交流电机无速度传感器矢量控制系统变结构模型参考自适应转速辨识

Variable-structure MRAS Speed Identification for Speed Sensorless Vector Control of Induction Motor

针对变结构控制在参数变化等不确定因素满足“匹配条件”即变化有界时,具有抗外部干扰和内部参数摄动的鲁棒性,但实际中要确定不确定性参数变化的上下界有一定难度的特点。文中基于变结构控制和自适应控制理论,提出一种新颖的变结构模型参考自适应(VS-MRAS)观测器,用于交流电机无速度传感器矢量控制系统的转速辨识。该法将变结构和MRAS进行有机的整合,利用MRAS对系统参数变化的适应能力实时估算不确定因素变化的边界,以改善磁链观测的准确性,进而提高转速辨识的精度。利用仿真方法,对比分析了带速度检测装置的矢量控制系统、基于MRAS转速辨识的无速度传感器矢量控制系统和基于VS-MRAS转速辨识的无速度传感器矢量控制系统的3种情况下转速和磁链响应的动静态性能。证实该方法提高了转子磁链观测的准确度,改善了转速估计的动静态性能。

Variable structure control is robust to external disturbance and interior parameters perturbation while the uncertain factors such as parameters variation meet the ＂matching condition＂, but it is somewhat difficult to find the bounds of uncertain parameters. Proposes a novel stratery of variable structure model reference adaptive system（VS-MRAS） observers for the speed identification of a sensorless vector controlled induction motor drive. This strategy combines MRAS with variable structure control to improve the veracity of flux observer and to improve the accuracy of speed identification making use of MRAS adaptive ability of the variation of system parameters to estimate variable boundary of uncertainty factorss in real time. The dynamic and static performance of the speed estimate is compared and analyzed for three cases that are the vector control system with encoder, the speed sensorless vector control system based on MRAS, and the VS-MRAS respectively. Simulation results verified that this strategy improves the veracity of fluxes observers and the dynamic and static performance of the speed estimation.