考虑参数扰动的风力发电系统并网逆变器积分型连续滑模控制策略

In the burgeoning field of clean wind energy, integrating wind power generation systems (WPGSs) into the power grid has become a critical area of research and development. The performance and reliability of the grid-connected inverter (GCI) play a pivotal role in ensuring efficient and stable electricity supply from wind farms. However, the GCI often faces parameter disturbances during operation, which are exacerbated by fluctuations on both the generation side and the grid side, significantly impacting the quality of grid connection and the overall performance of the WPGSs. However, current GCI control approaches neglect disturbance parameters and compromise the system's resilience and stability, making it vulnerable to operational unpredictability. Therefore, a novel approach is introduced, considering the GCI's typical internal and external parameter perturbations. A robust control system is developed using the continuous higher-order super-twisting sliding mode control (STSMC) strategy. The specific design method of the STSMC strategy for the GCI system is as follows. Firstly, a model of the GCI system is constructed with a DC voltage outer loop, an AC current inner loop, and an AC voltage inner loop, considering DC bus instantaneous power fluctuations and internal parameter disturbances of the GCI. Compared with the traditional PI strategy, the designed parameter perturbation model effectively suppresses internal parameter disturbances in the GCI when paired with the STSMC. Additionally, the design of the DC voltage outer loop has successfully stabilized the DC input power of the GCI, thereby reducing the impact of disturbances on the WPGS generation side on the performance of the GCI. Secondly, the integral-type STSMC controllers with time-varying and predictable gains are constructed for the GCI current and voltage loops. Thus, the interaction power between the GCI and the grid side can be controlled, and energy loss from excessive control gain caused by system uncertainty boundaries can be improved. Compared with the traditional terminal sliding mode control (TSMC) and Twisting algorithm SMC, the time-varying gain of the current loop helps the STSMC overcome the slow current response speed inherent in the Twisting algorithm. Moreover, the predictable gain of the voltage loop produces a smaller steady-state error than the TSMC. Finally, the stability and finite-time convergence of the proposed control strategy are proved based on Lyapunov theory. Simulations and experiments are designed considering parameter disturbances under multiple conditions. The STSMC system exhibits a 79.55% reduction in output power disturbance error compared to PI control, with the output power ripple amplitude lowered by 25.4% and 52.72% compared to TSMC and Twisting, respectively. The designed predictable gain and integral control term reduce the harmonic content of the GCI output capacitor voltage and current by 0.27% and 5.62%. The adoption of time-varying control gain further constrains the output power error to within 4.92% and 0.15%. In subsequent research, the investigation will be expanded to account for the practical influence of switching losses within the GCI and the effects of grid-side harmonics on the control system, aiming to construct a complete mathematical model of the GCI. Furthermore, the strategy's adaptability for application under unbalanced grids will be examined in depth.