Search Results (41)

Underground coal mine safety monitoring relies on multi-source sensor networks, but abnormal-state detection remains challenging because methane, airflow, dust, and equipment-operation signals are non-stationary, heterogeneous, and constrained by ventilation and mining disturbances. This study proposes GasNet, an interpretable engineering-prior graph–temporal–spectral framework for precursor-related anomaly detection. Variables are organized into methane-related core sensors and environmental–operational modulation sensors. A directed sensor graph is constructed using ventilation causality, sensor deployment, and shearer-coupling relationships. GasNet then integrates a graph convolutional network for spatial–topological modeling, TimesNet for temporal–spectral pattern extraction, and cross-attention for adaptive feature fusion. An unsupervised reconstruction strategy identifies intervals deviating from learned normal production patterns. Field validation was conducted on the 31002 fully mechanized working face of the Xinyuan Coal Mine, where eight precursor-related abnormal intervals were annotated from monitoring data and field records. GasNet achieved a Precision of 0.881, a Recall of 1.000, an F1-score of 0.937, a false-alarm rate of 0.0017, and zero missed detections, with the highest F1-score among seven time-series baselines. Interpretability analysis further provided feature-fusion and sensor-time evidence for warning review. These results support the feasibility of GasNet for interpretable anomaly detection in the investigated working face. Full article

Accurate mine pressure prediction is essential for understanding roof–support interaction and supporting intelligent monitoring in fully mechanized longwall mining. In underground production, hydraulic-support pressure sensors provide continuous pressure sequences that reflect the mechanical response of the support–roof system. However, these sequences are affected by local mining disturbances, missing records, abnormal zero-value segments, nonstationary fluctuations, and periodic weighting, which make future pressure forecasting challenging. To address this issue, an LSTM–Transformer hybrid model is proposed for hydraulic-support pressure forecasting. The LSTM module extracts local nonlinear pressure-evolution features from recent historical windows, whereas the Transformer module captures temporal dependencies and periodic pressure patterns through global sequence modeling. Support-wise experiments were conducted using field monitoring data from Yili No. 1 Mine, and the pressure sequence of each support was processed independently to avoid mixing information from different support locations. In the representative test case, the proposed model achieved an R² of 0.971 and reduced the MAE to 0.471 MPa, while improving the phase consistency of predicted pressure peaks. Further analysis indicates that sufficient historical data coverage is necessary to capture complete pressure-evolution cycles, and that the 25-step forecasting case maintains stable accuracy for short-term mine pressure estimation. These findings demonstrate the feasibility of the proposed approach for hydraulic-support pressure prediction under the monitored conditions of the studied working face. Full article

Methane exceedance events in underground coal mines are often associated with progressive multisensor abnormal evolution processes involving operational, ventilation, environmental, and methane-drainage subsystems. Existing studies primarily focus on methane concentration prediction and provide limited interpretability regarding how abnormal evolution patterns emerge before threshold exceedance. To address this limitation, this study proposes an interpretable event-driven multisensor risk-evolution analysis framework for methane early warning. Methane exceedance events are first extracted from multisensor monitoring data, and a continuous multisensor risk representation together with a persistence-based trigger mechanism is developed to identify sustained abnormal evolution prior to methane exceedance. Event-specific temporal dependency networks are then constructed using lagged dependency analysis to characterize multisensor interaction structures within event windows. Representative evolution paths and recurrent critical variables are further identified to reveal interpretable abnormal evolution patterns. Experiments conducted on a real underground coal mine monitoring dataset containing 784 methane exceedance events demonstrate that the proposed framework achieved the highest early-warning performance among all compared baselines. Under the Any-Target-Sensor criterion, the framework attained an effective warning rate of 0.848 and significantly outperformed benchmark methods in event-level McNemar tests (p < 0.001). The results further indicate that methane exceedance events are generally associated with structured multisensor abnormal evolution processes rather than isolated methane fluctuations, providing an interpretable system-level perspective for methane early warning. Full article

