课程文献中心 更多>>

关键词： Tunnel boring machine   Time series forecasting   Cutterhead torque   Thrust force   Recurrent neural network   TUNNEL CONSTRUCTION   PERFORMANCE   PREDICTION  

摘要： Tunnel Boring Machines (TBMs) are instrumental in the construction of modern tunnels, known for their operational reliability and efficiency. The real-time prediction of cutterhead loads is essential for effective project scheduling, cost management, and risk reduction. This study develops machine learning models for predicting future cutterhead torque and thrust force, simultaneously. The dataset is derived from the Yingsong water diversion project and consists of 12,962 steady-state boring cycles. Because geological conditions and setting values are available in advance, we build a geological parameter-based model and a setting value-based model for regression analysis of cutterhead torque and thrust force. In one-step forecasts, the operational parameterbased model closely captures the trend of the measured results, employing the recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU). We propose an aware-context recurrent neural network (AC-RNN) model that integrates historical operational parameters and the aware context of setting values, leading to a significant improvement in forecasting accuracy. A sensitivity analysis is carried out to quantify the relative importance of input parameters, revealing that the setting value of revolutions per minute is crucial in forecasting. The results of this study offer valuable practical insights into real-time forecasting, thereby informing engineering applications in this field.

关键词： Tunneling boring machine   Shield scraper   Gravelly soil   Cutting force   Specific energy   TBM   MODEL   STRENGTH   TESTS  

摘要： Gravelly soil strata exhibit heterogeneity and nonlinearity in their physical and mechanical properties, leading to volatile fluctuations of shield scraper force. Understanding the performance of shield scrapers in gravelly soils is significant for the safe and efficient excavation of tunnel boring machines. This paper conducts unconsolidatedundrained triaxial tests to obtain the mechanical properties of gravelly soils with gravel content ranging from 0 % to 30 %. The process of shield scrapers cutting through gravelly soils is analyzed by combining the varying mechanical properties of gravelly soils with the limit equilibrium analysis method. Subsequently, modified models for predicting the scraper force and specific energy in gravelly soils are established. Based on these models, the impact of key factors on the scraper performance is analyzed. Laboratory experiments are further performed on a rotary test bench to validate these models. The experimental results demonstrate that the horizontal cutting force and specific energy of shield scrapers in gravelly soils can be predicted with mean average errors of 8.8% and 7.3%, respectively, with the errors of all predicted values falling within +/- 20 % of the experimental results. These established models can serve as useful references for the structural and operational design of shield scrapers in gravelly soil strata.

关键词： TBM tunnelling   Microseismic monitoring   Microseismic identification   Deep learning   Multi-algorithm fusion   CONVOLUTIONAL NEURAL-NETWORKS   WAVE-FORM RECOGNITION   HARD-ROCK TBM   CLASSIFICATION   BLASTS   DISCRIMINATION   PREDICTION   PROSPECT   FEATURES  

摘要： For TBM's safe and efficient tunnelling, the microseismic monitoring technique has been widely applied in rockburst warning. However, useful microseismic events are often mixed with noisy events, which seriously affects warning accuracy. To this end, this study proposes several hybrid deep learning-based microseismic identification approaches in TBM tunnelling, where recurrent neural networks directly treat monitored microseismic events as time series and avoid complex feature pre-extraction, and grey wolf optimization and attention mechanism are embedded to optimize model hyper-parameters and recognize value information. Additionally, the learning curve and early-stopping strategy are also integrated to reduce overfitting and underfitting risks. Results in a real TBM-excavated water conveyance tunnel indicate that the bidirectional long short-term memory network model combined with grey wolf optimization and attention mechanism achieves the greatest identification performance among discussed models and provides an effective support for intelligent identification of microseismic events under TBM tunnelling environments.

