Power Quality Disturbances Classification Based on Markov Transition Field and Multi-head Attention
Power quality disturbances in the new power systems have become more complex.In order to improve the classification accuracy of the complex power quality disturbances and enhance the noise robustness of the algorithm,a power quality disturbances classification based on the Markov transition field and the multi-head attention is proposed.Firstly,the Markov transition field is used to transform the power quality disturbances time sequence data,and the image modal data is obtained.Then,the image modal data is put into the convolutional neural network for the feature extraction.Finally,the multi-head attention is used to focus on the important part of the feature extraction of the convolutional neural networks and to classify the disturbances.Compared with the conventional image modal conversion,this method has better disturbance classification effect and anti-noise ability.
power quality disturbancesdeep learningMarkov transition fieldmulti-head attention
国家科技期刊平台
NSTL
万方数据
维普
