上海科技大学知识管理系统

Inverter based resources (IBRs) are widely adopted to integrate large scale renewables into modern power systems. When analyzing IBRs’ performances during power system transients, the control strategies of IBRs are typically required, to enable various applications such as monitoring, operation, control, protection, fault location, and more. However, in practice, control strategies are often encapsulated as black boxes by manufacturers, and utilities may not have the access to relevant information. Consequently, it is imperative to determine the detailed control strategies for practical IBRs. In this paper, a physics-informed data-driven classification method is proposed to identify the control strategy of IBRs during transients. The proposed method only requires single-end voltage and current measurements at the IBR side during transients such as faults. First, according to the physics information within typical control strategies, different physics-based feature extraction tools are employed to preprocess the measurement data. Then, the convolutional neural network (CNN) is adopted to extract data features. In addition, an independent set of testing data, distinct from the training dataset, is employed to assess the performance of the proposed method. Numerical experimental results clearly indicate the effectiveness of the proposed physics-informed CNN based control strategy classification method.