Dynamic equivalent modeling for microgrid based on GRU

• Main Authors: Yunlu Li, Junyou Yang, Haixin Wang, Jia Cui, Yihua Ma, Siyu Huang
• Format: Article
• Language: English
• Published: Elsevier 2020-12-01
• Series: Energy Reports
• Subjects: Dynamic equivalent model; Microgrid; Neural network
• Online Access: http://www.sciencedirect.com/science/article/pii/S2352484720314669

The dynamic behaviors of microgrid become more complicated due to the increasing implementation of distributed energy generation, energy storage. It is significant to study the dynamic response for power planning, analysis, and control of microgrid. Hence, an accurate dynamic equivalent model is essential, since it can help to evaluate the performance by simulation to avoid the loss and danger in practical test. However, the dynamic model based on differential equation cannot be established because of the lack of information in most of time. To build dynamic model when microgrid is a black-box system, a gated recurrent unit based neural network is proposed in this paper. The proposed neural network can be treated as a black-box differential–algebraic equations. The structure design and model training procedure are presented in detail. Study cases are implemented to evaluate the performance of modeling method. The comparison results show that the proposed dynamic modeling method can precisely estimate the dynamic response of microgrid.