A control strategy of microgrid voltage source inverter based on self-learning sliding mode control

In island mode, voltage source inverter (VSI) supports the frequency and voltage of microgrid. After the complex load is connected, the VSI control performance is degraded, and the output voltage has deviation, negative sequence, waveform distortion and other problems, which further deteriorate the power quality of the microgrid. Different from the traditional strategy that only focus on a single problem, the strategy proposed in this paper can deal with these three power quality problems simultaneously. In this paper, a self-learning sliding mode control strategy is proposed. First, a nonlinear smooth function is used to design an expansion observer, which can estimate the expansion state of the internal uncertainties and external disturbances of the control system. Second, the extended observer is combined with the self-learning synovial control technology to realize the self-learning synovial disturbance rejection control of VSI control strategy. This strategy can improve the stability of voltage control under various working conditions without precise mathematical model. The simulation results show that, compared with the traditional control strategy, this strategy has good robustness under different working conditions.