课程文献中心 更多>>

关键词： TBM   Reinforcement learning   Transformer   Data mining  

摘要： Research on automation and intelligent operation of tunnel boring machine(TBM)is receiving more and more attention,benefiting from the increasing construction ***,most studies on TBM operations optimization were trained by the labels of human drivers’decisions,which were subjective and *** a result,the control parameters suggested by these models could hardly surpass the performance of a human driver,even the possibility of subjective incorrect *** that the geomechanical feedback to TBM under drivers’actions is objective,in this paper,a transformer-based model called the geological response for tunnel boring machine(GRTBM),is proposed to learn the relationship between operation-adjust and TBM monitoring ***,with the model-based offline reinforcement learning,this paper provided a novel approach to optimizing the TBM excavation *** decision processes,recorded in the Yin-song TBM project for a waterway tunnel in Jilin Province of China,were used for the validation of the *** adopting an implicit perception of geological conditions in the GRTBM model,the suggested method achieved the desired state within a single action,greatly outperformed the practical adjustments where 500 s were taken,revealing the fact that the proposed model has the potential to surpass the capability of human beings.

关键词： Shallow tunnel   NorSand   Finite element method   Numerical modeling   Relative density   Sandy soil   COARSE-GRAINED SOILS   SANDS   MOVEMENTS   MODEL  

摘要： Tunnel construction in urban areas may result in ground deformations that pose a risk to existing buildings and infrastructure; thus, accurate prediction of these induced ground deformations during the design phase is crucial. The paper focuses on the effects of sandy soil relative density on the ground deformations induced by tunnels excavated with Tunnel Boring Machines (TBMs). The study utilizes the Finite Element Method (FEM) and the NorSand model to simulate the behavior of a sandy ground. The validity of the FEM modeling approach is established by comparing predictions with results from six centrifuge tunnel tests from the literature. The centrifuge tests were performed on sand at different relative densities, tunnel diameters, and tunnel depths. The parameters for the NorSand model were determined based on laboratory tests. Only the state parameter was modified to achieve the desired relative density in the numerical simulations. The effects of relative density observed in centrifuge tests (Franza et al., 2019) have been numerically reproduced with no further adjustments of the model parameters. The rich outputs from the numerical models enabled an in-depth investigation of tunnel behavior, yielding new insights into how tunnels respond under varying relative densities, depths, and diameters. A comprehensive analysis of the induced ground deformations caused by shallow tunnels in sandy ground and the potential to damage buildings is included.

关键词： TBM   Structural design   FEM   Structural monitoring   In-situ measurements   Simulation  

摘要： Different calculation methods and modeling strategies are commonly used in engineering practice to predict the internal forces in segmental tunnel linings. Accordingly, the calculation results can differ. The present paper compares internal forces resulting from five state-of-the-art calculation models, of which three continuum models and two bedded beam models, to new in-situ measurements of internal forces at the recently built TBM tunnel for the U5 metro line in Frankfurt, Germany, and discusses agreements and differences. Hereby, a generally good agreement is found for the normal forces. The predictions of the bending moments are satisfying, but there are some discrepancies. It is shown that it might be necessary to consider assembly imperfections in the calculation models to improve the agreement between the calculation results and the measurements. Also, the differences between the calculation methods themselves are addressed. It is analyzed that generally continuum models give more realistic internal forces, but that also bedded beam models can have advantages, e.g. when large parametric studies are required. Finally, recommendations are given regarding the application of calculation models for structural design.

关键词： Tunnel boring machine   Disc cutter   Frictional contact induced vibration   Discrete element method   System dynamic model   SYSTEM  

摘要： The cutterhead system of a tunnel boring machine (TBM) experiences violent vibrations owing to friction between the cutter and rock, which is a potential cause of abnormal damage accidents. However, at present, few numerical models can couple the two most important behaviours, that is, rock fragmentation and cutter vibration; thus, the cutter vibration mechanism has not been fully revealed. This paper presents an analysis method that combines discrete element method (DEM)-multibody dynamics (MBD) analysis method to link the cutter frictional contact, rock fragmentation, and vibration of a cutterhead-cutter system. The model was used to study the influence of the cutter profiles, rock properties, and cutterhead rotational speeds on the contact and vibration characteristics of the system. It was determined that, as the main external excitation of the cutterhead-cutter system, cutting force increases will lead to more violent vibration, and equivalent damping between the cutter and rock could restrain the vibration.

