基于切向—轮廓控制的凸轮磨削算法
发布时间：2018-12-18 21:05
【摘要】：作为机械设备的关键零部件,凸轮的磨削精度直接影响着发动机等机构的整体性能。近年来随着数控磨削技术的不断发展,凸轮磨削精度得到了大幅度提高。不同于传统靠模仿形的加工方式,目前多采用两轴联动系统精密加工。通过控制工件旋转轴与砂轮进给轴的协同运动,完成凸轮磨削。由于凸轮的磨削精度最终反映在凸轮的轮廓误差上,可以将磨削过程的精度问题看作是对轮廓误差的控制问题,因此相比于通过减小各单轴跟踪误差以间接减小轮廓误差的方法,直接减小轮廓误差的方法效果更佳。本文引入切向-轮廓控制算法,以减小凸轮轮廓误差、提高系统磨削精度为目标,对该控制算法提出一系列的改进,从而实现对轮廓误差直接有效的控制。主要研究内容如下:1、磨削前的准备工作。为了维持凸轮轮廓的原有特性,消除升程数据可能带有的误差,对厂家提供的原始升程数据进行光顺处理和三次样条插值。在此基础上通过反转法和联动数学模型,求得系统的输入序列值。2、引入切向-轮廓控制算法。为了促进切向-轮廓控制算法在凸轮磨削系统中的应用,选取合适的转换矩阵,实现坐标系间的转换。并提出了位置跟踪补偿算法,与切向-轮廓控制器、位置环控制器共同构成整体轮廓误差控制算法,实现轮廓误差的进一步减小。并由仿真实验验证上述算法的有效性。3、设计切向-轮廓控制器的参数优化算法。在整体轮廓误差控制算法中,切向-轮廓控制器的参数选取直接影响凸轮的磨削精度。为了能够有效减小轮廓误差,采用差分进化算法,对切向控制器与轮廓控制器的增益系数进行优化。以当前磨削过程中凸轮轮廓误差的最大绝对值为适应度函数,由增益系数构成的二维矢量作为进化的种群个体,经迭代计算得出个体最优解。仿真实验表明,具有最优解的系统其磨削精度得到了提高。4、设计切向-轮廓控制器的参数自适应控制算法。由于凸轮的轮廓曲率复杂多变,若要减小凸轮的轮廓误差,还可以从参数自适应控制方面展开研究。采用BP(Back Propagation)神经网络算法,令网络输出层的输出分别对应切向控制器与轮廓控制器的增益系数,通过实时修正各层权系数,实现控制器增益系数根据不同磨削点和轮廓误差的自适应变化,并由仿真实验验证了其有效性。另外,对比分析参数优化算法与参数自适应控制算法的优缺点,在时效性方面参数优化算法更省时,在轮廓误差控制方面参数自适应控制算法更具优势。
[Abstract]:As the key parts of mechanical equipment, the grinding accuracy of cam has a direct impact on the overall performance of the engine and other mechanisms. In recent years, with the development of NC grinding technology, the precision of cam grinding has been greatly improved. Different from the traditional way of machining by imitation, two-axis linkage system is widely used in precision machining at present. The cam grinding is accomplished by controlling the coordinated movement between the workpiece rotary shaft and the grinding wheel feed shaft. Because the precision of cam grinding is reflected in the profile error of cam, the precision of grinding process can be regarded as the control problem of contour error. Therefore, the method of direct reduction of contour error is better than the method of reducing the error of contour indirectly by reducing the tracking error of each single axis. In this paper, a tangential contour control algorithm is introduced to reduce cam profile error and improve grinding accuracy of the system. A series of improvements are put forward to realize direct and effective control of contour error. The main research contents are as follows: 1. Preparation work before grinding. In order to maintain the original characteristics of the cam profile and eliminate the possible errors in the lift data, the original lift data provided by the manufacturer are faired and cubic spline interpolation is carried out. On this basis, the input sequence value of the system is obtained by inversion method and linkage mathematical model. 2, and the tangential contour control algorithm is introduced. In order to promote the application of tangential contour control algorithm in cam grinding system, a suitable transformation matrix is selected to realize the transformation between coordinates. A position tracking compensation algorithm is proposed, which, together with the tangent-contour controller and the position loop controller, constitutes the overall contour error control algorithm, which can further reduce the contour error. The effectiveness of the above algorithm is verified by simulation experiments. 3. The parameter optimization algorithm of tangential-contour controller is designed. In the overall contour error control algorithm, the selection of the parameters of the tangential contour controller directly affects the grinding accuracy of the cam. In order to reduce the contour error effectively, the gain coefficients of tangential controller and contour controller are optimized by differential evolution algorithm. The maximum absolute value of cam profile error in the current grinding process is taken as fitness function, and the two-dimensional vector composed of gain coefficients is used as the evolutionary population individual, and the individual optimal solution is obtained by iterative calculation. The simulation results show that the grinding accuracy of the system with optimal solution is improved. 4. A parameter adaptive control algorithm for tangent-contour controller is designed. Because the contour curvature of the cam is complex and changeable, if we want to reduce the contour error of the cam, we can also study the parameter adaptive control. Using the BP (Back Propagation) neural network algorithm, the output of the network output layer corresponds to the gain coefficients of the tangential controller and the contour controller respectively, and the weight coefficients of each layer are corrected in real time. The gain coefficient of the controller is adaptively changed according to different grinding points and contour errors, and its effectiveness is verified by simulation experiments. In addition, by comparing the advantages and disadvantages of parameter optimization algorithm and parameter adaptive control algorithm, parameter optimization algorithm is more time-saving in terms of timeliness, and parameter adaptive control algorithm has more advantages in contour error control.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG580.6;TP273
