Search Results (685)

Multiple uncertainties in traffic demand fluctuations and infrastructure vulnerability during seismic events pose significant challenges for the resilience assessment of highway transportation networks (HTNs). While Monte Carlo simulation remains the dominant approach for uncertainty propagation, its high computational cost limits its scalability, particularly in metropolitan-scale networks. This study proposes an EQResNet framework for accelerated post-earthquake resilience assessment of HTNs. The model integrates network topology, interregional traffic demand, and roadway characteristics into a streamlined deep neural network architecture. A comprehensive surrogate modeling strategy is developed to replace conventional traffic simulation modules, including highway status realization, shortest path computation, and traffic flow assignment. Combined with seismic fragility models and recovery functions for regional bridges, the framework captures the dynamic evolution of HTN functionality following seismic events. A multi-dimensional resilience evaluation system is also established to quantify network performance from emergency response and recovery perspectives. A case study on the Sioux Falls network under probabilistic earthquake scenarios demonstrates the effectiveness of the proposed method, achieving 95% prediction accuracy while reducing computational time by 90% compared to traditional numerical simulations. The results highlight the framework’s potential as a scalable, efficient, and reliable tool for large-scale post-disaster transportation system analysis. Full article

Most public seismic recordings, sampled at hundreds of Hz, tend to be unlabelled, i.e., not catalogued, mainly because of the sheer volume of samples and the amount of time needed by experts to confidently label detected events. This is especially challenging for very low signal-to-noise ratio microseismic events that characterise landslides during rock and soil mass displacement. Whilst numerous supervised machine learning models have been proposed to classify landslide events, they rely on a large amount of labelled datasets. Therefore, there is an urgent need to develop tools to effectively automate the data-labelling process from a small set of labelled samples. In this paper, we propose a semi-supervised method for labelling of signals recorded by seismometers that can reduce the time and expertise needed to create fully annotated datasets. The proposed Siamese network approach learns best class-exemplar anchors, leveraging learned similarity between these anchor embeddings and unlabelled signals. Classification is performed via soft-labelling and thresholding instead of hard class boundaries. Furthermore, network output explainability is used to explain misclassifications and we demonstrate the effect of anchors on performance, via ablation studies. The proposed approach classifies four landslide classes, namely earthquakes, micro-quakes, rockfall and anthropogenic noise, demonstrating good agreement with manually detected events while requiring few training data to be effective, hence reducing the time needed for labelling and updating models. Full article

The advanced determination of the type (foreshock–aftershock–swarm) of an ongoing seismic cluster is quite challenging; only retrospective solutions have thus far been proposed. In the period of January–March 2025, a seismic cluster, recorded between Santorini volcano and Amorgos Island, South Aegean Sea, caused considerable social concern. A rapid increase in both the seismicity rate and the earthquake magnitudes was noted until the mainshock of M_L = 5.3 on 10 February; afterwards, activity gradually diminished. Fault-plane solutions indicated SW-NE normal faulting. The epicenters moved with a mean velocity of ~0.72 km/day from SW to NE up to the mainshock area at a distance of ~25 km. Crucial questions publicly emerged during the cluster. Was it a foreshock–aftershock activity or a swarm of possibly volcanic origin? We performed real-time discrimination of the cluster type based on a daily re-evaluation of the space–time–magnitude changes and their significance relative to background seismicity using earthquake statistics and the topological metric betweenness centrality. Our findings were periodically documented during the ongoing cluster starting from the fourth cluster day (2 February 2025), at which point we determined that it was a foreshock and not a case of seismic swarm. The third day after the M_L = 5.3 mainshock, a typical aftershock decay was detected. The observed foreshock properties favored a cascade mechanism, likely facilitated by non-volcanic material softening and the likely subdiffusion processes in a dense fault network. This mechanism was possibly combined with an aseismic nucleation process if transient geodetic deformation was present. No significant aftershock expansion towards the NE was noted, possibly due to the presence of a geometrical fault barrier east of the Anydros Ridge. The 2025 activity offered an excellent opportunity to investigate deciphering the type of ongoing seismicity cluster for real-time discrimination between foreshocks, aftershocks, and swarms. Full article

