表贴式永磁同步电机的多参数在线辨识

发布时间：2018-11-16 12:10

【摘要】：针对表贴式永磁同步电机的多参数辨识,提出一种基于模型参考自适应理论的在线辨识方法,可在同一模型中对定子电阻、交直轴电感和永磁体磁链进行辨识。该方法借助Popov超稳定理论设计自适应律,保证了特定条件下辨识的收敛性,同时采用分步辨识,克服多参数辨识的欠秩问题,保证了辨识结果的唯一性,从而实现多参数辨识。分步辨识首先利用电机q轴电流方程估算出交直轴电感,之后对d轴注入一个小电流值,利用已获得的电感值辨识出定子电阻和转子磁链,同时所辨识出的磁链可进一步用来在线估算定子电阻值的变化。仿真和实验结果表明,该方法可以在同一模型中准确辨识出电机的多个参数,且无需考虑电机的运行状态,具有实用性。
[Abstract]:An on-line identification method based on model reference adaptive theory is proposed for multi-parameter identification of permanent magnet synchronous motor (PMSM) with surface mount. It can be used to identify stator resistance, AC axis inductance and permanent magnet flux in the same model. In this method, the adaptive law is designed with the help of Popov's hyperstability theory to ensure the convergence of identification under certain conditions. At the same time, step by step identification is used to overcome the problem of lack of rank of multi-parameter identification, and the uniqueness of the identification result is guaranteed, thus the multi-parameter identification is realized. In stepwise identification, the AC straight axis inductance is estimated by using the Q axis current equation of the motor, then a small current is injected into the d axis, and the stator resistance and rotor flux are identified by the obtained inductance. At the same time, the identified flux can be further used to estimate the change of stator resistance on line. The simulation and experimental results show that the proposed method can accurately identify multiple parameters of the motor in the same model and is practical without considering the running state of the motor.
【作者单位】：北京航空航天大学自动化科学与电气工程学院;
【基金】：国家自然科学基金(50807002) 电力系统国家重点实验室(SKLD10KM05)资助项目
【分类号】：TM341

【参考文献】