关键词： Cerchar Abrasivity Index (CAI)   Mechanical properties   Rock mass classification parameters/system   Machine performance   Regression analysis   Estimation performance evaluation   PERFORMANCE PREDICTION   GEOMECHANICAL PROPERTIES   ABRASIVENESS INDEX   PENETRATION RATE   GRANITIC-ROCKS   GRANULAR SOIL   TOOL WEAR   CAI   REGRESSION   STRENGTH  

摘要： Rock abrasivity plays an important role in the machine design, construction scheduling, and budgeting of TBM projects. Establishing several faster and simpler estimation equations for the Cerchar Abrasivity Index (CAI) of rocks is, therefore, very important. This study investigated the correlation between the CAI and mechanical properties of rock, rock mass classification parameters, and machine performance. A TBM construction database including 159 tunnel sections is established. Several acceptable and practical estimation equations of CAI are developed using simple and multiple regression analysis (0.66 < R-2 < 0.76). In this process, a normalized specific energy is proposed to evaluate the machine performance. The results show that the rock compressive strength and brittleness index are the most dependent parameters to explain CAI, and the estimated rock mass strength also indicates a close correlation. In addition, the contribution of a rock mass classification system and machine performance index for estimating CAI cannot be ignored. Finally, the estimation performance of the developed equations is compared and evaluated, and a new method for selecting an optimal model based on ranking is proposed. Since the input parameters of the proposed equations can be readily available at the project planning stage, they are very practical for TBM designers, tunnel designers, and contractors.

关键词： Himalayan geology   Rock mass classification   TBM tunnelling   Machine learning   SMOTE   PREDICTION   FRONT   SMOTE  

摘要： The geological condition of the Himalayan region is very complex and challenging. So far, empirical and analytical approaches for rock mass characterization have been a common practice in the Himalayas. Due to the limitations of input parameters and governing equations in design practices, rock mass characterization in tunnel boring machine (TBM) excavated tunnels is crucial. This research introduces robust machine learning (ML) approaches to predict rock mass quality conditions in complex geological environments, leveraging a large database of TBM parameters and rock mass rating (RMR) values. To do so, a total of 6879 stable phase TBM cycle data were collected from 12 km long tunnel in Nepal. The pre-processed parameters were randomly split into a training set (80%) and a testing set (20%). Seven individual classifiers consisting of logistic regression (LR), support vector machine (SVM), decision tree (DT), random forest (RF), k-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and bagging, and stacking ensemble classifier were exploited with optimal hyperparameters. The comprehensive assessment carried out has shown that the ensemble classifier gave highest overall accuracy as compared to other individual classifiers. More importantly, the synthetic minority over-sampling technique (SMOTE) performs better to handle the imbalanced database, while the RF and stacking classifier demonstrated the best prediction performance with accuracy of 92%. Moreover, for the minority rock mass class, the RF shows better performance compared to stacking classifier. The authors emphasize that the effective application of ML-based data-driven approach shows substantial potential for rock mass characterization in TBM tunnelling. Developing a well-structured rock mass classification approach to handle imbalanced data of TBM tunnelling in complex and challenging *** pipeline of seven individual ML classifiers and stacking ensemble classifier incorporating compl

关键词： 地铁盾构隧道   掘进   同步注浆施工技术  

摘要： 地铁盾构施工是现阶段城市化进程建设背景下,地铁工程中常用的施工技术,此项施工技术的应用具有对周边建筑物、建筑用户及地质层影响小的特点,地铁盾构隧道掘进中同步注浆施工技术的应用可有效提升盾构隧道掘进施工环节的安全性及掘进质量,从而保障地铁工程建设的效率.基于此,本文主要研究了地铁盾构隧道掘进中同步注浆施工技术的详细应用,以期为相关建设人员提供参考.

