A Multistate Load State Identification Model Based on Time Convolutional Networks and Conditional Random Fields | IEEE Journals & Magazine | IEEE Xplore

A Multistate Load State Identification Model Based on Time Convolutional Networks and Conditional Random Fields

DatasetsAvailable

Impact Statement:According to the 2019 data, industrial loads account for 41.9% of total global electricity consumption. Improving the capability of industrial load operation status monit...Show More

Abstract:

The increasing concern for carbon emissions requires smarter monitoring solutions for industrial load operating conditions. Nonintrusive load monitoring is one possible m...Show More
Impact Statement:
According to the 2019 data, industrial loads account for 41.9% of total global electricity consumption. Improving the capability of industrial load operation status monitoring is a useful aid to improve energy use efficiency. NILM method is a low-cost method to enhance the monitoring capability, and NILM is not effective in industrial load condition monitoring due to factors such as the long single-state operation time of industrial loads and complex operating environment. The temporal convolutional network-conditional random field (TCN-CRF) method proposed in this paper effectively realizes the modeling of long-range state dependencies and achieves various advantages over conventional methods in terms of long-range feature capture, dependency modeling, and error identification suppression. It is expected to contribute to the improvement of energy utilization efficiency and alleviate the energy crisis after its promotion.

Abstract:

The increasing concern for carbon emissions requires smarter monitoring solutions for industrial load operating conditions. Nonintrusive load monitoring is one possible monitoring solution that can provide users with detailed information on the operating status of the load, thus improving energy management capabilities. We note that industrial loads have two outstanding characteristics compared to domestic loads: 1) long single-state operation times; and 2) highly directional state transfer probabilities. Considering these two characteristics, we propose the temporal convolutional network-conditional random field architecture to construct a deep learning architecture with long-term dependencies capability and probabilistic transfer modeling. The results show that the proposed architecture can achieve over 97% recognition accuracy on the industrial load dataset used, and can suppress the problem of frequent switching of recognition state results.
Published in: IEEE Transactions on Artificial Intelligence ( Volume: 4, Issue: 5, October 2023)
Page(s): 1328 - 1336
Date of Publication: 02 September 2022
Electronic ISSN: 2691-4581

Funding Agency:


Contact IEEE to Subscribe

References

References is not available for this document.