Intelligent diagnosis and recognition method of GIS partial discharge data map based on deep learning

Li, Jie; Wang, Peng; Lin, Lingqi; Shi, Wei; Li, Xiuwei; Wang, Jiangwei; Zhang, Pipei

doi:10.1109/psgec51302.2021.9542659

Cited by 3 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where y i represents the original data value corresponding to the moment when the set data value is empty, ŷi represents the predicted value of the vacancy, and y represents the average value of the original data value. This study compares the data cleaning of the random forest algorithm [26], the LSTM algorithm [27], and the deep neural network algorithm [28]. It also applies the CC-VC algorithm to the traditional extended Kalman filter (EKF).…”

Section: Experimental Results Verificationmentioning

confidence: 99%

Spatiotemporal Joint Cleaning of Distribution Network Measurement Data Based on Correntropy Criterion with Variable Center Unscented Kalman Filter

Sun

Liu

et al. 2022

Applied Sciences

View full text Add to dashboard Cite

Measurement data cleaning is a key step of edge computing in a distribution network; it is beneficial to improve the state perception and regional autonomy level of a distribution network. According to the temporal and spatial correlation of measurement data in the distribution network, a joint cleaning method of measurement data in a distribution network is proposed based on the correntropy criterion with variable center unscented Kalman filter (CC-VC-UKF). Initially, the mean square error (MSE) in the original unscented Kalman filter (UKF) is replaced by the correntropy criterion with variable center (CC-VC) to improve the accuracy of filtering the measurement data in the distribution network with a non-Gaussian non-zero mean measurement deviation. Then, the measured data of different measuring devices located on the same section of the line are filtered based on the CC-VC-UKF algorithm according to their respective reference time series to improve the signal-to-noise ratio of the measured data. Then, the filtered measured data are filtered and cleaned based on the CC-VC-UKF algorithm according to the space–time joint filtering and cleaning technology. Finally, the method is used to test the measurement data of the distribution network obtained by a power supply company in a city in north China to solve the problem of measurement deviation caused by the existence of space distance. Results show that this method can obtain FTU measurement data with higher precision from network topology based on the filtered TTU measurement data through the media of filtered spatial measurement deviation.

show abstract

Section: Experimental Results Verificationmentioning

confidence: 99%