| With the permanent magnet synchronous motor(PMSM) servo system has been widely used in high performance motion control field, it’s particularly important to improve the precision, stability, static and dynamic performance of the control system. The inertia of motor and load is one of the main factors which affect the controller parameters of the system. Therefore, if the inertia of PMSM servo system can be identified on time, then parameters of the controller can be adjusted to compensate for the impact brings by its change. Thus can effectively improve control performance.Firstly, some common parameter identification algorithms are analyzed. Then the recursive least square method to the inertia identification are presented since its good characteristics. However, if the sampling time is large enough, this recursive algorithm can become invalid. That’s because the new data lose the ability to modify the previous estimates. So the forgetting factor is proposed to put into the old data in order to reduce the amount of information provided by the old data. Then specific process of the PMSM inertia identification is provided.Secondly, since the velocity is usually derived from the angular position using simple differences. Thus it can inject a phase lag. So Luenberger observer is used to reduce or remove phase lag. It can improve the calculation accuracy of the velocity and inertia identified.Finally, an inertia identification simulation model in MATLAB / Simulink based on the PMSM closed-loop control system is designed. Then the experimental platform is built based on the VECTODRIVE VD600 drive, GOLDEN AGE PMSM and VECTONUM V8 IPC. Besides, some main software is designed.The inertia identification algorithm based on recursive least square method combined with speed observer is proved to be feasible and effective by simulations and experiments. The inertia identification is quick and accurate enough either on condition of no-load or with constant load when the velocity changes. |
PDF Full Text Request