Shallow-buried coal seams in western China are commonly overlain by deeply incised gully terrain, where mining is often accompanied by coal-wall spalling and abnormal increases in support resistance, which affect safe and efficient production. To investigate overburden failure during shallow-buried coal seam mining under gully terrain and to clarify the support–resistance mechanism, a typical working face was selected as the engineering background. Physical similarity simulation, 3DEC numerical simulation, and theoretical analysis were used to analyze overburden failure characteristics and the coupled evolution of the stress, displacement, and fracture fields. Mechanical models of key-stratum fracture and a support–resistance estimation model were established to reveal the influence of overburden-thickness variation on key-stratum fracture and support resistance. The results show that overburden failure in gully areas exhibits pronounced stage-dependent and asymmetric characteristics. In the similarity simulation, the initial fracture intervals of the key stratum in the downhill section were 32 m and 36 m, indicating an asymmetric fracture pattern with a shorter span on the left side and a longer span on the right side. In the uphill section, the periodic fracture interval of the key stratum decreased from 30 m to 24 m as the overburden thickness increased. During overburden failure in gully areas, the three fields exhibited a coupled relationship: stress concentration at the working face caused overburden failure and subsidence, which promoted fracture propagation, whereas stress redistribution in the goaf compacted the fractured overburden and promoted fracture closure. The overburden failure characteristics differed significantly between mining stages. During downhill mining, the key stratum behaved as a fixed-ended beam with a relatively large fracture interval, whereas during uphill mining, it formed a cantilever beam, and its fracture interval decreased with increasing overburden thickness. The loading mechanism of support resistance was shown to be jointly controlled by variations in gully overburden thickness and key-stratum fracture. During downhill mining, support loading increased gradually under the support of the fixed-ended beam key stratum. During uphill mining, support loading exhibited periodic abrupt increases under the combined effects of increasing overburden thickness and periodic fracture of the cantilever-beam key stratum. These findings provide a theoretical basis for strata pressure control at working faces in gully areas. Full article

Landslide event warning in mining areas is essential for geohazard risk mitigation and infrastructure safety. With the increasing use of Global Navigation Satellite System (GNSS) monitoring networks, warning decisions are often derived from abnormal deformation responses in continuous displacement records. However, deriving stable and transferable warning decisions from GNSS networks is challenged by spatially coupled station responses, time-varying displacement patterns, and incomplete or disturbed observations. To address these issues, this study proposes a graph-augmented gradient boosting decision tree framework, termed GA-GBDT (Graph-Augmented Gradient Boosting Decision Trees), for multi-station landslide event warning in mining areas. The framework first constructs a weighted station graph to encode spatial dependence across stations. Based on this graph, a Gated Recurrent Unit (GRU) and a Graph Convolutional Network (GCN) are integrated to learn spatio-temporal embeddings, which are then fused with station-wise features and fed into XGBoost (eXtreme Gradient Boosting) for warning decision-making. Experiments on a 90-station GNSS network show that GA-GBDT outperforms representative rule-based, machine-learning, and deep-learning baselines, achieving more robust warning performance with improved generalization and false-alarm control. These results indicate that GA-GBDT improves warning robustness, decision stability, and cross-zone generalization for GNSS-based landslide warning in mining areas, with potential transferability to other slope warning scenarios. Full article

Continuous casting is a key core link in steel production with characteristics of strong nonlinearity, multi-parameter coupling and dynamic fluctuations under working conditions. Traditional experience-dependent or mechanism-driven models are no longer suitable for the high-quality and high-efficiency production demands of modern steel industries. Machine learning provides an effective technical path for solving the complex control problems in the continuous casting process through its powerful data mining and pattern recognition capabilities. This paper systematically reviews the research progress of machine learning applications in the field of continuous casting, focusing on three core scenarios: abnormal prediction, quality defect detection and process parameter optimization. It sorts out the evolution from single models to feature optimization and integration, deep learning hybrid models, and mechanism-data dual-driven models. It summarizes the significant achievements of this technology in reducing production risks and improving the stability of cast billet quality, and it analyzes the prominent challenges currently faced such as data distortion and distribution imbalance, insufficient model interpretability and limited cross-scenario generalization ability. Finally, it looks forward to future technological innovation and application expansion directions, providing theoretical support and technical references for the digital and intelligent transformation of the steel industry. Full article

