Modeling Voltage-response Characteristics of Renewable Energy Sources Based on the Fusion of Deep Learning and Knowledge Graph

Abstract

The problem of modeling voltage-response characteristics of renewable energy sources (RESs) is considered. A deep neural network-based method is adopted to track time-variant response characteristics of renewable energy sources. Each deep learning (DL) model is only suitable for a certain range of variations of weather and system conditions. A knowledge graph is introduced to equip the DL models with the transferability among various conditions. Simulations on a 5-bus test system have demonstrated the effectiveness of the proposed approach.