关键词： TBM   振动监测   信号降噪   机器学习   围岩感知  

摘要： TBM破岩振动与掌子面在掘岩体状态密切相关。为探究TBM振动应用在围岩感知领域的可行性，本文以新疆某TBM隧道工程为例，采集不同围岩条件下的TBM主梁振动信号，采用小波阈值方法进行信号降噪，从时域、频域和时频角度分析不同围岩类别下的振动信号特征，并选取与岩石单轴饱和抗压强度（UCS）和岩体体积节理数（Jv）相关性最高的5项振动特征指标建立围岩参数感知数据集，在此基础上分别开发了BP神经网络模型和SVM模型，最终通过验证集和测试集数据对模型的感知效果进行验证和对比。研究结果表明：采用小波阈值降噪方法降噪后振动峰值及平均值与围岩类别的相关性分别提高了7.56%和12.20%。整体来看，TBM破岩振动能量主要集中在低频范围，而随围岩类别由IV类到II类，振动强度和振幅逐渐增加且高频成分的能量占比逐渐增大。在26项振动特征参数中，平均值、峰峰值、峭度、重心频率和均方根频率与围岩参数的相关性最高，Pearson相关系数都在0.7以上。模型测试发现BP神经网络模型对UCS和Jv的预测误差分别小于2%和8%，相比SVM模型同比减少50%和53%，预测效果更好。研究结果可为TBM施工围岩参数实时感知技术创新提供借鉴和参考。

关键词： 滚动轴承   振动谱图   故障诊断   双线性池化   多特征融合  

摘要： 为了深入优化滚动轴承在变负载驱动环境下特征提取不充分、轴承故障特征表征不足的问题，提出了基于双线性池化引导特征融合的轴承故障诊断算法。对读取到的原始信号数据进行预处理，通过去除直流分量、噪声滤波、抗混叠滤波、时域窗函数等操作，提高信号处理后的振动谱图质量；对预处理后的信号数据进行傅里叶变换，计算出变换后的幅值和频率数据，并绘制对应的振动谱图；利用通道注意力和空间注意力改进R es2Net网络，提取不同关注点下的视觉特征，并基于双线性池化方法进行多特征融合；利用全连接和s oftmax函数构建分类头，实现轴承故障分类。结果表明：所提出的方法在凯斯西储大学轴承数据集以及德国P aderborn数据集中的精确率分别为9 8.22%、97.94%，在轴承故障诊断中，所提算法不仅在理论上融合了自动化控制理论与控制工程原理，而且在实践中验证了其在轴承故障诊断中的有效性，为实现轴承故障的早期预警和智能诊断提供了新的技术途径。

关键词： 故障诊断   多尺度图神经网络   应急管理   风机状态参数   多状态预测   多步预测  

摘要： 大型风场中风机组分布集中， 突发的风机运行状态异常易产生传导效应， 导致风电中断或更严重问题. 事前故障预警是应急管理的关键， 风机组发生关联状态故障前采取应急管理能够有效地维护风电场的稳定运行. 然而， 现有的风机事前故障预警系统面临着预测周期短、识别精度低等挑战， 且无法实现多状态、自动化的高效预警. 基于此， 本文提出了一种面向大型风机集群的故障关联状态预测、识别及预警方法. 该方法融合了变分模态分解、卷积神经网络、多尺度图注意网络及灰狼优化算法， 通过对风机组故障关联状态的多节点多步预测及故障知识图谱实现了故障模式自动化识别及预警. 本方法在多个指标上均显著优于目前先进的基线算法， 呈现出较高的预测精度及良好的故障预警性能.

关键词： 轨道电路   时频增强残差网络   迁移学习   补偿电容故障诊断   连续小波变换  

摘要： 针对现有轨道电路补偿电容故障诊断方法在复杂环境中受到高噪声干扰，导致故障诊断精度低的问题，提出一种基于迁移学习、连续小波变换（Continuous Wavelet Transform， CWT）和时频增强残差网络（Time-Frequency Enhanced Residual Network， TFEResNet）的智能故障诊断算法.首先，采用CWT将原始感应电压信号的时域和频域信息相结合，生成小波时频图.该图能够将补偿电容故障的特征信息映射到不同时间和尺度的局部位置，有效增强模型对故障特征的捕捉能力.然后，将小波时频图输入到构建的TFEResNet模型中进行迁移学习训练，用于特征提取和故障分类.TFEResNet能够从时频图中提取复杂的时频特征，减轻信号中多余和无用噪声的不良影响，提升了诊断精度和模型的泛化能力.实验结果表明：在高噪声环境下，本文算法相较于其他方法在补偿电容故障诊断中表现出更高的准确率，达到了99.28%，同时精确率、召回率和F1评分等指标上也表现优异，证明了该方法的有效性，为基于数据驱动的轨道电路补偿电容故障诊断技术提供了一种新方法.