Aug 2026· Applied Sciences· 0 citations· 21 references
Abstract
Hydrogen refueling station (HRS) requires continuous safety monitoring, yet conventional management relies largely on periodic inspection and manual oversight, limiting proactive risk mitigation. This study presents a data-driven intelligent analysis platform for real-time monitoring and anomaly detection of HRS safety data, including pressure, temperature, and flow-rate measurements from compressors, storage tanks, and dispensers. The platform integrates data collection adapters, a time-series database, and machine learning-based diagnostic modules (regression, clustering, and classification) into a unified reference software framework. For anomaly detection, an unsupervised LSTM-Variational Autoencoder trained on normal operating data is combined with DBSCAN-based clustering and a Mann–Kendall trend test to jointly identify point anomalies and pattern-level drifts, addressing the scarcity of labeled abnormal data in HRS environments. A continual learning mechanism further adapts detection thresholds to gradual and abrupt pattern changes without full retraining. The system was deployed and validated at BAM’s demonstration hydrogen refueling station in Germany, integrated with a remote safety-monitoring system and confirmed through performance testing, demonstrating reliable, proactive hydrogen safety management.
Reliable early detection of degradation in gas turbines is essential for preventing failures and optimizing maintenance strategies. However, traditional fault-detection methods struggle to capture the complex temporal dynamics that precede abnormal conditions. This study investigates the use of Generative Adversarial Networks (GANs) for time-series modeling and anomaly datasets augmentation, aiming to learn the evolving behavior of turbine systems directly from sensor data and to address the insufficient amount of failure data for anomaly detection. In the first stage, GANs are trained on historical time-series recordings of turbines operating under normal conditions to address data scarcity and enhance model robustness. Validation results show that the generated sequences maintain strong statistical consistency with the original data, accurately reproducing characteristic temporal patterns of healthy operation. The second stage focuses on anomaly detection by integrating real-time sensor inputs with the learned temporal representations from both real and synthetic datasets. The discriminator detects deviations from expected behavior, enabling early identification of abnormal trends and providing insight into degradation trajectories. Experimental findings demonstrate that the proposed framework captures dynamic transitions more effectively than threshold-based methods, predicts the progression of faults, and supports the construction of an expanded anomaly catalog by generating synthetic profiles that resemble documented failure modes. Overall, the approach improves early-warning capabilities and strengthens data-driven maintenance planning.
Maria del Carmen Rubiales Mena, A. Roque, M. A. Sanz-Bobi et al.· IEEE Access· 0 citations
This study develops and evaluates an AI-based analytical system for detecting anomalies in industrial processes. The work reviews major sources of risk in industrial control systems, distinguishes point, contextual, and collective anomalies, and summarizes the principal machine-learning approaches used for industrial anomaly detection. A synthetic dataset modeled on the Secure Water Treatment (SWaT) testbed was created with 10 sensor and actuator variables and 10,000 one-second observations, including 1,000 anomalous samples. After missing-value interpolation, duplicate removal, low-variance filtering, and standardization for consistent analysis and visualization, an Isolation Forest with 200 trees was trained in a novelty-detection configuration using normal operating data. On the held-out test set, the model achieved 88.63% accuracy, 45.86% precision, 75.67% recall, and an F1-score of 57.11%. The results show that Isolation Forest can detect most simulated anomalies, although the relatively low precision indicates a substantial false-alarm burden. Future work should validate the approach on authorized real SWaT or PLC-SCADA data, investigate hybrid temporal models, and incorporate explainable-AI methods to support operator decision-making.
Mehdiyeva Almaz, Ahmedov Elmar, Uzakov Gulom et al.· 2026 International Conferenc...· 0 citations
Anomaly-detection methods typically fail to detect changes in operating conditions or are unresponsive in detecting changes in the aging of industrial assets. This paper outlines an unsupervised lightweight machine learning-based approach to adaptively detect anomalies in the reliability of industrial assets. It incorporates in its framework operating-range normalisation, principal component analysis, Euclidean-distance anomaly scoring, Gaussian-based dynamic thresholds, sensor-contribution ranking and autonomous baseline recalibration. The validation was performed by taking 10 sensor variables initially screened in a critical P-101 A/B pump in an HDPE slurry polymerisation process, sampled approximately every 15 min and nine variables were selected for sensor-quality screening and then used for the validation. The framework identified all 44 reported functional failures resulting in a 100% recall, 84.62% precision, an average warning lead time of 72 h and an F1-score of 91.67%. The findings show that a self-recalibrating unsupervised model can successfully detect faults in a scalable manner, with an interpretable model, low computational complexity and can adapt to ageing assets and varying operating baselines without labelled fault information.
