基于滑模观测器的直线电机无速度传感器控制

Speed Sensorless Control of Linear Motor Based on Sliding Mode Observer

研究了一种基于模型参考自适应算法(MRAS)和滑模观测器(SM0)的速度估计方案用于直线牵引电机无速度传感器控制系统。首先,给出考虑动态边端效应的直线牵引电机状态空间模型;在状态空间模型基础上,设计一种以定子电流为滑模面的SMO,并以此为可调模型。考虑直线牵引电机在运行过程中,定子电阻变化会带来速度估计误差增加甚至控制系统的不稳定性,在所研究的速度估计方案中引入定子电阻在线辨识方案,以降低速度估计误差并提高系统稳定性。最后,对该速度辨识方案进行仿真和硬件在环测试并与传统的SM0进行对比,结果验证了该方案的可行性。

A speed estimation scheme by combining model reference adaptive system (MRAS) with sliding mode observer (SMO) in the sensorless-vector-controlled linear induction motor(LIM) drives is proposed.Firstly,the state space-vector model of the LIM considering the dynamic end effects is shown.On the basis of that, the SMO model is built, which is considered as the adaptive model of the MRAS-based scheme.Because the variation of stator resistance may lead to large speed estimation error and even system instability, a parallel stator resistance on-line identification scheme is introduced in the proposed speed estimation scheme to reduce the speed estimation error and improve the stability of sensorless-control system.The effectiveness of the proposed speed estimation algorithm has been verified and compared with the performance of the conventional speed estimation scheme based on single-manifold SMO by the simulation and hardware-in-the-loop tests.