A Deep Learning Method based on Long Short Term Memory and Sliding Time Window for Type Recognition and Time Location of Power Quality Disturbance

摘要

It is extremely important to recognize type and locate stating-ending times of power quality disturbance for adopting corresponding measures to suppress disturbances. The development of machine learning and artificial intelligence technology provides an effective way for dealing with power quality disturbance. In this paper, a deep learning method based on long short term memory and sliding time window for type recognition and time location of power quality disturbance is proposed. To be specific, the collected power quality disturbance wave is firstly transformed into a gray scale image and then the model based on deep learning with long short term memory (LSTM) stacked is constructed to automatically learn features. After that, the type of power quality disturbance is recognized, and furthermore, the starting-ending times are also located. Finally, experiment is carried out to verify the effectiveness of the proposed method.