盾构掘进会引起地表变形,当地表变形严重时会导致财产损失和人员伤亡。开展盾构掘进引起地表变形研究,可以提前感知盾构隧道施工风险,为盾构隧道施工安全提供保障。本文采用人工水母算法优化(AJSO)标准BP神经网络算法,建立了AJSO-BP神经网络预测模型,并将AJSO-BP、GA-BP与BP神经网络预测模型进行对比,为实际工程选择了精度最高的预测模型,并验证了预测模型的工程适用性。结果表明:标准BP神经网络模型的预测精度最低,预测值与实测值之间的相对误差为43.6%;AJSO-BP神经网络模型的预测精度最高,预测值与实测值之间的相对误差为6%,人工水母搜索算法能够显著优化标准BP神经网络;工程案例分析中,AJSO-BP神经网络模型的预测值与实测值较为接近,证明了AJSO-BP神经网络模型在实际工程应用效果较好。
Shield tunneling can cause surface deformation, which can lead to property damage and casualties when the surface deformation is severe. Conducting research on surface deformation caused by shield tunneling can provide early awareness of shield tunnel construction risks and guarantee the safety of shield tunnel construction. This article uses the artificial jellyfish algorithm to optimize the standard BP neural network algorithm, establishes an AJSO-BP neural network prediction model, and compares the AJSO-BP, GA-BP, and BP neural network prediction models to select the most accurate prediction model for practical engineering, and verifies the engineering applicability of the prediction model. The research results indicate that the standard BP neural network model has the lowest prediction accuracy, with a relative error of 43.6% between predicted and measured values. The AJSO-BP neural network model has the highest prediction accuracy, with a relative error of 6% between predicted and measured values. The artificial jellyfish search algorithm can significantly optimize the standard BP neural network. The engineering case analysis shows that the predicted values of the AJSO-BP neural network model are relatively close to the measured values, proving that the AJSO-BP neural network model has a good effect in practical engineering applications.
[1] Kong F C, Lu D C, Ma Y D, et al. Analysis and intelligent prediction for displacement of stratum and tunnel lining by shield tunnel excavation in complex geological conditions: A case study[J]. Ieee Transactions on Intelligent Transportation Systems, 2022, 23(11): 22206-22216.
[2] 杨帆, 胡晋, 孙彩霞. InSAR技术结合深度学习的范围性地表沉降预测[J]. 测绘科学,2022, 47(7): 60-68, 134.
[3] 戴田. 基于长短期记忆网络的盾构掘进地表纵向沉降预测研究[D].长沙:湖南大学, 2021.
[4] 陈仁朋, 戴田, 张品, 等. 基于机器学习算法的盾构掘进地表沉降预测方法[J]. 湖南大学学报(自然科学版), 2021, 48(7): 111-118.
[5] 王宇哲. 基于统计机器学习的盾构隧道地表沉降预测模型及优化方法研究[D]. 北京:北京交通大学, 2021.
[6] 王鹏. 基于机器学习的地铁隧道施工扰动控制研究[J]. 现代隧道技术, 2019, 56(增2): 368-373.
[7] 甘鹏路. 富水软弱地层浅埋暗挖隧道地层变形规律及预测研究[D]. 杭州:浙江大学, 2016.
[8] 崔庆龙. 岩溶地层盾构施工引起地表沉降的规律及预测研究[J]. 现代隧道技术, 2021, 58(增2): 110-119.
[9] 杨志勇, 朱钢. 富含生活垃圾土层浅埋暗挖隧道变形实测分析[J].地下空间与工程学报, 2023, 19(增1): 345-352.
[10] 马伟亮, 李顺群, 叶茂松, 等. 基坑开挖对邻近隧道变形影响的解析方法研究[J].地下空间与工程学报, 2023, 19(2): 447-455.
[11] Deng H S, Fu H L, Yue S, et al. Ground loss model for analyzing shield tunneling-induced surface settlement along curve sections[J]. Tunnelling and Underground Space Technology, 2022, 119: 104250.
[12] Peck R B. Deep excavations and tunnelling in soft ground[A] // In: 7th International Conference on Soil Mechanics and Foundation Engineering[C]. Mexico, 1969: 225-290.
[13] 刘宇豪. 基于机器学习算法的浅埋隧道地表沉降预测模型研究[D]. 重庆:重庆交通大学, 2022.
[14] Abdel-basset M, Mohamed R, Abouhawwash M, et al. An improved jellyfish algorithm for multilevel thresh olding of magnetic resonance brain image segmentations[J]. Computers, Materials and Continua, 2021, 68(3): 2961-2977.
