Jul 2026· 2026 International Conference on Intelligent and Sustainable AI Systems (ICOSAAS)· pp. 392-398· 0 citations· 17 references
Abstract
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.
This study presents a comparative evaluation of unsupervised anomaly detection techniques using the PyOD Python library, applied to three representative industrial scenarios: Intrusion Detection, Fault Detection, and Predictive Maintenance. Three public datasets, each corresponding to one of these domains, are used to assess the generalizability and robustness of the methods. The methodology includes data normalization, feature selection, and class balancing to address the specific challenges of each dataset, such as class imbalance and heterogeneous feature spaces. Performance is assessed using metrics such as Precision, Recall, Specificity, F1-score, AUC-PR, AUC-ROC, Balanced Accuracy, and Execution Time (training and testing). The findings offer valuable guidance for selecting robust anomaly detection models in industrial applications beyond cybersecurity, including those related to reliability and maintenance. The code and results will be publicly available on GitHub
https://github.com/hvelesaca/PyOD_OPC-UA
to support future research.
Henry O. Velesaca, Melissa Ayllon Gutierrez, Luis Barrera Muñoz et al.· Manufacturing Review· 0 citations
This work explores the application of pre-trained time-series foundation models (FMs) for detecting anomalies in industrial processes and introduces a new time-series forecasting method that filters out suspicious data and uses previously predicted data as input, called Forecast Fallback (FF).
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.
Minsu Kim, Seongseop Kim, Seungwoo Lee et al.· Applied Sciences· 0 citations
Monitoring chemical processes is essential to prevent catastrophic failures, optimize costs and profits, and ensure the safety of employees and the environment. A key component of modern monitoring systems is the automated detection of anomalies in sensor data over time, called time series, enabling partial automation of plant operation and adding additional layers of supervision to crucial components. The development of anomaly detection methods in this domain is challenging, since real chemical process data is usually proprietary, and simulated data is generally not a sufficient replacement. In this paper, we present NoBOOM, the first collection of datasets for anomaly detection in real-world chemical process data, including labeled data from a running process at our industry partner BASF SE — one of the world’s leading chemical companies —
Dennis Wagner, Fabian Hartung, J. Arweiler et al.· Neural Information Processin...· 2 citations
Machine-learning-based anomaly detection is increasingly used in industrial control systems (ICS), yet most studies assume that detector training data is trustworthy. In practice, training data may be corrupted through compromised logs, labeling errors, manipulated historian records, or unsafe retraining processes. This paper evaluates the robustness of offline ICS anomaly-detection pipelines on the Secure Water Treatment (SWaT) benchmark under training-time contamination. We assess 11 heterogeneous anomaly detectors under three contamination strategies: random injection, similarity-targeted injection, and feature-noise injection. The first two insert attack samples into the nominal training pool, while the third adds bounded Gaussian noise to selected normal training samples. These attacks are contamination-based rather than gradient-driven poisoning methods. Contamination budgets from 1% to 10% are evaluated using clean validation and test sets under a unified offline protocol. The results show that robustness is strongly model-dependent and cannot be predicted from clean-data performance alone. Injection-based contamination causes the greatest degradation, particularly for local-density and distance-based detectors, whereas feature-noise contamination has a comparatively limited effect. PCA, SVM, HBOS, and IForest remain relatively stable, while the tuned neural detectors demonstrate intermediate robustness. Overall, the findings highlight the importance of training-data integrity in ML-enabled ICS monitoring, subject to the evaluated dataset, models, and threat assumptions.
Mustafa Umut Ozbek, Taiwo P. Ojo, Pooria Madani et al.· 0 citations
Motors are crucial elements in the industry, where unexpected failures can interrupt production cycles, reduce profits, and raise safety concerns; and therefore an early anomaly detection in motor behavior is highly appreciated. As an extension of the known internet of things (IoT), industrial IoT or IIoT allows connection of motors and their drive units through distributed sensing platforms capable of acquiring operational data related to power, temperature, vibration, and rotational speed. Once these data are transferred through the available IIoT infrastructure and stored appropriately for later off-line processing, the limited availability of labeled fault data remains a major obstacle in practical industrial applications. As a remedy, this study proposes a semi-supervised anomaly detection framework that relies exclusively on non-intrusive three-phase electrical telemetry. Focusing on a commercial offset printing press, high-frequency power measurements were collected from failure-sensitive dryer motors. Time-domain statistical features, including mean, clearance factor, and shape factor extracted from active power, power factor, and current signals, were employed to train an unsupervised One-Class Support Vector Machine (OC-SVM). Experimental results obtained from 84 hours of real industrial telemetry demonstrated the effectiveness of the proposed approach in modeling normal operating behavior, achieving a Recall of 93.79%, a False Positive Rate of 5.12%, and an F1-Score of 94.59%. The developed framework enables early anomaly detection, lowers maintenance expenses, reduces operational downtime, and enhances overall system reliability, supporting the advancement of smart Industry 4.0 environments.
M. Zeidan, S. Aldalahmeh, Z. Haymoor et al.· IEEE Jordan Conference on Ap...· 0 citations