Local site conditions can significantly influence the amplitude, duration, and frequency content of seismic recordings, making the characterization of subsoil properties a critical component in seismic hazard assessment. However, despite extensive research, standardized methodologies for assessing site effects are still lacking. This study presents preliminary steps in the site characterization of a small area of Campi Flegrei caldera (Italy), with the aim of enhancing understanding of local lithology and seismic wave propagation. The analysis focuses on the broad-band seismic station CSTH, installed in 2021, and incorporates data from a temporary 2D array of five short-period sensors deployed around the station. These sensors recorded both ambient noise and seismic events associated with caldera dynamics. To improve the robustness of the characterization, data from two additional permanent broad-band stations (CPIS and CSOB) of the Istituto Nazionale di Geofisica e Vulcanologia—Osservatorio Vesuviano’s monitoring network, also located nearby a hydrothermal field, were included. Spectral analyses such as Power Spectral Density (PSD), Horizontal-to-Vertical (H/V) spectral ratios, and f-k array technique were performed to evaluate the frequency-dependent response of the site and to support the development of a comprehensive seismic site model. Full article

Environmental noise is inevitable during seismic data acquisition, with major sources including heavy machinery, rivers, wind, and other environmental factors. During field data acquisition, it is important to assess the impact of environmental noise and evaluate data quality. In subsequent seismic data processing, these noise components also need to be eliminated. Accurate identification of noise traces facilitates rapid quality control (QC) during fieldwork and provides a reliable basis for targeted noise attenuation. Conventional environmental noise identification primarily relies on amplitude differences. However, in seismic data, high-amplitude signals are not necessarily caused by environmental noise. For example, surface waves or traces near the shot point may also exhibit high amplitudes. Therefore, relying solely on amplitude-based criteria has certain limitations. To improve noise identification accuracy, we use the Mel-spectrogram to extract features from seismic data and construct the dataset. Compared to raw time-series signals, the Mel-spectrogram more clearly reveals energy variations and frequency differences, helping to identify noise traces more accurately. We then employ a Vision Transformer (ViT) network to train a model for identifying noise in seismic data. Tests on synthetic and field data show that the proposed method performs well in identifying noise. Moreover, a denoising case based on synthetic data further confirms its general applicability, making it a promising tool in seismic data QC and processing workflows. Full article

China is one of the countries severely affected by earthquakes, making precise and timely identification of earthquake precursors essential for reducing casualties and property damage. A novel method is proposed that combines a rock acoustic emission (AE) detection technique with deep learning methods to facilitate real-time monitoring and advance earthquake precursor detection. The AE equipment and seismometers were installed in a granite tunnel 150 m deep in the mountains of eastern Guangdong, China, allowing for the collection of experimental data on the correlation between rock AE and seismic activity. The deep learning model uses features from rock AE time series, including AE events, rate, frequency, and amplitude, as inputs, and estimates the likelihood of seismic events as the output. Precursor features are extracted to create the AE and seismic dataset, and three deep learning models are trained using neural networks, with validation and testing. The results show that after 1000 training cycles, the deep learning model achieves an accuracy of 98.7% on the validation set. On the test set, it reaches a recognition accuracy of 97.6%, with a recall rate of 99.6% and an F₁ score of 0.975. Additionally, it successfully identified the two biggest seismic events during the monitoring period, confirming its effectiveness in practical applications. Compared to traditional analysis methods, the deep learning model can automatically process and analyse recorded massive AE data, enabling real-time monitoring of seismic events and timely earthquake warning in the future. This study serves as a valuable reference for earthquake disaster prevention and intelligent early warning. Full article

