基于模糊小脑模型神经网络的直线伺服跟踪控制

以永磁直线同步电机为被控对象，在高精度直线伺服跟踪控制系统中，为使输出响应快速地跟踪输入指令，需要克服系统滞后、未建模动态、不确定性以及负载的变化因素等的影响.为此，提出了一种模糊小脑模型(FCMAC)神经网络直接逆控制的方案，可动态地克服上述这些因素的影响.同时，还给出了较详细的原理分析及实现过程.仿真结果表明，此控制方案是十分有效的，能够明显地提高伺服系统的快速跟踪能力，并使系统具有较强的适应性和鲁棒性.

Linear servo tracking control based on fuzzy-CMAC neural network

A permanent magnet linear synchronous motor is served as the controlled plant.In a high precision servo tracking control system,in order to make the output response track perfectly the input command,the influences of lagging phenomenon,unmodelled dynamic,uncertain factors and the variation of the load will be overcome.In the paper,a direct inverse control scheme of the fuzzy cerebellar model articulation controller(FCMAC) neural network is presented.The principle analysis and design realizing process of the system are given in detail.The simulation results show that the proposed scheme is quite effective,the method can obviously improve the tracking ability of the servo system,and make the system have stronger adaptability and robustness.