Electromagnetic Transient Simulation Method of Power System with High Proportion of New Energy Based on FCN-LSTM

To address the challenges of complex modeling and low computational efficiency in electromagnetic transient simulation for high-penetration new energy power systems, this paper deeply studies the electromagnetic transient simulation method suitable for engineering practice. Firstly, the impedance distribution characteristics of new energy stations are analyzed to establish transient models for renewable energy units, with systematic investigation of collector line topology conversion. Secondly, based on the above characteristics, a new energy station aggregation method is proposed, including parameter conversion of cluster, collector line and power grouping. Finally, a hybrid deep learning framework combining Fully Convolutional Networks (FCN) and Long Short-Term Memory (LSTM) networks is employed to process feature data and generate optimized parameters. The effectiveness of the proposed modeling approach is rigorously validated through electromagnetic transient simulation studies. This research provides substantial technical support for addressing simulation challenges in high-penetration new energy power systems.