Research on Wind / PV / Storage Hybrid Microgrid and Grid/Islanded Switching Based on Neural Network

In the process of control mode switching of the micro-grid controlled by the master-slave, the output value of the controller will be abruptly changed, causing voltage and current shock and waveform distortion. For this problem, a neural-based control strategy and improved loop controller, realizes the smooth transition of the micro-grid and off-grid, ensures stable operation of the important load. The traditional PI controller has a simple structure, but has limitations: when the controlled object is too complicated, the parameters of the PI controller are difficult to adjust. The self-tuning of PI parameters is realized by combining BP neural network with traditional PI controllers. Finally, the switch between the grid connection and islanding is simulated by Matlab to verify the feasibility and effectiveness of the control strategy.