The Study Of AC Servo System And Parameters Identification Algorithms

With the development of modern industrialization and the improved performance of system, AC servo system has come into digital and intelligent era. All digital AC servo system is excellently flexible so that we can easily implement new control strategies without ameliorating system hardware, which is greatly convenient to realize intelligent control strategies. The performance of AC servo system is highly influenced by uncertainties of unpredictable variation of mechanical parameters and external load disturbances. In order to enhance the system static and dynamic performances, it is necessary to identify the mechanical parameters and auto-tune the speed controller. Several parameter identification methods especially the online identification method of system inertia based on MRAC are proposed in this dissertation. Then, the scheme for auto-tuning the speed controller is presented as well. The experiment results suggest that this algorithm is feasible. The load torque is also difficult to be measured directly, so it needed to be online identified. This dissertation adopt a compensate control algorithm, which could dynamically compensate speed variation. The experiment results show that, this compensatory control strategy obviously enhanced system robustness. According to the characteristics of LF2407A, specific designs of software and program flowcharts are presented. Based on those identification and compensation results mentioned above, more experiments studies on the control performance of multi-loop system are conducted. A summery of whole thesis is given in the end.