摘要
考虑机械传动环节结构柔性对进给驱动系统整体动态性能的影响,首先构建滚珠丝杠进给系统状态空间模型,在此基础上,基于数字化模块仿真,进一步实现与伺服控制系统的集成建模,并通过实验对集成模型进行了验证。在此基础上,鉴于级联控制及其动态特性非线性特征下伺服控制参数最优值并不固定,通过控制参数值优化选取对进给驱动整体动态性能进行优化具有一定的局限性,提出神经网络自适应电流与速度前馈控制设计策略(NNFFC),基于控制参数时变最优参数组合自适应调节与电流速度前馈,对滚珠丝杠进给驱动动态响应性能进行优化,优化后整体进给驱动系统跟随误差由0.252 6降为0.111 5,干扰峰值由0.019降为0.007 0,上升时间由原来的0.128 3 s减少为0.107 5 s。
Considering the influence of the mechanical transmission flexibility on the servo control system and its control parameters, and the feed drive dynamic performance, the ball screw feed system state space model was firstly constructed in order to integrate to servo control system. Then, this state space model was integrated with servo control system model based on the digital module simulation method. And the correctness of this integration model was verified by experiment test. The test result showed that the error between experiment curve and the simulation curve was small, and the simulation result was 50. 45 Hz, -1.78 dB in the first order natural frequency while the experiment result was 47.12 Hz, -4.46 dB in the first order natural frequency. In view of uncertainty of servo control parameter optimal value resulting from the nonlinear of cascade control and its dynamic characteristic, feed drive dynamic performance optimization was limited to some extent by the method of servo control parameter optimization choice. Hence, the neural network adaptive current and speed feed forward control design strategy was put forward. Based on this ball screw feed drive integration modeling, servo control parameter time-varying optimal selection as well as current and speed feed forward control strategy were combined to realize the ball screw feed drive dynamic response optimization. The following error decreased from 0. 252 6 to 0. 111 5, and interference peak decreased from 0. 019 to 0. 007 0, rising time decreased from 0. 128 3 s to 0. 107 5 s. It can be seen that the response speed of ball screw feed drive system was faster while the following error was smaller, and the system anti-interference ability was better at the same time. The research of this paper provided the important theoretical basis and method guidance for the study of the overall dynamic performance and optimization of the ball screw feed drive system.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2016年第11期376-383,共8页
Transactions of the Chinese Society for Agricultural Machinery
基金
"十二五"国家科技重大专项(2012ZX04005-031)
关键词
数控机床
滚珠丝杠进给驱动
神经网络自适应
电流速度前馈
动态响应性能
NC machine tool
ball screw feed drive
neural network adaptive
current and velocity feedforward
dynamic response property