Deep Flexible Transmitter Networks for Non-Intrusive Load Monitoring of Power Distribution Networks

• Main Authors: Chenxiao Ma, Linfei Yin
• Format: Article
• Language: English
• Published: IEEE 2021-01-01
• Series: IEEE Access
• Subjects: Non-intrusive load monitoring; deep backpropagation neural networks; flexible transmitter networks; load modeling; embedded calculation
• Online Access: https://ieeexplore.ieee.org/document/9502688/

With the massive development of various new energy sources, the balance of supply and demand in the power grid faces a considerable challenge. A reliable way is to perform demand-side management on energies. For demand-side management, monitoring the load connected to power distribution networks is necessary. This paper proposes deep flexible transmitter networks to quickly monitor the load when the load is connected to power distribution networks. The proposed algorithm combines flexible transmitter networks and deep backpropagation neural networks. After measuring the waveforms of loads, the algorithm is trained by the measured operational data. The testing results show that the proposed deep flexible transmitter networks can accurately monitor the load connected to power distribution networks. Compared with the deep backpropagation neural networks, the proposed algorithm improves the monitoring accuracy by more than 5%. The speed of the proposed algorithm is verified in experiments. After testing on embedded devices, the proposed algorithm can satisfy the requirements of edge computing systems.