A novel hybrid deep learning approach including combination of 1D power signals and 2D signal images for power quality disturbance classification.
Hatem SindiMajid Kamal A. NourMuhyaddin Jamal H. RawaSaban ÖztürkKemal PolatPublished in: Expert Syst. Appl. (2021)
Keyphrases
- power quality disturbance
- power quality
- signal processing
- deep learning
- decision trees
- power system
- electric power
- wavelet transform
- power supply
- single phase
- image classification
- image features
- input image
- lifting wavelet
- pattern recognition
- probabilistic neural network
- unsupervised learning
- machine learning
- multiple images
- training set
- object recognition
- test images
- multiscale
- learning algorithm
- high frequency
- maximum likelihood
- image processing