关键词： Full-scale cutting test   TBM   Cutting force distribution   Disc cutter   Rock-breaking process   Mawan tunnel   CONFINING STRESS   ROCK FRAGMENTATION   PERFORMANCE   EFFICIENCY   EXCAVATION   PREDICTION  

摘要： Understanding the distribution of cutting force provides a foundation for the design and optimization of cutterhead layout. The simulation of rotary cutting with a large cutting radius remains a challenge for laboratory tests. In this study, taking the driving condition of the Mawan Tunnel located in the Guangdong-Hong Kong-Macao Greater Bay Area as a case study, a full-scale rotary cutting platform with large cutting radius was used to investigate the rock-breaking effects. Rotary cutting with cutting depths of 1-8 mm and cutting radii of 200-1400 mm were carried out on the granite testing platform using a 19-in disc cutter. The normal force, rolling force, and side force during the cutting process were systematically monitored. Moreover, the time-frequency characteristics of the rock breaking process and the effects of cutting depth, cutting radius and stress state on the average and peak loads were investigated. The results show that the average normal force and rolling force increased with increasing cutting depth, while the side force was less affected by cutting depth through the Boltzman model analysis. The average normal force and rolling force were not significantly affected by the cutting radius, but the average side force decreased as a power function with increasing cutting radius, and showed a stable side force when the installation radius exceeded 1000 mm. The dominant frequency of the cutting force was 0-1.2 Hz, exhibiting notable low-frequency characteristics. Additionally, the ratio of peak to average (RPA) with respect to triaxial cutting force demonstrated an increase with the growth of cutting depth, while RPA in terms of side force exhibited a corresponding growth with the increasing cutting radius. In particular, the cutting force characteristics of the cutterhead obtained in this study can inform the layout of disc cutters and real-life TBM operations, thereby avoiding severe overloading of internal disc cutters. A multifunctional rotary cutti

关键词： 桩基位移   双线盾构隧道   数值模拟   正交试验  

摘要： 为研究双线盾构隧道施工对邻近单桩的位移影响，以长沙市地铁6号线为背景，建立三维有限元模型，分析了桩前3m处地表位移与桩身位移的变化规律。同时，通过正交试验定量分析了桩隧间距、隧道间距和土仓压力对桩身位移的影响。结果表明：地表沉降曲线受桩基影响表现出偏态分布，且竖向、水平两个方向的地表最大位移均出现在右侧。桩身水平位移呈现“顺时针旋转”的形式，其旋转点随着施工隧道的接近而向桩端移动；桩身竖向位移表现为先隆起再下沉；桩身纵向位移在盾构穿越过程中变化明显，施工完成后基本复原。此外，对于桩身最大竖向、水平位移，桩隧间距的影响程度分别为I级显著和不显著，隧道间距的影响程度分别为III级显著和II级显著，土仓压力的影响程度均为不显著。基于试验结果，建立多元线性回归方程，可为类似工程预测建筑物桩基最大位移提供参考。

关键词： 循环制动   鼓式制动器   温升   迭代计算   数学模型  

摘要： 针对某载重特种车辆循环制动工况下的鼓式制动器温升计算问题，基于热力学理论和一定的假设条件，将其制动过程分解为若干个循环，并将每个循环划分为4个阶段（最后一个循环仅有减速阶段），建立循环制动工况下鼓式制动器温升计算的数学模型。结合特定的循环制动试验参数，经过迭代计算，得到每个循环各阶段制动器温度并绘制前、后轮制动器的温升曲线。温升试验结果表明，采用所建立的数学模型计算得到的温升结果与制动试验温升结果基本吻合，误差不超过5.3%，验证了计算模型的有效性。

关键词： 故障诊断   自编码器   深度神经网络   特征提取   未知故障检测  

摘要： 在实际的工业环境中，随着机械设备运行时间的增加，往往会出现一些新的未知故障。此时，传统的故障诊断方法并不能及时发现这些新的故障，从而影响正常生产进度，造成生产损失。因此，本文提出了一种基于多解码器的自编码分类（Multi-decoders AE Classifier, MDAEC）模型，在对已知样本进行分类的同时保持对未知故障样本的检测。利用自编码器的特征提取能力，将中间的隐藏特征输入到一个分类网络中，使用已知样本对自编码器以及分类网络进行训练。根据已知样本类别训练各自的解码器，以提高模型对未知样本的检测能力。在检测时，通过自编码器的重构误差以及阈值可判断输入样本是否为新的未知故障，如果判断输入样本属于已知样本，则将隐藏层的特征输入到分类网络中即可得到具体样本类型。实验结果表明，本文所提出方法在对未知故障样本进行准确识别的同时保持着对已知样本较高的分类准确率。