A robotic belt grinding method for accurate control of the blade leading edge profile was proposed in this paper. The stress distribution in the contact area between the flexible abrasive tool and the leading edge of the blade was obtained by combining the semi-Hertzian contact theory and finite element simulation, and the material removal function was solved based on the Preston equation. The global material removal matrix was established by traversing the grinding depth of the control point. The resident time was set up to solve the nonlinear equations. Tikhonov regularization with a damping factor was used to eliminate the influence of a large sparse-ill-conditioned matrix on the fluctuation of solving accuracy, and then the desired dwell time was converted to the feed speed of the corresponding cutter point so as to generate the robot machining code. The grinding experiment results show that the robotic belt grinding method based on resident time control can achieve accurate machining of the leading edge profile of the blade within a given tolerance range, and the profile error can be controlled within 0.02 mm.
获取去除函数矩阵 R 前,由于叶片边缘曲率变化大,并不像平面一样可以直接判断控制点是否位于驻留点的接触范围之内,需要对此作出适当的处理. 加工路径上的接触判断图如图8所示.本文以任一驻留点为原点,求出当前驻留点的切平面和法矢,以指向下一个驻留点的方向为x轴,法矢和x方向的叉乘作为y轴,建立一个新的局部坐标系. 在新的坐标系下以x、y轴作为投影平面将所有控制点投影到该平面上,再通过式(8)的坐标变换矩阵 G 得到所有控制点在局部坐标系下的坐标.
R (m,n)矩阵的每一列表示的是路径上砂带在每一个驻留点时对该控制点的单位时间材料去除率. 为了保证加工精度,控制点的数量一般是远远大于驻留点的数量,并且由于接触轮的接触区域相对于工件较小,去除函数矩阵 R 是一个大型的稀疏矩阵. 由于该稀疏矩阵往往是个病态矩阵,向量行数m远大于列数n,因此无法获取精确解,众多学者从奇异值分解、最小二乘正交分解法(Least Squares QR-Factorization Method,LSQR)、二次规划寻优等多种角度进行研究和优化,在实际应用中,加工前缘并非需要精确解,大多数情况下,LSQR和Tikhonov正则化下的解即可满足加工要求.
Tikhonov正则化则是从提高解的稳定性和减少驻留总时间的角度出发,对去除函数矩阵 R 进行正则化处理,以优化驻留时间 t . 在去除函数矩阵 R 中引入一个数量阵 T,数量阵 T 引入的阻尼因子w表示驻留点密度与表面误差间的权重关系,理论上w的取值区间为[0,+∞),但w的值越小,取得非负驻留时间的可能性越小;而当w的值越大时,加工后的表面误差也就越大. 矩阵如式(10)所示.
数量阵 T 将去除函数矩阵 R 由m行n列扩展为(m+n)行n列的矩阵 RW,同时将加工余量列向量 b 补充0以扩充成(m+n)维,基于式(6)和改进后的 式(11),将最小二乘问题转换成亏秩矩阵的求解问题.
合理地选择阻尼因子w的值,以保证叶片前缘廓形高精度的同时,符合实际加工特点是Tikhonov正则化的关键所在. 去除函数矩阵 R 如图10所示.
LANR H, HUANGY, CHENG L, et al. Self-adaptive belt grinding technology and its experimental research on aero-engine blade[J].Aeronautical Manufacturing Technology,2018, 61(15):16-24.(in Chinese)
[5]
PRAMANIKA .Problems and solutions in machining of titanium alloys[J].The International Journal of Advanced Manufacturing Technology, 2014, 70(5): 919-928.
[6]
YANGZ Y, CHUY, XUX H,et al .Prediction and analysis of material removal characteristics for robotic belt grinding based on single spherical abrasive grain model[J].International Journal of Mechanical Sciences,2021,190:106005.
[7]
黄云, 黄智 .现代砂带磨削技术及工程应用[M].重庆: 重庆大学出版社, 2009:10-14.
[8]
HUANGY, HUANGZ .Modern abrasive belt grinding technology and its engineering application[M].Chongqing:Chongqing University Press,2009:10-14.(in Chinese)
[9]
GUOY M, RENJ X, LIANGY S .Aerodynamic performance-preserving construction method for a near-net-shape blade cross section[J].Aerospace Science and Technology,2023,138:108313.
[10]
WANN, SHENX G, CHANGZ Y,et al .A new localization theory of adaptive machining of near-net-shape blades[J].Chinese Journal of Aeronautics,2021,34(6):18-32.
[11]
WANGY D, XUZ Y, MENGD M,et al .Study on surface roughness of large size TiAl intermetallic blade in electrochemical machining[J].Journal of Manufacturing Processes,2022,76:1-10.
[12]
MUY L, WANGZ L, ZOUL,et al .A novel regional force control strategy based on seven-axis linkage grinding system to improve blade machining accuracy[J].Journal of Manufacturing Processes,2023,97:235-247.
[13]
SARMAR, DUTTAD .Tool path generation for NC grinding[J].International Journal of Machine Tools and Manufacture, 1998, 38(3): 177-195.
[14]
ZOUL, WANGT T, WANGC, et al .Modelling and analysis of the effect of nonlinear time-varying contact deformation on flexible precision grinding process[J].The International Journal of Advanced Manufacturing Technology, 2021, 115(1):77-89.
[15]
MOONEYM .A theory of large elastic deformation[J].Journal of Applied Physics, 1940, 11(9): 582-592.
[16]
PRESTONF W. Glass technology[J]. Journal of the Society of Glass Technology, 1927(11):277-281.
[17]
王文玺 .面向钢轨砂带打磨的材料去除机理及表面完整性研究[D].北京:北京交通大学,2019.
[18]
WANGW X. Research on material removal mechanisms and surface integrity for abrasive belt rail grinding[D]. Beijing:Beijing Jiaotong University, 2019.(in Chinese)
[19]
AYASSEJ B, CHOLLETH .Determination of the wheel rail contact patch in semi-Hertzian conditions[J].Vehicle System Dynamics, 2005, 43(3): 161-172.
[20]
MAX C, WANGP, XUJ M,et al .Assessment of non-Hertzian wheel-rail contact models for numerical simulation of rail damages in switch panel of railway turnout[J].Wear,2019,432/433:102912.
[21]
QUOSTX, SEBESM, EDDHAHAKA, et al .Assessment of a semi-Hertzian method for determination of wheel-rail contact patch[J].Vehicle System Dynamics, 2006, 44(10): 789-814.
[22]
ZHUB, ZENGJ, ZHANGD F,et al .A non-Hertzian wheel-rail contact model considering wheelset yaw and its application in wheel wear prediction[J].Wear, 2019, 432/433:202958.