The accurate identification of seismic faults, which serve as crucial fluid migration pathways in hydrocarbon reservoirs, is of paramount importance for reservoir characterization. Traditional interpretation is inefficient. It also struggles with complex geometries, failing to meet the current exploration demands. Deep learning boosts fault identification significantly but struggles with edge accuracy and noise robustness. To overcome these limitations, this research introduces SwiftSeis-AWNet, a novel lightweight and high-precision network. The network is based on an optimized MedNeXt architecture for better fault edge detection. To address the noise from simple feature fusion, a Semantics and Detail Infusion (SDI) module is integrated. Since the Hadamard product in SDI can cause information loss, we engineer an Attention-Weighted Semantics and Detail Infusion (AWSDI) module that uses dynamic multi-scale feature fusion to preserve details. Validation on field seismic datasets from the Netherlands F3 and New Zealand Kerry blocks shows that SwiftSeis-AWNet mitigates challenges like the loss of small-scale fault features and misidentification of fault intersection zones, enhancing the accuracy and geological reliability of automated fault identification. Full article

Research on reliability control for enhancing power systems under random loads holds significant and undeniable importance in maintaining system stability, performance, and safety. The primary challenge lies in determining the reliability index while optimizing system parameters. To effectively address this challenge, we developed a novel intelligent algorithm and conducted an optimal reliability assessment for a Negative Stiffness Device (NSD) seismic isolation structure incorporating fractional-order damping. This algorithm combines the Gaussian Radial Basis Function Neural Network (GRBFNN) with the Particle Swarm Optimization (PSO) algorithm. It takes the reliability function with unknown parameters as the objective function, while using the Backward Kolmogorov (BK) equation, which governs the reliability function and is accompanied by boundary and initial conditions, as the constraint condition. During the operation of this algorithm, the neural network is employed to solve the BK equation, thereby deriving the fitness function in each iteration of the PSO algorithm. Then the PSO algorithm is utilized to obtain the optimal parameters. The unique advantage of this algorithm is its ability to simultaneously achieve the optimization of implicit objectives and the solution of time-dependent BK equations.To evaluate the performance of the proposed algorithm, this study compared it with the algorithm combines the GRBFNN with Genetic Algorithm (GA-GRBFNN)across multiple dimensions, including performance and operational efficiency. The effectiveness of the proposed algorithm has been validated through numerical comparisons and Monte Carlo simulations. The control strategy presented in this paper provides a solid theoretical foundation for improving the reliability performance of mechanical engineering systems and demonstrates significant potential for practical applications. Full article

The process of geothermal energy development may cause induced seismic activities, posing a potential threat to the sustainable utilization and safety of geothermal energy. To effectively evaluate the danger of induced seismic activities, this paper establishes an artificial neural network model and selects nine influencing factors as the input parameters of the neurons. Based on the results of induced seismic activity under different parameter conditions, a sensitivity analysis is conducted for each parameter, and the influence degree of each parameter on the magnitude of induced seismic activity is ranked from largest to smallest as follows: in situ stress state, fault presence or absence, depth, degree of fracture aggregation, maximum in situ stress, distance to fault, injection volume, fracture dip angle, angle between fracture, and fault. Then, the weights of each parameter in the model are modified to improve the accuracy of the model. Finally, through data collection and the literature review, the Pohang EGS project in South Korea is analyzed, and the induced seismic activity influencing factors of the Pohang EGS site are analyzed and evaluated using the induced seismic activity evaluation model. The results show that the induced seismicity are all located below 3.7 km (drilling depth). As the depth increases, the seismicity magnitude also shows a gradually increasing trend. An increase in injection volume and a shortening of the distance from faults will also lead to an increase in the seismicity magnitude. When the injection volume approaches 10,000 cubic meters, the intensity of the seismic activity sharply increases, and the maximum magnitude reaches 5.34, which is consistent with the actual situation. This model can be used for the induced seismic evaluation of future EGS projects and provide a reference for project site selection and induced seismic risk warning. Full article

