rate in hard rock condition based on artificial neural network and genetic programming techniques[J]. Bulletin of Engineering Geology and the Environment, 2019, 79 (2): 1- 16
6 JING L, LI J, ZHANG N, et al A TBM advance rate prediction method considering the effects of operating factors[J]. Tunnelling and Underground Space Technology, 2021, 107: 103620
doi: 10.1016/j.tust.2020.103620
7 段晓晨, 张小平 TBM掘进工时利用率动态优化系统研究[J]. 铁道学报, 2000, (6): 104- 108
DUAN Xiao-chen, ZHANG Xiao-ping Research on dynamic optimized system of increasing constantly the TBM boring operation rate[J]. Journal of the China Railway Society, 2000, (6): 104- 108
doi: 10.3321/j.issn:1001-8360.2000.06.023
8 龚秋明, 卢建炜, 魏军政, 等 基于岩体分级系统(RMR)评估预测TBM利用率研究[J]. 施工技术, 2018, 47 (5): 92- 98,127
GONG Qiu-ming, LU Jian-wei, WEI Jun-zheng, et al Study on estimation and prediction of TBM utilization rate using rock mass rating (RMR)[J]. Construction Technology, 2018, 47 (5): 92- 98,127
9 ALBER M, RÜHL S. Tunnel support by fully grouted rock bolts for fast TBM advance [C]// Proceedings of 31st ITA-AITES World Tunnel Congress. London: Taylor & Francis Group, 2005: 743-748.
10 BRULAND A. Hard rock tunnel boring [D]. Trondheim: Norwegian University of Science and Technology, 2000.
11 FARROKH E. Study of utilization factor and advance rate of hard rock TBMs [D]. Stetter Klich: Pennsylvania State University, 2012.
12 SIMOES M G, KIM T. Fuzzy modeling approaches for the prediction of machine utilization in hard rock tunnel boring machines [C]// Conference Record of the 2006 IEEE Industry Applications Conference 41st IAS Annual Meeting. Tampa: IEEE, 2006: 947-954.
13 FROUGH O, TORABI S R An application of rock engineering systems for estimating TBM downtimes[J]. Engineering Geology, 2013, 157: 112- 123
doi: 10.1016/j.enggeo.2013.02.003
14 NOORI A M, MIKAEIL R, MOKHTARIAN M, et al Feasibility of intelligent models for prediction of utilization factor of TBM[J]. Geotechnical and Geological Engineering, 2020, 38 (3): 3125- 3143
doi: 10.1007/s10706-020-01213-9
15 王超, 龚国芳, 杨华勇, 等 NSVR硬岩隧道掘进机刀盘扭矩预测分析[J]. 浙江大学学报:工学版, 2018, 52 (3): 479- 486
WANG Chao, GONG Guo-fang, YANG Hua-yong, et al NSVR based predictive analysis of cutterhead torque for hard rock TBM[J]. Journal of Zhejiang University: Engineering Science, 2018, 52 (3): 479- 486
16 KENNEDY J, EBERHART R. Particle swarm optimization [C]// Proceedings of ICNN'95-International Conference on Neural Networks. Perth: IEEE, 1995, 4: 1942-1948.
17 姜长泓, 张永恒, 王盛慧 基于改进粒子群优化算法的PID控制器参数优化[J]. 应用科学学报, 2017, 35 (5): 667- 674
JIANG Chang-hong, ZHANG Yong-heng, WANG Sheng-hui PID parameter optimization based on improved particle swarm optimization algorithm[J]. Journal of Applied Sciences, 2017, 35 (5): 667- 674
doi: 10.3969/j.issn.0255-8297.2017.05.012
18 PARSOPOULOS K E, VRAHATIS M N Particle swarm optimization method for constrained optimization problems[J]. Intelligent Technologies-Theory and Application: New Trends in Intelligent Technologies, 2002, 76 (1): 214- 220
[1] 刘蕴, 董冠华, 殷国富. 非接触密封失稳振动分析与结构优化[J]. 浙江大学学报(工学版), 2018, 52(7): 1390-1397.
[2] 李忠超, 陈仁朋, 孟凡衍, 叶俊能. 软黏土中盾构掘进地层变形与掘进参数关系[J]. 浙江大学学报(工学版), 2015, 49(7): 1268-1275.
[3] 李忠超, 陈仁朋, 孟凡衍, 叶俊能. 软黏土中盾构掘进地层变形与掘进参数关系[J]. 浙江大学学报(工学版), 2015, 49(4): 2-3.
[4] 卢浩,王明洋,夏沅谱,戎晓力. 土压平衡盾构刀盘扭矩计算模型[J]. 浙江大学学报(工学版), 2014, 48(9): 1640-1645.
[5] 梁荣柱,潘金龙,林存刚,单华峰,孙廉威. 软土地区盾构施工沉降界限[J]. 浙江大学学报(工学版), 2014, 48(7): 1148-1154.
[6] 梁荣柱,潘金龙,林存刚,单华峰,孙廉威. 软土地区盾构施工沉降界限[J]. J4, 2014, 48(2): 0-00.
[7] 周俊虎 宋国良 刘建忠 岑可法. 浓相粉体给料器流化特性影响因素的分析[J]. J4, 2006, 40(7): 1168-1173.