Optimization of energy storage VSG Control strategy based on RBF neural networks

Abstract

In response to the issue that traditional energy storage VSGs （virtual synchronous generators） cannot simultaneously possess good disturbance resistance and rapid dynamic response capabilities， a control strategy for energy storage VSGs is proposed， optimizing the dynamic synchronizer using RBF （radial basis function） neural networks. First， a mathematical model for VSG is established， analyzing the impact of rotor inertia and damping coefficient configuration on VSG performance. This analysis reveals the conflicting relationship between parameter configuration and dynamic response versus system dynamic stability. Subsequently， the transient unbalanced power of the rotor is taken as input for a three-layer forward structure RBF neural network algorithm. Through online learning with the RBF neural network algorithm， the optimal transient compensation power is obtained to dynamically adjust the input power of VSG， thereby reducing unbalanced rotor torque and enhancing the transient stability of VSG. Finally， simulation and comparative experiments are conducted to validate the effectiveness of the proposed control strategy.

Keywords

• virtual synchronous generator control
• rbf neural network
• dynamic synchronizer control
• energy storage inverter
• transient stability