Analysis of Early Detection of Automatic Weather Station Sensor Failures: A Comparative Study of Supervised Deep Neural Networks and Unsupervised Autoencoders
DOI:
https://doi.org/10.35870/ijsecs.v6i2.7593Keywords:
Automatic Weather Station, Deep Neural Network, Autoencoder, Anomaly Detection, Predictive MaintenanceAbstract
Automatic Weather Stations (AWS) continuously collect and record meteorological data in real time and support weather information services. The reliability of AWS observations depends on sensor performance, as sensor failures, equipment degradation, and communication disturbances may produce anomalous data and reduce data quality. This study aims to develop an early-detection approach for AWS sensor anomalies using a Deep Neural Network (DNN) and an Autoencoder, compare their performance, and identify the sensors most frequently associated with anomalies. The dataset consists of 385,237 observations collected from the AWS Ancol station in North Jakarta from January to September 2025, covering nine meteorological and oceanographic parameters. The study involved data preprocessing, Min-Max normalization, model training, and evaluation using accuracy, precision, recall, F1-score, confusion matrix, and ROC-AUC. The DNN achieved 99.6% accuracy, 0.922 precision, 0.996 recall, 0.958 F1-score, and 1.000 AUC. The Autoencoder achieved 95.8% accuracy, 0.578 precision, 0.023 recall, 0.043 F1-score, and 0.584 AUC. The AUC of 1.000 obtained by the DNN should be interpreted within the characteristics of the labeled dataset used in this study and should not be considered evidence of perfect generalization. Sensor analysis identified wind direction as the parameter most frequently associated with anomalies, contributing 46.67% of the detected anomalies in the DNN and 66.02% in the Autoencoder. These findings indicate that the supervised DNN performed better than the Autoencoder for anomaly detection on the labeled AWS dataset used in this study.
Downloads
References
Akbar, G., Prajitno, P., Ariffudin, & Ananda, N. (2024). Multivariate imputation chained equation on solar radiation in automatic weather station. Jurnal Penelitian Pendidikan IPA, 10(7), 3633–3639. https://doi.org/10.29303/jppipa.v10i7.7679
Ariffudin, A., & Musa, P. (2022). Analisa sistem komunikasi data berbasis Internet of Things (IoT) menggunakan metode PIECES pada sistem pengamatan cuaca otomatis di Badan Meteorologi Klimatologi dan Geofisika (BMKG). Jurnal Meteorologi dan Geofisika, 23(2), 81. https://doi.org/10.31172/jmg.v23i2.831
Duraj, A., Szczepaniak, P. S., & Sadok, A. (2025). Detection of anomalies in data streams using the LSTM-CNN model. Sensors, 25(5), 1610. https://doi.org/10.3390/s25051610
Goodfellow, I. (2016). Deep learning. MIT Press.
Heaton, J. (2018). Ian Goodfellow, Yoshua Bengio, and Aaron Courville: Deep learning: The MIT Press, 2016, 800 pp, ISBN: 0262035618. Genetic Programming and Evolvable Machines, 19(1), 305–307.
Kaya, Ş. M., İşler, B., Abu-Mahfouz, A. M., Rasheed, J., & AlShammari, A. (2023). An intelligent anomaly detection approach for accurate and reliable weather forecasting at IoT edges: A case study. Sensors, 23(5), 2426. https://doi.org/10.3390/s23052426
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444.
Malhotra, P., Ramakrishnan, A., Anand, G., Vig, L., Agarwal, P., & Shroff, G. (2016). LSTM-based encoder-decoder for multi-sensor anomaly detection. arXiv. https://arxiv.org/abs/1607.00148
Pahlepi, R., Soekirno, S., & Wicaksana, H. S. (2023). Solar radiation intensity imputation in pyranometer of automatic weather station based on long short term memory. Ultima Computing: Jurnal Sistem Komputer, 35–40. https://doi.org/10.31937/sk.v15i2.3348
Rhamadania, K., Makhsun, M., Basir, C., & Ananda, N. (2026). Imputation of missing weather data in automatic weather station using the GRU algorithm. Journal of Applied Informatics and Computing, 10(2), 1165–1171. https://doi.org/10.30871/jaic.v10i2.12301
Sakurada, M., & Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. In Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis (pp. 4–11). Association for Computing Machinery. https://doi.org/10.1145/2689746.2689747
Sevruk, B., Ondrás, M., & Chvíla, B. (2009). The WMO precipitation measurement intercomparisons. Atmospheric Research, 92(3), 376–380. https://doi.org/10.1016/j.atmosres.2009.01.016
Utku, A., & Kaya, S. K. (2025). LSTM-based deep learning model for air temperature prediction. IDUNAS Natural & Applied Sciences, 26(32), 26–32. https://doi.org/10.38061/idunas.1596669
Wicaksana, H. S., Winarko, S., & Kurniadi, D. (2023). Treatment on missing data of wind speed measurement in automatic weather station using multivariate imputation chained equation. In 2023 8th International Conference on Instrumentation, Control, and Automation (ICA) (pp. 81–85). IEEE. https://doi.org/10.1109/ICA58538.2023.10273106
World Meteorological Organization. (2021). Guide to instruments and methods of observation (WMO-No. 8).
Yoon, S., Lee, Y., Lee, J.-G., & Lee, B. S. (2022). Adaptive model pooling for online deep anomaly detection from a complex evolving data stream. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 2347–2357). Association for Computing Machinery. https://doi.org/10.1145/3534678.3539348
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Hasbullah Zuhri Hasibuan, Sudarno Wiharjo, Yan Mitha Djaksana
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
1. Copyright Retention and Open Access License
Authors retain copyright of their work and grant the journal non-exclusive right of first publication under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license allows unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
2. Rights Granted Under CC BY 4.0
Under this license, readers are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, including commercial use
- No additional restrictions — the licensor cannot revoke these freedoms as long as license terms are followed
3. Attribution Requirements
All uses must include:
- Proper citation of the original work
- Link to the Creative Commons license
- Indication if changes were made to the original work
- No suggestion that the licensor endorses the user or their use
4. Additional Distribution Rights
Authors may:
- Deposit the published version in institutional repositories
- Share through academic social networks
- Include in books, monographs, or other publications
- Post on personal or institutional websites
Requirement: All additional distributions must maintain the CC BY 4.0 license and proper attribution.
5. Self-Archiving and Pre-Print Sharing
Authors are encouraged to:
- Share pre-prints and post-prints online
- Deposit in subject-specific repositories (e.g., arXiv, bioRxiv)
- Engage in scholarly communication throughout the publication process
6. Open Access Commitment
This journal provides immediate open access to all content, supporting the global exchange of knowledge without financial, legal, or technical barriers.
