基于RFNN补偿的直线伺服系统反推控制

针对永磁直线同步电机(PMLSM)伺服系统,在分析影响直线伺服跟踪精度因素的基础上,采用智能反推控制策略对该伺服系统进行有效的补偿控制。考虑参数变化、外部负载扰动和摩擦力等不确定因素对系统伺服性能的影响,设计基于递归模糊神经网络(RFNN)的反推控制器,利用了递归神经网络具有捕获系统动态信息的优点,可实时补偿不确定因素对跟踪性能的影响。仿真结果表明,控制策略明显降低了不确定因素对系统性能的影响,从而显著提高了直线伺服系统的位置跟踪精度。

The Backstepping Control Based on RFNN Compensation for Linear Servo System

For permanent magnet linear synchronous motor （ PMLSM ） servo systems, on the base of analyzing the factors that influence the tracking precision of servo system, an intelligence backstepping control is introduced to suppress further the tracking error. Considering the influence of the lumped uncertainties with parameter variations, external force disturbances and frictions for PMLSM, a backstepping controller based on recurrent fuzzy neural network （ RFNN ） is designed. The advantages of RFNN with capturing the dynamic information of system are used, can compensate real time the influence of tracking performance. Simulation results show that the proposed algorithm can reduce effectively the influence of the lumped uncertainties on system performances. The tracking performances of the PMLSM servo system are significantly improved.