Identification of Low Frequency Oscillation Mechanism Based on Deep Learning

With the development of power grid technology, the scale of modern power grid is expanding, and the operation mode is becoming more and more complex. Low frequency oscillations are always important factors threatening the security and stability of power grid. It is difficult to distinguish the negative damping mechanism oscillation from the forced power oscillation. In recent years, the rapid development of the third generation artificial intelligence technology provides a technical basis for the identification of power oscillations with different mechanisms. This kind of artificial intelligence technology, represented by deep learning, has made remarkable achievements in classification, aggregation and other fields. This paper studies the construction of deep learning model and its application in the discrimination of different oscillation mechanisms, and proposes a recognition method. This method is based on the deep belief network of negative damping oscillation and forced power oscillation to build a deep learning model, which can be applied to feature extraction of negative damping oscillation and forced power oscillation.