| AC servo system plays an important role in industrial applications and our daily life. After decades of development, servo technology has become increasingly mature. However, as the era of intelligent manufacturing arrives, the upgrading of industrial automation has put forward higher requirements on servo technology. Improving the performance of servo system with finite stiffness connection between motor and load has drawn increasing attention. And this paper focus on solving this issue.Firstly, the mathematical model of flexible servo system was built and the transfer function between speed and torque of flexible system is derived. Then the causes and factors of resonance phenomena were analyzed and the influence law was summarized based on that. A method of resonance suppression based on 4-order Luenberger observer was proposed. The torsional torque could be observed quickly and precisely by pole assignment. By feeding back the estimated torque to current control block, the resonance can be suppressed effectively without additional cost in the hardware system.Meanwhile, the study has found that the moments of motor and load inertia are important factors affecting the resonance phenomenon, which means that identification of motor and load inertia is necessary. Several identification methods were introduced in this paper. Comparing the simulation results of these methods, the identification scheme based on model reference adaptive control was chosen. The identification scheme of inertia of flexible connected motor and load was also introduced.Finally, the effectiveness of proposed resonance suppression method was proved by various simulation results of sets of 2-inertia servo system with different motor/load inertia ratio. The chosen inertia identification algorithm was realized in a 1.5kw PMSM control system based on DSP. The identification result was promising. |
PDF Full Text Request