Abstract
In the non-invasive load identification task, feature overlap exists when only a single load feature is used to identify equipment, which cannot meet the requirements of fine-grained equipment classification. Therefore, this paper proposes a non-invasive fine-grained load identification method based on color coding. Firstly, the Fryze power theory is used to decompose the high-frequency sampling current into active and reactive current, and the high-frequency sampling voltage and trace image are obtained. Then, the trajectory image is processed by color coding technology, and the active current, trajectory change information and instantaneous power are fused in R, G and B channels respectively to obtain the color V-I trajectory image. Finally, the Region-based fully Convolutional Networks (R-FCN) model is constructed to extract features from color V-I track images and realize device classification. On this basis, an autonomous learning method is proposed to update the load identification model autonomously. The recognition effect of the algorithm and the autonomous learning method were tested using PLAID data set. The results show that the proposed method increases the amount of information carried by V-I trajectory and enhances the uniqueness of load characteristics, thus realizing fine-grained identification of equipment. The self-learning method can learn new electrical appliances and update the model, which improves the adaptability of load identification model.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.