Optimization of Processing Parameters in Grinding and Polishing Coupling Neural Networks with Genetic Algorithms
-
摘要: 针对机器人磨抛系统工艺参数的自主选择与优化问题, 提出一种基于神经网络与遗传算法的磨抛工艺参数优化方法, 采用基于人工神经网络的工件表面粗糙度预测模型解决各工艺参数间复杂的非线性问题, 结合粗糙度预测模型与磨抛效率公式, 通过遗传算法对各工艺参数进行全局寻优解决加工质量和效率的双目标优化问题并得到最优工艺参数组合。在满足加工质量要求的前提下, 加工效率提高了近三分之一, 证明此工艺参数优化方法是可行有效的。Abstract: Aiming at the problem of autonomous selection and optimization of process parameters of robot grinding and polishing systems, an optimization method of the processing parameters in the grinding and polishing based on neural networks and genetic algorithms is proposed. Adopting artificial neural networks based workpiece surface roughness prediction model to solve complex non-linear problems among the processing parameters. Combining the roughness prediction model with the grinding and polishing efficiency formula, by using genetic algorithms to globally optimize each processing parameter to solve the dual-objective optimization problem of processing quality and efficiency and finally obtain the optimal processing parameter combination. On the premise of meeting the requirements of processing quality, the processing efficiency has been improved by nearly one third, which proves that this processing parameter optimization method is feasible and effective.
-
Key words:
- processing parameters /
- neural networks /
- genetic algorithms /
- parameter optimization
-
表 1 磨抛工艺参数表
方法 步骤 G/# P/kPa ns/(r·min-1) va/(mm·min-1) N/次 未优化 1 150 20 1 000 120 6 2 180 20 1 000 120 4 ANN-GA 1 150 15 1 000 180 7 表 2 磨抛工艺参数表
方法 磨抛后Ra 预测Ra ΔRa/μm 总时间/s 未优化 0.768 0.665 0.103 171 0.192 0.176 0.016 ANN-GA 0.216 0.195 0.021 116 -
[1] 卢燕明. 智能制造与工业机器人[J]. 金属加工(冷加工), 2013(20): 6 doi: 10.3969/j.issn.1674-1641.2013.20.003LU Y M. Intelligent manufacturing and industrial robots[J]. Metal Processing (Cold Processing), 2013(20): 6 (in Chinese) doi: 10.3969/j.issn.1674-1641.2013.20.003 [2] 杨洁. 超精密研磨抛光工艺优化及智能控制系统的研究[D]. 南京: 南京航空航天大学, 2006YANG J. Research on the process optimization and intelligent control system for ultra-precision lapping and polishing[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2006 (in Chinese) [3] KIM N H, CHOI M H, KIM S Y, et al. Design of experiment (DOE) method considering interaction effect of process parameters for optimization of copper chemical mechanical polishing (CMP) process[J]. Microelectronic Engineering, 2006, 83(3): 506-512 doi: 10.1016/j.mee.2005.11.016 [4] LIN T R. The use of reliability in the Taguchi method for the optimisation of the polishing ceramic gauge block[J]. The International Journal of Advanced Manufacturing Technology, 2003, 22(3): 237-242 doi: 10.1007/s00170-002-1465-1 [5] LIN T R, CHUI H C, HUANG M F. Optimising removal rate and reliability of polishing of ceramic blocks using a combination of Taguchi and grey methods[J]. Materials Science and Technology, 2004, 20(12): 1649-1654 doi: 10.1179/026708304X6077 [6] 李振国, 田风杰, 陈洪玉, 等. 复杂曲面机器人自动化研抛加工试验人动加[J]. 制造技术与机床, 2016(8): 151-155LI Z G, TIAN F J, CHENG H Y, et al. Experiment of robot automatic polishing for curved surfaces[J]. Manufacturing Technology & Machine Tool, 2016(8): 151-155 (in Chinese) [7] 袁楚明, 张雷, 陈幼平, 等. 模具曲面机器人智能抛光系统的研究[J]. 