A novel hybrid deep learning approach including combination of 1D power signals and 2D signal images for power quality disturbance classification

作者：

Highlights：

• It is the first method for power quality analysis on combining 1D and 2D CNN features.

• A robust and effective framework is created by combining signal and image features.

• It provides better performance than state-of-the-art power quality classification methods.

摘要

•It is the first method for power quality analysis on combining 1D and 2D CNN features.•A robust and effective framework is created by combining signal and image features.•It provides better performance than state-of-the-art power quality classification methods.

论文关键词：PQD,1D CNN,2D CNN,Classification,Power signal,Signal disturbance

论文评审过程：Received 15 November 2020, Revised 29 January 2021, Accepted 22 February 2021, Available online 27 February 2021, Version of Record 11 March 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114785