Dhirendra Singh, Durga Prasad Panday, Manish Kumar· International journal of com...· 0 citations
The rapid expansion of renewable energy infrastructures has introduced significant challenges for monitoring system performance, ensuring regulatory compliance, and maintaining transparency in energy production and emissions reporting. Modern renewable energy systems generate large volumes of heterogeneous operational data through smart meters, sensor networks, supervisory control and data acquisition (SCADA) systems, and distributed generation platforms. Traditional monitoring approaches, which rely primarily on rule-based thresholds and periodic audits, often struggle to process such complex and dynamic data streams in real time. This paper proposes an artificial intelligence (AI)-driven intelligent monitoring framework designed to enhance operational oversight and sustainability monitoring in renewable energy systems. The proposed architecture integrates machine learning and anomaly detection techniques to analyze energy production data, detect abnormal operational patterns, and assess compliance with environmental and regulatory requirements. A design-science research methodology is adopted to develop and evaluate the framework using simulated renewable energy datasets representing solar and wind energy production scenarios. Experimental results demonstrate that the AI-based monitoring system significantly improves anomaly detection accuracy and reduces reporting delays compared with traditional rule-based monitoring methods. The proposed approach supports intelligent renewable energy infrastructure management by enabling proactive monitoring, improved operational transparency, and enhanced sustainability reporting.
Badreddine Said, Ashraf Rashid, Omari Asem et al.· E3S Web of Conferences· 1 citation
For the urban environment monitoring scenario, this paper constructs a multi-source time-series data feature system and preprocessing process. Through exponential smoothing and linear interpolation, noise is suppressed and missing data is repaired. On this basis, a joint anomaly identification framework combining traditional supervised models and LSTM time-series models is designed. Further, a multi-model output weighted fusion and adaptive weight update mechanism based on batch performance indicators is proposed. The fusion results are embedded into the monitoring business process and real-time data stream interface to achieve online anomaly determination. Experimental results show that in typical monitoring scenarios, the accuracy of anomaly identification in this system reaches 97.3%, which is approximately 2.5 and 1.7 percentage points higher than that of the two comparison systems (94.8% and 95.6%), respectively. The false alarm rate is reduced to 3.1%, and the end-to-end average latency is about 180 ms. This verifies the effectiveness of the proposed mechanism in balancing recognition accuracy and real-time performance, providing a feasible technical path for engineering environment anomaly monitoring.
Luming Zhang, Xian Zheng· International Conference on...· 0 citations
During the operation of power-system industrial control and monitoring platforms (e.g., SCADA/EMS and substation automation), strong coupling among components and multi-source heterogeneous data often lead to alarm flooding and complicate root cause identification. To address this, this paper proposes an intelligent abnormal-alarm identification and on-site operation behavior analysis method, combining artificial intelligence with an abductive inference framework. Under fixed parameters, the method first aggregates raw alarm streams by events to enhance structure and interpretability. Then, a diagonal-covariance Gaussian Hidden Markov Model (HMM) is trained with normal data, and a path-deviation metric ranks root cause candidates. Multi-source evidence chains—integrating temporal, network, and semantic features—further improve inference interpretability for grid operation and maintenance. Using annotated operation logs, four quantitative metrics (MTTA, MTTR, action rate, consistency) assess the link between model outputs and actual handling behaviors. Experiments on five test sets show the method achieves a 78% alarm compression rate and a 0.43 average silhouette coefficient. Top-1 and Top-3 root cause localization hit rates are 71.8% and 88.5%, with path score fluctuations under 0.05 nats. The average MTTA and MTTR are 186s and 792s, with an 84% action rate and 72% consistency. These results confirm the method’s effectiveness in mitigating alarm flooding, improving root cause localization, and supporting on-site decision-making in power-grid operational scenarios.
Kan Shi, Jiao Yang, Yongmei Chen et al.· EAI Endorsed Transactions on...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