高技术通讯, 2001, 11(9): 76-80 doi: 10.3321/j.issn:1002-0470.2001.09.017YUAN C M, ZHANG L, CHEN Y P, et al. Study on a robotic intelligent polishing system for mold curved surface[J]. High Technology Letters, 2001, 11(9): 76-80 (in Chinese) doi: 10.3321/j.issn:1002-0470.2001.09.017 [8] LEE H S, PARK M S, KIM M T, et al. Systematic finishing of dies and moulds[J]. International Journal of Machine Tools and Manufacture, 2006, 46(9): 1027-1034 doi: 10.1016/j.ijmachtools.2005.07.049 [9] WU B H, WANG J J. A neuro-fuzzy approach to generating mold/die polishing sequences[J]. Journal of Materials Processing Technology, 2009, 209(7): 3241-3250 doi: 10.1016/j.jmatprotec.2008.07.031 [10] 王瑞芳, 徐方. 机器人研磨抛光工艺研究与实现[J]. 新技术新工艺, 2008(9): 19-22 doi: 10.3969/j.issn.1003-5311.2008.09.007WANG R F, XU F. Research and realization of robot grinding and polishing processing technology[J]. New Technology & New Process, 2008(9): 19-22 (in Chinese) doi: 10.3969/j.issn.1003-5311.2008.09.007 [11] TSAI M J, HUANG J F. Efficient automatic polishing process with a new compliant abrasive tool[J]. The International Journal of Advanced Manufacturing Technology, 2006, 30(9): 817-827 doi: 10.1007/s00170-005-0126-6 [12] AHN J H, SHEN Y F, KIM H Y, et al. Development of a sensor information integrated expert system for optimizing die polishing[J]. Robotics and Computer-Integrated Manufacturing, 2011, 17(4): 269-276 http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.40.8431 [13] 雷英杰, 张善文, 李续武, 等. MATLAB遗传算法工具箱及应用[M]. 西安: 西安电子科技大学出版社, 2005LEI Y J, ZHANG S W, LI X W, et al. MATLAB genetic algorithm toolbox and application[J]. Xi'an: Xidian University Press, 2005 (in Chinese) [14] 宋倩, 郭宇, 王发麟. 基于人工神经网络的线缆敷设质量综合评判[J]. 计算机工程与应用, 2018, 54(14): 256-263, 270 doi: 10.3778/j.issn.1002-8331.1705-0182SONG Q, GUO Y, WANG F L. Comprehensive evaluation method of cable harness wiring quality based on artificial neural network[J]. Computer Engineering and Applications, 2018, 54(14): 256-263, 270 (in Chinese) doi: 10.3778/j.issn.1002-8331.1705-0182 [15] 温海骏, 孟小玲, 许向川, 等. 基于神经网络和遗传算法的激光熔覆工艺参数多目标优化[J]. 应用激光, 2019, 39(5): 734-740 http://www.cnki.com.cn/Article/CJFDTotal-YYJG201905003.htmWEN H J, MENG X L, XU X C, et al. Multi-objective optimization of laser cladding process parameters based on neural network and genetic algorithm[J]. Applied Laser, 2019, 39(5): 734-740 (in Chinese) http://www.cnki.com.cn/Article/CJFDTotal-YYJG201905003.htm [16] 金爱娟, 杨晓洁, 王居正, 等. 基于智能算法的感应电机多目标优化设计[J]. 农业装备与车辆工程, 2020, 58(1): 44-48, 70 doi: 10.3969/j.issn.1673-3142.2020.01.011JIN A J, YANG X J, WANG J Z, et al. Multi-objective optimization design of induction motor based on intelligence algorithm[J]. Agricultural Equipment & Vehicle Engineering, 2020, 58(1): 44-48, 70 (in Chinese) doi: 10.3969/j.issn.1673-3142.2020.01.011 [17] 顾徐波, 马元巍, 王德忠, 等. 基于改进BP神经网络和遗传算法的多风扇风洞风速廓面快速形成方法[J]. 原子能科学技术, 2018, 52(10): 1889-1896 doi: 10.7538/yzk.2018.youxian.0102GU X B, MA Y W, WANG D Z, et al. Improved BP neural network and genetic algorithm based fast formation method of wind speed profile in multi-fan wind tunnel[J]. Atomic Energy Science and Technology, 2018, 52(10): 1889-1896 (in Chinese) doi: 10.7538/yzk.2018.youxian.0102