Identifying abnormal group behavior formed by multi-type participants from large-scale historical industry and tax data is important for regulators to prevent potential criminal activity. We propose an Abnormal Alliance detection framework comprising two methods. For detecting joint behavior among multi-type participants, we present DyCIAComDet, a dynamic community identification and tracking method for large-scale, time-varying bipartite multi-type participant networks, and introduce three community-splitting measurement indicators—cohesion, integration, and leadership—to improve community division. To verify whether joint behavior is abnormal, termed an Abnormal Alliance, we propose BMPS, a frequent-sequence identification algorithm that mines key features along community evolution paths based on bitmap matrices, sequence matrices, prefix-projection matrices, and repeated-projection matrices. The framework is designed to address sampling limitations, temporal issues, and subjectivity that hinder traditional analyses and to remain scalable to large datasets. Experiments on the Southern Women benchmark and a real tax dataset show DyCIAComDet yields a mean modularity Q improvement of 24.6% over traditional community detection algorithms. Compared with PrefixSpan, BMPS improves mean time and space efficiency by up to 34.8% and 35.3%, respectively. Together, DyCIAComDet and BMPS constitute an effective, scalable detection pipeline for identifying abnormal alliances in tax datasets and supporting regulatory analysis. Full article

Bitcoin transaction anomaly detection is essential for maintaining financial market stability. A significant challenge is capturing the dynamically evolving transaction patterns within transaction networks. Dynamic graph models are effective for characterizing the temporal evolution of transaction systems. However, current methods struggle to mine long-range temporal dependencies and address the class imbalance caused by the scarcity of abnormal samples. To address these issues, we propose a novel approach, the Bidirectional EvolveGCN with Class-Balanced Learning Network (Balanced-BiEGCN), for Bitcoin transaction anomaly detection. This model integrates two key components: (1) a bidirectional temporal feature fusion mechanism (Bi-EvolveGCN) that enhances the capture of long-range temporal dependencies and (2) a Sample Class Transformation (CSCT) classifier that generates difficult-to-distinguish abnormal samples to balance the positive and negative class distribution. The generation of these samples is guided by two loss functions: the adjacency distance adaptive loss function and the symmetric space adjustment loss function, which optimize the spatial distribution and confusion of abnormal samples. Experimental results on the Elliptic dataset demonstrate that Balanced-BiEGCN outperforms existing baseline methods in anomaly detection. Full article

Anomaly detection aims to identify abnormal instances that significantly deviate from normal samples. With the natural connectivity between instances in the real world, graph neural networks have become increasingly important in solving anomaly detection problems. However, existing research mainly focuses on static graphs, while there is less research on mining anomaly patterns in dynamic graphs, which has important application value. This paper proposes a Transformer-based semi-supervised anomaly detection framework for dynamic graphs. The framework adopts the Transformer architecture as the core encoder, which can effectively capture long-range dependencies and complex temporal patterns between nodes in dynamic graphs. By introducing time-aware attention mechanisms, the model can adaptively focus on important information at different time steps, thereby better understanding the evolution process of graph structures. The multi-head attention mechanism of Transformer enables the model to simultaneously learn structural and temporal features of nodes, while positional encoding helps the model understand periodic patterns in time series. Comprehensive experiments on three real datasets show that TSAD significantly outperforms existing methods in anomaly detection accuracy, particularly demonstrating excellent performance in label-scarce scenarios. Full article

