Design and Implementation of a Low-Cost Edge-AI-Based Smart Energy Monitoring and Protection System for Single-Phase Power Networks

Abdulaziz Usman, A. E Airoboman, Nathaniel Dalla

Abstract


Conventional single-phase energy monitoring and protection systems predominantly rely on fixed threshold techniques and centralized or cloud-based data processing. These approaches often suffer from high false trip rates, poor adaptability to dynamic load conditions, and increased latency, particularly in regions with unstable power infrastructure. This paper presents the design and implementation of a low-cost smart energy monitoring and protection system based on edge artificial intelligence (Edge AI). The proposed system integrates voltage and current sensing with on-device machine learning for real-time anomaly detection, eliminating dependence on internet connectivity. A lightweight neural network model is deployed on a resource-constrained microcontroller using TensorFlow Lite for Microcontrollers. Experimental results demonstrate that the proposed system achieves faster response time, higher detection accuracy, and lower nuisance tripping compared to conventional threshold-based protection schemes. The developed solution is cost-effective, scalable, and well-suited for single-phase residential and small commercial power networks. 


Full Text:

PDF

References


H. G. Rodriguez, A. Patel, and J. Kim, “Smart energy monitoring systems: A comprehensive review,” IEEE Access, vol. 8, pp. 189332–189345, 2020.

S. R. Ghorbani and M. J. Hossain, “IoT-based energy monitoring and management systems: Architectures and challenges,” Sustainable Energy, Grids and Networks, vol. 25, 2021.

TensorFlow, “TensorFlow Lite for Microcontrollers: Embedded machine learning,” 2023.

A. E. Amoran, G. K. Ijemaru, O. Akinsanmi, and W. Agbramu, “Design and construction of a low-cost automatic single-phase transfer switch,” FUOYE Journal of Engineering and Technology, vol. 6, no. 1, pp. 55–63, 2021.

A. Sharma and R. Kumar, “Machine learning-based anomaly detection in power systems,” Measurement, vol. 152, 2020.

M. H. Yousuf and A. Abu-Siada, “Application of artificial intelligence techniques in smart grid protection,” Electric Power Systems Research, vol. 189, 2020.

P. G. Kini and R. C. Bansal, “Embedded intelligence for power quality monitoring,” International Journal of Electrical Power & Energy Systems, vol. 117, 2020.

J. Lee, F. Wu, and Y. Zhao, “Edge computing and AI for real-time power system monitoring,” IEEE Transactions on Industrial Informatics, vol. 17, no. 6, pp. 4231–4240, 2021.

K. E. Nwankwo and S. A. Oyetola, “Low-cost embedded systems for power quality assessment in developing grids,” Measurement, vol. 174, 2021.

M. Chen and Y. Hao, “Edge intelligence: Concepts, architectures, and applications,” IEEE Network, vol. 33, no. 3, pp. 6–12, 2019.

R. Kumar, D. Singh, and P. Verma, “Performance evaluation of smart protection schemes using statistical analysis,” IET Generation, Transmission & Distribution, vol. 15, no. 4, pp. 589–598, 2021.


Refbacks

  • There are currently no refbacks.