Abstract In tunnel boring machine (TBM) operations, a large volume of feedback data related to the excavation process is collected in real time, which directly reflects the interaction between the TBM and surrounding rock. The key challenge in ensuring efficient and safe tunneling is to accurately extract sensitive features from vast, fluctuating, and complex multidimensional time series data to understand the geological conditions ahead of the cutterhead. In this study, a novel intelligent recognition method for TBM rock mass classification is proposed, based on piecewise aggregate approximation (PAA) and a multichannel one-dimensional convolutional neural network (MC-1D-CNN). The method effectively maps the interaction between the TBM and surrounding rock to the associated rock classes information ahead of the cutterhead. Five tunneling parameters are selected as input features. First, the PAA technique is applied to transform the multidimensional, raw TBM time series data, which exhibit different scales, into fixed-length one-dimensional feature vectors. This process reduces dimensionality and fuses the data along spatial channels. Then, a MC-1D-CNN model is constructed, which takes the multichannel PAA sequence matrix as input for training. Through multilayer nonlinear mappings and dynamic optimization, the model automatically learns the features of univariate time series within each channel, further reducing the dimensionality and extracting high-order features more strongly correlated with the rock classes. Additionally, to enhance the feature extraction ability and generalization capability of the model, a MC-1D dilated CNN (MC-1D-DCNN) is developed by incorporating dilated convolutions. The method is validated using TBM monitoring data from the second section of the Yinchao project, demonstrating its capability to accurately classify the rock classes. Comparative analysis with several ensemble machine learning models based on key tunneling parameter time series statistical features shows that the proposed TBM surrounding rock level intelligent method PAA-MC-1D-CNN has better feature mining ability and higher recognition accuracy and robustness.
Abstract This study introduces a physics-informed deep learning framework for modeling the viscoelastic behavior of asphalt binders using a fractional-order Kelvin–Voigt series model embedded within a physics-informed neural network (PINN). By enforcing Kramers–Kronig constraints and integrating depth-frequency-dependent input features, the proposed fractional-order constitutive PINN (FO-PINN) framework captures broadband dynamic modulus behavior with high accuracy (global R 2 > 0.96 ). A unique material embedding strategy enables the network to generalize across different binder types while preserving material-specific characteristics. Sensitivity analysis reveals that long-term viscosities ( η 2 , η 3 ) and the fractional derivative order ( α ) dominate the rheological response, reflecting the multiscale nature of asphalt microstructure. The model demonstrates superior performance compared to conventional identification methods in terms of accuracy, generalizability, and physical consistency. Furthermore, its capability to decode depth-resolved viscoelastic gradients and support cross-material parameter estimation highlights its potential as a foundational tool for digital twins of pavement systems. These findings promote the adoption of physics-constrained machine learning in civil material informatics and infrastructure resilience assessment.
Abstract The tunneling parameters of tunnel boring machines (TBMs) exhibit distinct spatiotemporal characteristics under different rock mass conditions, particularly their non-Euclidean spatial properties. This paper developed a real-time rock mass identification model based on the spatial-temporal graph convolutional network (STGCN) to simultaneously capture the spatiotemporal characteristics of tunneling parameters. TBM tunneling data and geological data were collected from an 8.41-km-long tunneling section of the YinChaoJiLiao (YCJL) project in China. By matching the data based on tunnel mileage, 2,367 valid samples were obtained. A downsampling strategy was applied to reduce data redundancy and computational cost while maintaining model performance. The optimized model achieved an accuracy of 93.04% and an F 1 -score of 89.75%, and comparative experiments verified its superior performance compared with the mainstream model in this task. A statistical analysis of the adjacency matrix derived from the graph convolutional network (GCN) module was conducted and the influence of input time length on model performance was analyzed to interpret the spatial and temporal characteristics of the tunneling parameters, respectively. Furthermore, the model was validated on a extension section of the same tunnel, confirming its accuracy and efficiency for practical application. The results indicate that the STGCN model can simultaneously capture the complex spatiotemporal characteristics of TBM data and provide accurate real-time rock mass identification results, showing great potential for engineering application in intelligent tunneling.
Abstract Intelligent transportation systems rely on accurate perception technologies for traffic monitoring, with light detection and ranging (LiDAR) emerging as a critical sensing modality. However, effective background–foreground separation is challenging due to the varying measurement reliability across point clouds caused by distance, reflectivity, and atmospheric conditions. This paper presents fast reliability-weighted Gaussian background modeling for 3D LiDAR traffic monitoring (FRGB3D), a novel unsupervised approach for LiDAR point cloud background modeling and foreground detection that explicitly incorporates physics-based reliability assessment with adaptive learning mechanisms. The key contributions include (1) a physics-based reliability metric that quantifies measurement quality using range and reflectivity characteristics, (2) a reliability-weighted adaptive learning approach for Gaussian background modeling that dynamically adjusts learning rates based on point reliability, and (3) an adaptive thresholding mechanism that incorporates both measurement reliability and learned model uncertainty. Through comprehensive experiments using both the Car Learning to Act (CARLA) simulator and real-world LiDAR data collected in Lowell, Massachusetts, we demonstrate that FRGB3D achieves superior foreground detection performance compared with baseline methods. FRGB3D achieved F1-scores of 0.84, 0.68, and 0.77 across base, adverse-weather, and high-density traffic scenarios, respectively, consistently outperforming simple difference, moving average, and standard Gaussian background modeling approaches. Real-world experiments demonstrated competitive object-level performance average precision (AP) (AP@0.5 = 0.48) while maintaining real-time processing capabilities with 19.1 frames per second (FPS), making it suitable for practical traffic monitoring applications.
Abstract In order to overcome the limitation of traditional road damage detection algorithms such as single detection type, real-time, and lightweight scenarios, a lightweight multilevel feature fusion algorithm HCC-You Only Look Once (HCC-YOLO) is proposed in this paper. First, improved HGNetv2 is adopted as a backbone network; a lightweight spatial attention module is integrated; activation function is optimized to enhance damage area location and spatial expression, and a very low-cost innovative module (SiGE) replaces C2PSA to improve feature enhancement efficiency. Secondly, a lightweight neck network CHP with deep optimization is designed to reduce redundancy through feature screening and cross-scale interaction and improve adaptability to complex scenes. Thirdly, a multitask lightweight task dynamic alignment detection head (LTADH) is proposed to enhance multiscale damage feature extraction by combining task decoupling and dynamic alignment. Experiments on RDD2022 data set show that the HCC-YOLO model reduces 61.82% in size, 44.44% in computation, and 66.67% in parameters (less than 1 M), while mAP@0.5 reaches 88.2%. (1.6% higher than the baseline module YOLOv11n). The algorithm model can be deployed on edge computing devices carried by drones to quickly detect cracks, potholes, and other damages on road surfaces, so as to repair roads in time, ensure road driving safety, and realize green, intelligent, and efficient operation and maintenance.
