基于TSK型递归模糊神经网络的加工中心双直线电机交叉耦合同步控制

为解决高精密龙门移动式镗铣床加工中心X轴的两台直线电机的同步跟踪问题,采用一种TSK型递归模糊神经网络(TSK-type recurrent fuzzy neural network,TSKRFNN)与交叉耦合控制(cross-coupled control,CCC)相结合的控制方法。利用TSKRFNN解决单轴PMLSM受到参数变化和外界扰动等不确定性影响的问题,估计并补偿总不确定性因素并在线调整网络参数,从而抵抗外界干扰,提高系统的鲁棒性和跟踪性。其次,为解决双直线电机运行时存在的参数不匹配性和耦合问题,将CCC与TSKRFNN相结合,CCC可以将单轴跟踪误差按照一定比例分配给两台永磁直线同步电动机(permanent magnet linear synchronous motor,PMLSM),以抑制由不同步问题引起的不平衡扭矩,从而使系统高精度同步运行。最后,通过双直线电机平台上证明所提方法的有效性,实验结果表明该方法鲁棒性及跟踪性优良,可以较好地满足加工中心同步控制的要求。

Cross-Coupled Synchronization Control of Dual Linear Motor in Machining Center Based on TSK-type Recurrent Fuzzy Neural Network

In order to solve the asynchronous tracking problem of dual linear motors on X axis of high precision gantry moving machining centers, a control method combined TSK-type recurrent fuzzy neural network (TSKRFNN) with cross-coupled control (CCC) was adopted. TSKRFNN is used to solve the problem that single axis PMLSM is affected by the uncertainty of parameter variation and external disturbance, estimate and compensate the total uncertainty factors and adjust the network parameters on-line, so as to resist the external disturbance and improve the robustness and tracking performance of the system. Secondly, in order to solve the problem of parameter mismatch and coupling in the operation of dual linear motor, CCC and TSKRFNN were combined. CCC can distribute the single-axis tracking error to two PMLSMs according to a certain proportion, so as to restrain the unbalanced torque caused by asynchronization and ensure the high-precision synchronous operation of the system. Finally, the validity of the proposed method is proved on the dual linear motor platform. The experimental results show that the method has good robustness and tracking performance, and can meet the requirements of machining center synchronous control.