【摘要】 针对数控加工中常用的滚珠丝杆高精度伺服系统，提出了一种基于双编码器的反馈控制技术。该方法将基于模型的摩擦补偿与神经网络滑模控制相结合，可以有效补偿全闭环系统中的非线性干扰因素，提升伺服系统的动态响应特性。在Matlab/Simulink环境下进行了仿真实验，结果证明，RBF神经网络可以有效逼近伺服系统所受的非线性干扰力矩。搭建了基于TwinCAT3的双编码器实验平台，进行了位移跟踪实验。实验结果表明，相比于PID控制，自适应控制可有效提升伺服系统的动态位移精度，改善运动特性。更多还原

【Abstract】 Aiming at the high-precision ball screw servo system commonly used in CNC machining, a feedback control technology based on dual encoders is proposed. This method combines model-based friction compensation with neural network sliding mode control, which can effectively compensate the nonlinear interference factors in the fully closed loop system and improve the dynamic response characteristics of the servo system. The simulation experiment is carried out in the Matlab/Simulink environment, and the result proves that the RBF neural network can effectively approximate the nonlinear disturbance torque suffered by the servo system. A dual-encoder experimental platform based on TwinCAT3 is built, and displacement tracking experiments are carried out. The experimental results show that, compared with PID control, adaptive control can effectively improve the dynamic displacement accuracy of the servo system and improve the motion characteristics.更多还原