Data-Driven Low Frequency Oscillation Mode Identification and Preventive Control Strategy Based on Gradient Descent

Accurate mode identification and effective preventive control strategy of low frequency oscillation (LFO) are vital to improve the small signal stability of power system. This paper proposes a novel data-driven method based on Convolutional Neural Network (CNN) to identify the low frequency modes. The application of feature selection and feature fusion makes the CNN model well adapted to the complexity of large-scale power system. The model can predict LFO modes of a power system in operation scenarios with different power injections and topologies. By invoking the trained CNN model, a preventive control method based on gradient descent is developed to increase the damping of critical modes. Case study demonstrates that the proposed method can efficiently identify the oscillation modes and the obtained preventive control strategy can effectively prevent the occurrence of LFO.