This study proposes a Swin-ReshoUnet architecture with a three-level enhancement mechanism to address inefficiencies in multi-scale feature extraction and gradient degradation in deep networks for high-precision seismic exploration. The encoder uses a hierarchical convolution module to build a multi-scale feature pyramid, enhancing cross-scale geological signal representation. The decoder replaces traditional self-attention with ORCA attention to enable global context modeling with lower computational cost. Skip connections integrate a residual channel attention module, mitigating gradient degradation via dual-pooling feature fusion and activation optimization, forming a full-link optimization from low-level feature enhancement to high-level semantic integration. Simulated and real dataset experiments show that at decimation ratios of 0.1–0.5, the method significantly outperforms SwinUnet, TransUnet, etc., in reconstruction performance. Residual signals and F-K spectra verify high-fidelity reconstruction. Despite increased difficulty with higher sparsity, it maintains optimal performance with notable margins, demonstrating strong robustness. The proposed hierarchical feature enhancement and cross-scale attention strategies offer an efficient seismic profile signal reconstruction solution and show generality for migration to complex visual tasks, advancing geophysics-computer vision interdisciplinary innovation. Full article

Radon anomalies have long been explored as potential geochemical precursors to seismic activity due to their responsiveness to subsurface stress variations. However, before this study, the scientific progression of this research domain had not been systematically examined through a quantitative lens. This study presents a comprehensive bibliometric analysis of 379 articles published between 1977 and 2025 and indexed in Scopus and Web of Science. Utilizing the Bibliometrix R-package and its Biblioshiny interface, the analysis investigates temporal publication trends, leading countries, institutions, international collaboration networks, and thematic evolution. The results reveal a marked increase in research output since 2010, with China, India, and Italy emerging as the most prolific contributors. Thematic mapping indicates a shift from conventional geochemical monitoring toward the integration of artificial intelligence techniques, such as decision trees and neural networks, for anomaly detection and predictive modeling. Notwithstanding this methodological evolution, core research themes remain centered on radon concentration monitoring and the analysis of environmental parameters. Overall, the findings highlight the coexistence of traditional and emerging approaches, emphasizing the importance of standardized methodologies and interdisciplinary collaboration. This bibliometric synthesis provides strategic insights to inform future research and strengthen the role of radon monitoring in seismic early warning systems. Full article

To achieve more precise and effective microseismic magnitude estimation, a classification model based on transfer learning with an improved deep residual network is proposed for predicting microseismic magnitudes. Initially, microseismic waveform images are preprocessed through cropping and blurring before being used as inputs to the model. Subsequently, the microseismic waveform image dataset is divided into training, testing, and validation sets. By leveraging the pretrained ResNet18 model weights from ImageNet, a transfer learning strategy is implemented, involving the retraining of all layers from scratch. Following this, the CBAM is introduced for model optimization, resulting in a new network model. Finally, this model is utilized in seismic magnitude classification research to enable microseismic magnitude prediction. The model is validated and compared with other commonly used neural network models. The experiment uses microseismic waveform data and images of magnitudes 0–3 from the Stanford Earthquake Dataset (STEAD) as training samples. The results indicate that the model achieves an accuracy of 87% within an error range of ±0.2 and 94.7% within an error range of ±0.3. This model demonstrates enhanced stability and reliability, effectively addressing the issue of missing data labels. It validates that using ResNet transfer learning combined with an attention mechanism yields higher accuracy in microseismic magnitude prediction, as well as confirming the effectiveness of the CBAM. Full article

