LSTM-MFAC Based Secondary Frequency Control Strategy for Microgrid

The parameter design of the secondary frequency controller for microgrids requires the use of a microgrid model. However, due to the complexity and diversity of the microgrid system structure, as well as the time-varying of the microgrid system structure caused by operation in extreme environments, microgrid modeling is very challenging. To address the above problems, this paper proposes a model-free adaptive control (MFAC) method. And for the problems that some parameters of the MFAC control need to be pre-tuned before application and the parameters cannot be updated online, an improvement scheme based on the Long Short-Term Memory (LSTM) neural network is proposed. The proposed LSTM-MFAC based secondary frequency control method for microgrids can dynamically update the control model through real-time operation feedback data, thus optimizing the control performance, improving the adaptability of the microgrid frequency stability to environmental influences, and improving the quality of power supply. Finally, the paper carries out simulation analyses to verify the effectiveness of the proposed method.