[Abstract]:As the key parts of mechanical equipment, the grinding accuracy of cam has a direct impact on the overall performance of the engine and other mechanisms. In recent years, with the development of NC grinding technology, the precision of cam grinding has been greatly improved. Different from the traditional way of machining by imitation, two-axis linkage system is widely used in precision machining at present. The cam grinding is accomplished by controlling the coordinated movement between the workpiece rotary shaft and the grinding wheel feed shaft. Because the precision of cam grinding is reflected in the profile error of cam, the precision of grinding process can be regarded as the control problem of contour error. Therefore, the method of direct reduction of contour error is better than the method of reducing the error of contour indirectly by reducing the tracking error of each single axis. In this paper, a tangential contour control algorithm is introduced to reduce cam profile error and improve grinding accuracy of the system. A series of improvements are put forward to realize direct and effective control of contour error. The main research contents are as follows: 1. Preparation work before grinding. In order to maintain the original characteristics of the cam profile and eliminate the possible errors in the lift data, the original lift data provided by the manufacturer are faired and cubic spline interpolation is carried out. On this basis, the input sequence value of the system is obtained by inversion method and linkage mathematical model. 2, and the tangential contour control algorithm is introduced. In order to promote the application of tangential contour control algorithm in cam grinding system, a suitable transformation matrix is selected to realize the transformation between coordinates. A position tracking compensation algorithm is proposed, which, together with the tangent-contour controller and the position loop controller, constitutes the overall contour error control algorithm, which can further reduce the contour error. The effectiveness of the above algorithm is verified by simulation experiments. 3. The parameter optimization algorithm of tangential-contour controller is designed. In the overall contour error control algorithm, the selection of the parameters of the tangential contour controller directly affects the grinding accuracy of the cam. In order to reduce the contour error effectively, the gain coefficients of tangential controller and contour controller are optimized by differential evolution algorithm. The maximum absolute value of cam profile error in the current grinding process is taken as fitness function, and the two-dimensional vector composed of gain coefficients is used as the evolutionary population individual, and the individual optimal solution is obtained by iterative calculation. The simulation results show that the grinding accuracy of the system with optimal solution is improved. 4. A parameter adaptive control algorithm for tangent-contour controller is designed. Because the contour curvature of the cam is complex and changeable, if we want to reduce the contour error of the cam, we can also study the parameter adaptive control. Using the BP (Back Propagation) neural network algorithm, the output of the network output layer corresponds to the gain coefficients of the tangential controller and the contour controller respectively, and the weight coefficients of each layer are corrected in real time. The gain coefficient of the controller is adaptively changed according to different grinding points and contour errors, and its effectiveness is verified by simulation experiments. In addition, by comparing the advantages and disadvantages of parameter optimization algorithm and parameter adaptive control algorithm, parameter optimization algorithm is more time-saving in terms of timeliness, and parameter adaptive control algorithm has more advantages in contour error control.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG580.6;TP273
【相似文献】
相关期刊论文 前10条
1 刘兴富;邹常荣;;确定凸轮桃尖位置方法的探讨[J];计量技术;1979年01期
2 贾正锋;梁海顺;刘锋;李军宁;王华蓉;;高精度凸轮轮廓反求的数据处理及分析[J];纺织器材;2009年06期
3 刘兴富;刘瑞玲;;凸轮测量容易出现的问题及解决的途径[J];制造技术与机床;2012年07期
4 金其文;;高精度凸轮磨削[J];磨床与磨削;1981年01期
5 林章东;尹效志;;凸轮磨削加工的转速控制[J];汽车技术;1986年07期
6 李智;;组合盘状凸轮制造工艺[J];食品机械;1986年02期
7 吴冠林;;也谈车削凸轮工艺分析[J];机械制造;1989年08期
8 朱伟成;;双凸轮的设计制造研究[J];汽车技术;1989年01期
9 刘兴富;;凸轮检测起始转角的求解方法[J];宇航计测技术;1993年02期
10 姜e，
本文编号：2386499
本文链接：https://www.wllwen.com/kejilunwen/jiagonggongyi/2386499.html