Electricity theft in power grids results in significant economic losses for utility companies. While machine learning (ML) methods have shown promising results in detecting such frauds, they often suffer from low detection rates, leading to excessive physical inspections. In this study, we attempted to solve the above-mentioned problem using a novel approach. The proposed framework utilizes the intelligence of Siamese network architecture with the Triplet Loss function to detect electricity theft using a labeled dataset obtained from Multan Electric Power Company (MEPCO), Pakistan. The proposed method involves analyzing and comparing the consumption patterns of honest and fraudulent consumers, enabling the model to distinguish between the two categories with enhanced accuracy and detection rates. We incorporate advanced feature extraction techniques and data mining methods to transform raw consumption data into informative features, such as time-based consumption profiles and anomalous load behaviors, which are crucial for detecting abnormal patterns in electricity consumption. The refined dataset is then used to train the Siamese network, where the Triplet Loss function optimizes the model by maximizing the distance between dissimilar (fraudulent and honest) consumption patterns while minimizing the distance among similar ones. The results demonstrate that our proposed solution outperforms traditional methods by significantly improving accuracy (95.4%) and precision (92%). Eventually, the integration of feature extraction with Siamese networks and Triplet Loss offers a scalable and robust framework for enhancing the security and operational efficiency of power grids. Full article

Liquid rocket engines occasionally experience abnormal phenomena with unclear mechanisms, causing difficulty in design improvements. To address the above issue, a data mining method that combines ante hoc explainability, post hoc explainability, and prediction accuracy is proposed. For ante hoc explainability, a feature selection method driven by data, models, and domain knowledge is established. Global sensitivity analysis of a physical model combined with expert knowledge and data correlation is utilized to establish the correlations between different types of parameters. Then a two-stage optimization approach is proposed to obtain the best feature subset and train the prediction model. For the post hoc explainability, the partial dependence plot (PDP) and SHapley Additive exPlanations (SHAP) analysis are used to discover complex patterns between input features and the dependent variable. The effectiveness of the hybrid feature selection method and its applicability under different noise combinations are validated using synthesized data from a high-fidelity simulation model of a pressurization system. Then the analysis of the causes of a large vibration phenomenon in an active engine shows that the prediction model has good accuracy, and the feature selection results have a clear mechanism and align with domain knowledge, providing both accuracy and interpretability. The proposed method shows significant potential for data mining in complex aerospace products. Full article

Takotsubo syndrome (TS), also known as stress-induced cardiomyopathy, is classically characterized by an acute onset mimicking myocardial infarction and by distinctive transient wall motion abnormalities detectable via echocardiography, often resembling a Japanese octopus trap (the so-called “takotsubo”). The possibility that a genetic background may contribute to TS susceptibility emerged early, supported by several familial case reports. Despite a large number of investigations, no definitive genetic markers associated with TS risk have been conclusively identified. The lack of a clear Mendelian inheritance pattern suggests a multifactorial etiology and pathogenesis, likely involving complex gene–environment interactions and a polygenic background. This review analyzes the genetic variants implicated in the different functional pathways contributing to TS pathogenesis and discusses the current state of knowledge regarding its genetic underpinnings. Finally, we propose future directions for research aimed at identifying a multigene susceptibility panel that could be useful in diagnosis, prevention strategies, and the identification of novel therapeutic targets for individuals at high risk. We conclude that innovative approaches based on data-mining algorithms and nonlinear analytic methods applied to large patient datasets may be instrumental in resolving the genetic complexity of TS. Full article

