高精度数控凸轮磨削的速度优化与轮廓误差补偿

发布时间：2018-05-31 18:42

本文选题：数控凸轮磨削 + 轮廓误差　；参考：《吉林大学》2017年博士论文

【摘要】：目前数控加工最大的问题就是过于依赖人工操作经验,加工效率低。本论文在吉林省科技厅项目(20150101031JC)“高精度数控凸轮磨削过程的速度优化与轮廓误差补偿”的资助下,以数控凸轮磨削为研究对象,为实现磨削加工的最优化和智能化为目标,展开以下几方面的研究工作:第一,建立了数控凸轮磨削系统中传动机构的数学模型。在机理分析和非线性因素分析的基础上,分别对X轴和C轴的传动机构建立了数学模型;并考虑到数控凸轮磨削中存在的重复问题,设计了基于扩张状态观测器的重复控制机械系统,提高了两个单轴伺服系统的跟踪精度,也为数控磨削加工的优化提供理论基础。第二,在单轴精度保证的前提下,设计了基于重复学习控制的交叉耦合控制系统。考虑到两个单轴即使精度得到提高,但仍然无法完全没有跟踪误差,且存在滞后等问题,对轮廓误差与伺服跟踪误差的几何关系进行分析,建立了基于同步滞后控制策略的轮廓误差模型。基于此模型,开展了实时补偿的交叉耦合控制。考虑到轮廓误差中的重复信号,设计了重复学习控制与PID控制相结合的交叉耦合控制器。数字仿真实验结果证明了利用此控制器可以对轮廓误差进行实时补偿,使得轮廓精度得以提高。第三,提出一种数控磨床凸轮磨削的速度优化控制方法。即使两个单轴的动态特性良好,高速数控凸轮磨削依然会引起两个轴的不同滞后,将导致轮廓误差不可避免的超过规定的范围带。为解决这个问题,本文提出一种新的速度优化方案。第一、导出了轮廓误差与两轴速度之间的数学关系,提出砂轮架相对于凸轮的相对速度概念,构造了有关相对速度的参数方程;第二、构造了一个有关于相对速度的指数型函数来优化两个单轴的速度使轮廓误差尽可能的减小。所提出的速度优化方案可以实现在凸轮升程斜率较大处,即凸轮敏感区域,速度降低,在升程变化缓慢处,提高速度。在保证效率的同时,提高磨削加工精度。与恒角速度磨削相对比,提出的速度优化算法可以有效提高轮廓精度。并通过实际对三种凸轮片的加工实验进一步来证明算法的有效性和可靠性。应用结果说明了轮廓误差可以有效的控制在0.012mm里。第四,提出了基于GCTC(generalized cycle-to-cycle)控制的双闭环控制方案。数控加工的凸轮轮廓只能在一个周期结束后才能得到真正的测量,即使过程中的测量可行,付出的代价也很大。针对这个问题,本文建立了基于GCTC的凸轮轮廓误差的双闭环控制。其中,内环是利用本地控制器以保证跟踪精度;外环利用GCTC反馈控制,将其作为内环的一个优化模块以实时更新给定值。加工系统的等价动态模型可以通过内环的基于扩张状态观测器的重复控制系统而获得。论文给出了GCTC反馈控制系统稳定的充要条件。通过对基于GCTC控制的数控磨削系统的仿真实验来验证了提出的控制方案的优越性。第五,进一步提出了基于测量误差的双层优化轮廓曲线误差控制方案。在内层应用基于重复控制的更新策略来提高可行性和收敛速度;从CTC控制和约束自适应中推导出的CTC误差修正作为外层的优化方案。两层目标函数一致,双层互相作用,内层作为外层的反馈,外层作为内层的指导。仿真实验结果证明了其有效性。
[Abstract]:At present, the biggest problem of NC machining is relying too much on manual operation experience and low processing efficiency. In this paper, under the support of "speed optimization and contour error compensation of high precision CNC cam grinding process" in the Jilin science and Technology Department (20150101031JC) project, CNC cam grinding is used as the research object, and the optimization of grinding process is realized. The following research work is carried out in the following aspects: first, the mathematical model of the transmission mechanism in the CNC cam grinding system is set up. On the basis of the mechanism analysis and the nonlinear factor analysis, the mathematical model of the transmission mechanism of the X axis and the C axis is established, and the repeated problems in the CNC cam grinding are taken into consideration. The repetitive control mechanical system based on the extended state observer improves the tracking accuracy of two single axis servo systems and provides a theoretical basis for the optimization of CNC grinding machining. (second) a cross coupling control system based on repeated learning control is designed under the premise of single axis precision guarantee. The accuracy of two single axes is considered. To improve, but still can not complete no tracking error, and there is a lag and other problems, the geometric relationship between the contour error and the servo tracking error is analyzed, and the contour error model based on the synchronization lag control strategy is established. Based on this model, the real-time compensation cross coupling control is carried out. A cross coupling controller combined with repetitive learning control and PID control is designed. The results of digital simulation prove that the contour error can be compensated in real time by using this controller, and the contour precision can be improved. Third, a speed optimization control method for cam grinding of CNC grinding machine is proposed. Even the dynamic characteristics of two single axes are presented. Well, the grinding of high-speed CNC cam will still cause different lag of the two axes, which will lead to the inevitable contour error exceeding the prescribed range. In order to solve this problem, a new speed optimization scheme is proposed in this paper. First, the mathematical relationship between the contour error and the velocity of the two axis is derived, and the phase of the grinding wheel frame relative to the cam is proposed. For the concept of speed, a parameter equation about relative velocity is constructed. Second, an exponential function with relative velocity is constructed to optimize the speed of two uniaxial forces to minimize the contour error. The proposed speed optimization scheme can be achieved in the higher slope of the cam lift, that is, the cam sensitive area, the speed decrease, and the increase of the speed. At the same time, the speed is improved. While the efficiency is ensured, the grinding precision is improved. Compared with the constant angular velocity grinding, the proposed speed optimization algorithm can effectively improve the contour precision. The efficiency and reliability of the algorithm are further proved by the processing experiments of the three kinds of cam sheets. The application results show the outline error. The difference can be effectively controlled in 0.012mm. Fourth, a double closed loop control scheme based on GCTC (generalized cycle-to-cycle) control is proposed. The cam profile of NC machining can only be measured at the end of one cycle. Even if the measurement is feasible in the process, the cost is very high. The double closed loop control of the cam profile error based on GCTC. The inner loop is used to ensure the tracking accuracy, and the outer ring uses the GCTC feedback control to use it as an optimization module of the inner loop to update the given value in real time. The equivalent dynamic model of the machining system can be controlled by the repeated control of the inner loop based on the extended state observer. The paper gives the necessary and sufficient conditions for the stability of the GCTC feedback control system. The superiority of the proposed control scheme is verified by the simulation experiment on the CNC grinding system based on GCTC control. Fifth, the double layer optimized contour error control scheme based on the measurement error is further proposed. The application based on the duplication is based on the internal layer. The updated strategy is controlled to improve the feasibility and convergence speed; the CTC error correction derived from the CTC control and the constraint adaptation is used as the optimization scheme for the outer layer. The two layer objective function is consistent, the double layer interaction, the inner layer as the outer layer feedback, the outer layer as the inner guide. The simulation results prove its effectiveness.
【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TG596

【相似文献】