Accurate and rapid prediction of structural responses to seismic loading is critical for ensuring structural safety. Recently, there has been active research focusing on the application of deep learning techniques, including Physics-Informed Neural Networks (PINNs) and Long Short-Term Memory (LSTM) networks, to predict the dynamic behavior of structures. While these methods have shown promise, each comes with distinct limitations. PINNs offer physical consistency but struggle with capturing long-term temporal dependencies in nonlinear systems, while LSTMs excel in learning sequential data but lack physical interpretability. To address these complementary limitations, this study proposes a hybrid LSTM-PINN model, combining the temporal learning ability of LSTMs with the physics-based constraints of PINNs. This hybrid approach allows the model to capture both nonlinear, time-dependent behaviors and maintain physical consistency. The proposed model is evaluated on both single-degree-of-freedom (SDOF) and multi-degree-of-freedom (MDOF) structural systems subjected to the El-Centro ground motion. For validation, the 1940 El-Centro NS earthquake record was used, and the ground acceleration data were normalized and discretized for numerical simulation. The proposed LSTM-PINN is trained under the same conditions as the conventional PINN models (e.g., same optimizer, learning rate, and loss structure), but with fewer training epochs, to evaluate learning efficiency. Prediction accuracy is quantitatively assessed using mean error and mean squared error (MSE) for displacement, velocity, and acceleration, and results are compared with PINN-only models (PINN-1, PINN-2). The results show that LSTM-PINN consistently achieves the most stable and precise predictions across the entire time domain. Notably, it outperforms the baseline PINNs even with fewer training epochs. Specifically, it achieved up to 50% lower MSE with only 10,000 epochs, compared to the PINN’s 50,000 epochs, demonstrating improved generalization through temporal sequence learning. This study empirically validates the potential of physics-guided time-series AI models for dynamic structural response prediction. The proposed approach is expected to contribute to future applications such as real-time response estimation, structural health monitoring, and seismic performance evaluation. Full article

The Moxi area in the Sichuan Basin hosts abundant deep geothermal resources, but their thermal regime and accumulation mechanisms remain poorly understood. Using 2D/3D seismic data, drilling records, and temperature measurements (DST), we analyze deep thermal fields, reservoir–caprock systems, and structural features. The following are our key findings: (1) Heat transfer is conduction-dominated, with thermal anomalies in Late Permian–Early Cambrian strata. Four mudstone/shale caprocks and three carbonate reservoirs occur, with the Longtan Formation as the key seal. Reservoir geothermal gradients (25.05–32.55 °C/km) exceed basin averages. (2) Transtensional strike-slip faults form E-W/NE/NW networks; most terminate at the Permian Longtan Formation, with few extending into the Lower Triassic while penetrating the Archean–Lower Proterozoic basement. (3) Structural highs positively correlate with higher geothermal gradients. (4) The deep geothermal reservoirs and thermal accumulation mechanisms in the Moxi area are jointly controlled by crustal thinning, basement uplift, and structural architecture. Mantle-derived heat converges at basement uplift cores, generating localized thermal anomalies. Fault networks connect these deep heat sources, facilitating upward fluid migration. Thick Longtan Formation shale seals these rising thermal fluids, causing anomalous heating in underlying strata and concentrated thermal accumulation in reservoirs—enhanced by thermal focusing effects from uplift structures. This study establishes a theoretical framework for target selection and industrial-scale geothermal exploitation in sedimentary basins, highlighting the potential for repurposing oil/gas infrastructure. Full article

Physics-informed neural networks (PINNs) have emerged as a transformative methodology integrating deep learning with scientific computing. This review establishes a three-dimensional analytical framework to systematically decode PINNs’ development through methodological innovation, theoretical breakthroughs, and cross-disciplinary convergence. The contributions include threefold: First, identifying the co-evolutionary path of algorithmic architectures from adaptive optimization (neural tangent kernel-guided weighting achieving 230% convergence acceleration in Navier-Stokes solutions) to hybrid numerical-deep learning integration (5× speedup via domain decomposition) and second, constructing bidirectional theory-application mappings where convergence analysis (operator approximation theory) and generalization guarantees (Bayesian-physical hybrid frameworks) directly inform engineering implementations, as validated by $72\%$ cost reduction compared to FEM in high-dimensional spaces ($p<0.01,n=15$ benchmarks). Third, pioneering cross-domain knowledge transfer through application-specific architectures: TFE-PINN for turbulent flows ($5.12\pm 0.87\%$ error in NASA hypersonic tests), ReconPINN for medical imaging ($▵\mathrm{SSIM}=+0.18\pm 0.04$ on multi-institutional MRI), and SeisPINN for seismic systems ($0.52\pm 0.18$ km localization accuracy). We further present a technological roadmap highlighting three critical directions for PINN 2.0: neuro-symbolic, federated physics learning, and quantum-accelerated optimization. This work provides methodological guidelines and theoretical foundations for next-generation scientific machine learning systems. Full article