In the present study, major oxide, trace, and rare-earth element (REE) contents in the stream sediments of the Ikuno and surrounding areas of the central part of Hyogo Prefecture in the Kinki district in southwestern Japan were analyzed. Several abandoned mines that contain Au, Ag, Cu, Pb, Zn, Fe, W, and As exist in these areas, including the Ikuno and Akenobe mines, which are famous historical mines. A total of 156 stream sediments over approximately 1300 km² in these areas were analyzed using X-ray fluorescence (XRF) and inductively coupled plasma mass spectrometry (ICP-MS). The spatial distribution patterns of elemental concentrations in the stream sediments in the Ikuno area were determined by three primary factors: the surface geology, the localized deposition of ore minerals, and the influence of the sedimentation of heavy minerals in the basin on local distribution. The mean value of the spatial distributions of the ore deposits was greater than the median, primarily due to the presence of concentrated regions near the mining sites. A Kolmogorov–Smirnov test indicated abnormal distribution patterns of Pb, Zn, Cu, Cr, and Ni due to the presence of exceptionally high concentrations of these elements at the mine sites. The stream sediments showed higher levels of light REEs, mainly La, Ce, and Nd, in comparison with the heavy REEs. This pattern, deviating from the global abundance, suggests the dominating influence of mining sites on local REE distributions. These findings are essential for assessing the environmental impacts of historical mining and developing strategies for responsible resource management in the region. By understanding the geochemical signatures of mining-affected areas, these data could contribute to future environmental monitoring and mitigation efforts, enhancing our understanding of environmental sustainability and responsible resource utilization. Full article

This study addresses the abnormal emission of pressure-relieved methane under high-intensity mining conditions by integrating geostatistical inversion, FLAC^3D-COMSOL coupled numerical simulations, and stable carbon–hydrogen isotopic tracing. Focusing on the 12023 working face at Wangxingzhuang Coal Mine, we established a heterogeneous methane reservoir model to analyze the mechanical responses of surrounding rock, permeability evolution, and gas migration patterns under mining intensities of 2–6 m/d. Key findings include the following: (1) When the working face advanced 180 m, vertical stress in concentration zones increased significantly with mining intensity, peaking at 12.89% higher under 6 m/d compared to 2 m/d. (2) Higher mining intensities exacerbated plastic failure in floor strata, with a maximum depth of 47.9 m at 6 m/d, enhancing permeability to 223 times the original coal seam. (3) Isotopic fingerprinting and multi-method validation identified adjacent seams as the dominant gas source, contributing 77.88% of total emissions. (4) Implementing targeted long directional drainage boreholes in floor strata achieved pressure-relief gas extraction efficiencies of 34.80–40.95%, reducing ventilation air methane by ≥61.79% and maintaining return airflow methane concentration below 0.45%. This research provides theoretical and technical foundations for adaptive gas control in rapidly advancing faces through stress–permeability coupling optimization, enabling the efficient interception and resource utilization of pressure-relieved methane. The outcomes support safe, sustainable coal mining practices and advance China’s Carbon Peak and Neutrality goals. Full article

Maritime behavior detection is vital for maritime surveillance and management, ensuring safe ship navigation, normal port operations, marine environmental protection, and the prevention of illegal activities on water. Current methods for detecting anomalous vessel behaviors primarily rely on single time series data or feature point analysis, which struggle to capture the relationships between vessel behaviors, limiting anomaly identification accuracy. To address this challenge, we proposed a novel vessel anomaly detection framework, which is called the BPEF-TSD framework. It integrates a ship behavior pattern recognition algorithm, Smith–Waterman, and text similarity measurement methods. Specifically, we first introduced the BPEF mining framework to extract vessel behavior events from AIS data, then generated complete vessel behavior sequence chains through temporal combinations. Simultaneously, we employed the Smith–Waterman algorithm to achieve local alignment between the test vessel and known anomalous vessel behavior sequences. Finally, we evaluated the overall similarity between behavior chains based on the text similarity measure strategy, with vessels exceeding a predefined threshold being flagged as anomalous. The results demonstrate that the BPEF-TSD framework achieves over 90% accuracy in detecting abnormal trajectories in the waters of Xiamen Port, outperforming alternative methods such as LSTM, iForest, and HDBSCAN. This study contributes valuable insights for enhancing maritime safety and advancing intelligent supervision while introducing a novel research perspective on detecting anomalous vessel behavior through maritime big data mining. Full article