基于结构耦合的数控十字滑台机电系统动态特性研究

发布时间：2019-05-07 17:38

【摘要】：数控十字滑台是一种典型的数控机床类设备,它由XY两根空间上相互垂直的电机丝杠系统驱动工作,它具有一般数控机床进给驱动系统的共性特征。因此,深入的研究数控十字滑台双轴耦合进给驱动系统的动态性能对数控机床的后续发展有着十分重要的意义。就一般数控机床而言,其空间结构往往都不是单一的一根轴,多轴结构在高速运行的过程中由于移动部件之间存在着很大的空间时变性,结构与结构之间的相互影响不可忽略。本文针对数控十字滑台结构耦合的特征对其机电系统进行了理论、实验、计算机仿真三方面的综合研究。所做的主要工作包括有:(1)首先,基于数控十字滑台XY轴的结构耦合作用,利用有限元—集中参数法对其电机丝杠系统和移动部件滑块系统分别单独进行了动力学数学建模的工作,并对其模型参数的确定做了理论分析。其次根据动力学数学模型利用三维建模软件和Adams动力学分析软件建立起了数控十字滑台XY双轴耦合的虚拟样机模型,并在Adams软件中完成了其固有模态的分析,讨论了Y轴工作台在不同位置结构耦合作用时十字滑台系统固有模态特性的变化规律。(2)利用Matalab/Simulink建立了数控十字滑台双轴进给驱动系统的控制模型,并结合Adams完成了其双轴耦合作用下十字滑台的机电联合仿真工作,分析研究了其在不同轨迹、速度、加速度进给情况下动态特性的区别。(3)建立了数控十字滑台动态特性实验平台,并利用高分辨率伺服电机码盘、直线光栅尺以及PCI高速脉冲计数卡,测得了实验平台双轴耦合作用下动态特性的振动信号,同时利用Matalb软件对测得的信号进行了数值计算与时域范围下的分析研究。(4)基于BP神经网络理论,利用Matalab建立了数控十字滑台伺服控制系统PID参数自适应调整优化的仿真运行脚本,通过仿真结果表明数控十字滑台在BP神经网络PID自适应调整的控制下,其系统控制精度、稳定性和抗干扰性能都得到了很大提高,证明采用BP神经网络算法能够很好的满足数控十字滑台伺服参数时变性和不确定性影响的要求,提高了系统的动态特性性能。
[Abstract]:The CNC cross slide table is a typical CNC machine tool. It is driven by the XY motor screw system, which is perpendicular to each other in two spaces. It has the common characteristics of the feed drive system of the general NC machine tool. Therefore, the in-depth study of the dynamic performance of the two-axis coupling drive system of the CNC cross slide table is of great significance for the follow-up development of the NC machine tool. In general CNC machine tools, the spatial structure is not always a single axis. In the process of high-speed operation, the interaction between structures can not be ignored because of the large spatial time-variation between moving components in the process of multi-axis structure. In this paper, the mechanical and electrical system of numerical control crosshairs is studied in theory, experiment and computer simulation according to the structural coupling characteristics of numerical control crosshairs. The main works are as follows: (1) first, the structural coupling action of the XY axis of the numerical control cross slide table is presented. By using the finite element-lumped parameter method, the dynamic mathematical modeling of the motor screw system and the sliding block system of the moving parts is carried out separately, and the determination of the parameters of the model is analyzed theoretically. Secondly, according to the dynamic mathematical model, the virtual prototype model of XY biaxial coupling of NC cross slide table is established by using 3D modeling software and Adams dynamic analysis software, and the inherent modal analysis of the virtual prototype is completed in Adams software. The inherent modal characteristics of the cross slide platform system are discussed when the Y-axis table is coupled at different positions. (2) the control model of the two-axis feed drive system of the numerical control cross-slide table is established by using Matalab/Simulink. Combined with Adams, the electro-mechanical simulation of the cross slide platform under the action of biaxial coupling is completed, and its different trajectories and velocities are analyzed and studied. (3) the dynamic characteristic experiment platform of numerical control cross slide table is established, and the high resolution servo motor code plate, linear grating ruler and PCI high-speed pulse counting card are used. The vibration signals of dynamic characteristics under biaxial coupling are measured, and the measured signals are calculated numerically and analyzed in time domain by using Matalb software. (4) based on the theory of BP neural network, the vibration signals are analyzed and analyzed in time domain. (4) based on the theory of BP neural network, the vibration signals of dynamic characteristics are measured. The simulation running script of PID parameter adaptive adjustment and optimization for servo control system of NC crosshairs platform is established by using Matalab. The simulation results show that the control precision of NC cross slide table is controlled by BP neural network PID adaptive adjustment. The stability and anti-jamming performance have been greatly improved. It is proved that the BP neural network algorithm can meet the requirements of time-varying and uncertain servo parameters of CNC crosshairs and improve the dynamic performance of the system.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TG659

【参考文献】