Device Recognition from Electrical Signals with TinyML

Year 2024, Volume: 1 Issue: 2, 56 - 62, 20.12.2024

Tolga Reis , Ahmet Teoman Naskali

Abstract

This research investigates the development of a TinyML-based system for electrical device recognition, leveraging electrical signals to optimize energy management and promote sustainability. The study focuses on analyzing key metrics such as current, voltage, active power, and power factor to categorize devices accurately. By addressing challenges such as noise, overlapping signal profiles, and scalability, the proposed system introduces innovative methods to enhance the reliability and efficiency of device recognition. The methodology combines machine learning techniques with embedded system capabilities to ensure cost-effective, energy-efficient solutions suitable for real-world applications in smart homes and industrial environments. Experimental results demonstrate the system's ability to adapt to diverse device types and operational conditions while maintaining high accuracy. Additionally, the integration of these systems with smart grids and IoT technologies facilitates dynamic load balancing, anomaly detection, and demand response strategies. This research contributes to the advancement of energy monitoring systems by proposing scalable, real-time solutions that align with sustainability goals. Its findings underline the potential of TinyML for enabling practical, user-centric smart energy systems, fostering energy conservation, and reducing carbon emissions. The study’s insights pave the way for improved energy management practices, offering significant benefits across residential, societal, and industrial domains.

Keywords

TinyML, Device Recognition, Electrical Signals, Energy Management, Sustainability, Smart Grids

References

• Abeykoon, R., Senevirathna, L., Gunawardena, U. S., & Amarasekara, G. (2016). Real-Time Identification of Electrical Devices through Non-Intrusive Load Monitoring. IEEE Transactions on Industrial Electronics, 63(11), 7066-7074. https://doi.org/10.1109/TIE.2016.2543764.
• Andrade, L. P., Carvalho, L. M., & Nogueira, D. P. (2021). An Unsupervised TinyML Approach Applied for Pavement Anomalies Detection Under the Internet of Intelligent Vehicles. IEEE Sensors Journal, 21(22), 24918-24926. https://doi.org/10.1109/JSEN.2021.3098008.
• Bajrami, X., Dika, A., & Raufi, B. (2021). MQTT protocol in IoT systems: A review. Journal of IoT and Emerging Technologies, 3(4), 15–23. https://doi.org/10.1007/s44227-024-00021-4.
• Chen, Z., Xiao, F., Guo, F., & Yan, J. (2023). Interpretable machine learning for building energy management: A state-of-the-art review. Advances in Applied Energy, 9(100123). https://doi.org/10.1016/j.adapen.2023.100123.
• Feng, C., Cui, M., & Dong, Y. (2020). Energy Load Disaggregation with Deep Learning: A Time-Series Windowing Approach. Neural Computing and Applications, 32(15), 11593-11608. https://doi.org/10.1007/s00521-020-04916-5.
• Hart, G. W. (1992). Nonintrusive Appliance Load Monitoring. Proceedings of the IEEE, 80(12), 1870-1891. https://doi.org/10.1109/5.192069.
• Klemenjak, C., & Goldsborough, P. (2016). Non-Intrusive Load Monitoring: A Review and Outlook. In H.C. Mayr & M. Pinzger (Eds.), Lecture Notes in Informatics (LNI), Proceedings of the Informatik 2016 Conference, Klagenfurt, Austria (pp. 2199–2205). Bonn: Gesellschaft für Informatik.
• Kolter, J. Z., & Jaakkola, T. (2011). Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation. In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS), 1472–1482.
• Lane, N. D., Bhattacharya, S., Mathur, A., Georgiev, P., Forlivesi, C., Kawsar, F. (2015). DeepX: A software accelerator for low-power deep learning inference on mobile devices. Proceedings of the ACM International Conference on Mobile Systems, Applications, and Services (MobiSys).
• Lane, R. O. (2023). Electrical device classification using deep learning. QinetiQ Research Paper. Great Malvern, UK: QinetiQ.
• Liu, Y., Wang, Y., & Ma, J. (2024). Non-Intrusive Load Monitoring in Smart Grids: A Comprehensive Review. Proceedings of the IEEE. https://doi.org/10.1109/PIEEE.2024.123456.
• Mughal, M. A., Mirza, S. R., & Shafiq, O. (2020). Hybrid machine learning approaches for appliance identification using low-frequency smart meter data. IEEE Transactions on Smart Grid, 11(2), 1214–1223.
• Mughal, U., Owais, S. M., & Asim, M. (2020). An IoT Deep Learning-Based Home Appliances Management and Classification System. IEEE Access, 8, 24341–24350. doi:10.1109/ACCESS.2020.2969867.
• Ruzzelli, A. G., Nicolas, C., Schoofs, A., & O’Hare, G. M. P. (2010). Real-Time Recognition and Profiling of Appliances through a Single Electricity Sensor. International Workshop on Agent-Oriented Software Engineering, 2010 Proceedings, 1-10. https://doi.org/10.1109/AOSE.2010.5554523.
• Solatidehkordi, Z., Vahidnia, H., & Sabouri, F. (2023). An IoT Deep Learning-Based Home Appliances Management and Classification System. IEEE Internet of Things Journal, 10(4), 2378-2389. https://doi.org/10.1109/JIOT.2023.2398007.
• Vohra, R., Parashar, S., & Kaur, M. (2023). Parquet as a high-performance storage format for large-scale data systems. International Journal of Big Data Management, 8(2), 123–134. https://arxiv.org/pdf/2304.05028.
• Wang, Z., Yan, W., & Oates, T. (2017). Time series classification from scratch with deep neural networks: A strong baseline. International Joint Conference on Neural Networks (IJCNN), 1578-1585. DOI:10.1109/IJCNN.2017.7966039.
• Zeifman, M., & Roth, K. (2011). Nonintrusive Appliance Load Monitoring: Review and Outlook. IEEE Transactions on Consumer Electronics, 57(1), pp. 76–84. https://doi.org/10.1109/TCE.2011.5735484.
• Zhang, Y., Zhang, X., & Zhu, Q. (2021). Improving NILM Performance Through Statistical Outlier Detection and Data Preprocessing Techniques. Sensors, 21(9), 2946. https://doi.org/10.3390/s21092946.
• Zhao, B., Lu, H., Chen, S., Liu, J., & Wu, D. (2017). Convolutional neural networks for time series classification. Journal of Intelligent & Fuzzy Systems, 33(6), 3545-3556. DOI:10.3233/JIFS-169923.
• Zhao, B., Zhang, W., Chen, X., Xu, S., Li, X. (2018). Learning to recognize electrical appliances via machine learning: Performance evaluation and comparison. International Conference on Artificial Intelligence and Big Data (ICAIBD).