摘要
针对硅锭电火花线切割电极丝存在张力不稳定、幅度变化大等问题,设计一种双边张力伺服控制系统。建立张力伺服系统数学模型,根据该模型设计一种基于RBF神经网络的PID控制算法并与传统PID控制算法进行仿真对比。仿真结果表明:优化后的RBF-PID可在15 ms内完成PID参数整定,实现了超调量<5%、调整时间<25 ms的控制性能,克服了普通PID系统参数改变后需再次整定、自适应性差的缺点。硅锭电火花切割实验表明:RBF-PID控制算法的张力波动率比普通PID算法降低了40%,显著抑制了走丝速度、预紧力变化等因素对电极丝张力稳定性的影响。
For the problems of unstable tension and large amplitude changes in the wire electrode wire of silicon ingot WEDM,a bilateral tension servo control system was designed.A mathematical model of the tension servo system was established,and according to the model,a PID control algorithm based on RBF neural network was designed and compared with the traditional PID control algorithm.The simulation results show that the optimized RBF-PID can realize rapid self-tuning of PID parameters within 15 ms,and achieve the control performance of less than 5%overshoot and less than 25 ms adjustment time.It overcomes the shortcoming of poor adaptability that needs to be readjusted after the parameters of the ordinary PID system are changed.The silicon ingot WEDM cutting experiment shows that the tension fluctuation rate of the RBF-PID control algorithm is 40%lower than that of the ordinary PID algorithm,which significantly suppresses the influence of factors such as wire speed and pre-tightening force change on the tension stability of the electrode wire.
作者
刘鹏
罗福源
孙凌云
LIU Peng;LUO Fuyuan;SUN Lingyun(College of Mechanic and Electronic Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《机械制造与自动化》
2021年第6期202-206,238,共6页
Machine Building & Automation
基金
江苏省自然科学基金优秀青年基金项目(BK20160084)
南京市产学研后补助项目资助(201722014)。