基于主成分分析方法的并联机构参数优化

摘要/Abstract

摘要： 以DELTA并联机构为研究对象，求解该机构的全域条件数、全域各向同性系数、全域最小奇异值指标数据，建立多维全域性能指标数据。基于主成分分析(PCA)法，对上述多维全域性能指标数据进行降维、有效融合，去除指标间的冗余信息，减小目标函数对权重系数的人为依赖性，从而确定目标函数。通过对不同结构参数样本进行PCA分析，选出基于综合全域性能最优的结构尺寸。将PCA优化结果与各性能指标值分布图进行比较分析，最优和最差样本均与各性能指标分布图的最优和最差性能区域相对应，从而验证PCA方法在并联机构参数优化时的有效性。研究结果为多目标下并联机构的优化提供了一种新思路。

关键词: 参数优化, 并联机构, 主成分分析, 全域性能指标

Abstract: A DELTA parallel mechanism was treated as the research object, the global conditioning index, the global isotropic and the global minimum singular value of the mechanism were solved to build multi-dimensional global performance indices. Based on PCA, the multi-dimensional global performance indices were dimensionally reduced, effectively integrated. Redundant data dimensions among kinematic indicators and target function dependence of weight coefficient were reduced to obtain more objective target function. The different structure parameter samples were analyzed by PCA thereby optimal structure dimensions were obtained based on the comprehensive global performance. The optimization results of PCA and the distribution of the performance index values were compared and analyzed, the best and the worst samples corresponded to the best and the worst performance areas of distribution map of every global performance index respectively, and then the effectiveness of applying PCA in parameter optimization of parallel mechanisms was verified. These contents of the research provide a new idea for multi-target optimization of the parallel mechanisms.

Key words: parameter optimization, parallel mechanism, principal component analysis（PCA）, global performance index

中图分类号:

TH112

曾达幸;王娟娟;樊明洲;岳晓博;侯雨雷. 基于主成分分析方法的并联机构参数优化[J]. 中国机械工程.

ZENG Daxing;WANG Juanjuan;FAN Mingzhou;YUE Xiaobo;HOU Yulei. Parameter Optimization of Parallel Mechanisms Based on PCA[J]. China Mechanical Engineering.

参考文献

[1]黄田, 曾宪菁, 倪雁冰,等. 球面并联操作机的尺度综合方法[J]. 自然科学进展, 2000, 10(8): 751-756.
HUANG Tian, ZENG Xianjing, NI Yanbing, et al. The Dimensional Synthesis of Spherical Parallel Manipulator[J]. Progress in Natural Science, 2000, 10(8): 751-756.
[2]刘辛军, 吴泽启, 张立杰. 基于力传递性能的一种平面并联机器人的优化设计[J]. 清华大学学报（自然科学版）, 2008, 48(11): 1751-1754.
LIU Xinjun, WU Zeqi, ZHANG Lijie. Optimum Design of a Planar Parallel Robot Based on Force Transmissibility[J]. Journal of Tsinghua University (Science and Techonlogy),2008, 48(11):1751-1754.
[3]赵亮, 雎刚, 吕剑虹. 一种改进的遗传多目标优化算法及其应用[J]. 中国电机工程学报, 2008, 28(2): 96-102.
ZHAO Liang, JU Gang, LYU Jianhong. An Improved Genetic Algorithm in Multi-objective Optimization and Its Application[J]. Proceedings of the CSEE, 2008, 28(2): 96-102.
[4]侯媛彬, 薛斐, 郑茂全, 等. 基于多层编码的遗传-粒子群融合算法流水线优化控制[J]. 机械工程学报, 2015, 51(9): 159-164．
HOU Yuanbin, XUE Fei, ZHENG Maoquan, et al. Optimization of Flow-shop Control by Using Genetic-particle Swarm Algorithm of Multilayer-coded[J]. Journal of Mechanical Engineering, 2015, 51(9): 159-164.
[5]SCHMIDHUBER J. Discovering Neural Nets with Low Kolmogorow Complexity and High Generalization Capability[J]. Neural Networks the Official Journal of the International Neural Network Society, 1997, 10(5): 857-873.
[6]张渝, 胡启国, 张甫仁. 基于粒子群算法的冷挤压模具结构参数优化设计[J]. 中国机械工程, 2010, 21(8): 1003-1007.
ZHANG Yu, HU Qiguo, ZHANG Furen. Optimal Design of Structural Parameters of Cold Extrusion Dies Based on Partical Swarm Optimization[J]. China Mechanical Engineering, 2010, 21(8): 1003-1007.
[7]KURODA M,MORI Y, LIZUKA M, et al. Acceleration of the Alternating Least Squares Algorithm for Principal Components Analysis[J]. Computational Statistics and Data Analysis, 2011, 55(1): 143-153.
[8]黄姗, 丁坚平, 徐洋. 基于主成分分析法对开阳县岩溶地下水水质的评价[J]. 环保科技, 2014, 20(6): 48-52.
HUANG Shan, DING Jianping, XU Yang. The Evaluation of Karst Groundwater Quality Based on Principal Component Analysis in Kaiyang[J]. Environmental Protection and Technology, 2014, 20(6): 48-52.
[9]刘臣辉, 吕信红, 范海燕. 主成分分析法用于环境质量评价的探讨[J]. 环境科学与管理, 2011, 36(3): 183-186.
LIU Chenhui, LYU Xinhong, FAN Haiyan. Study of Applying Principal Component Analysis to Environmental Quality Assessment[J]. Environmental Science and Management, 2011, 36(3): 183-186.
[10]周桐. 基于PCA的人脸识别系统的设计与实现[D]. 哈尔滨：哈尔滨工业大学, 2007.
ZHOU Tong. Face Recognition System Design and Implementation Based on Principal Component Analysis[D]. Harbin:Harbin Institute of Technology, 2007.
[11]王婧瑶, 许勇, 曹本希. 基于主成分分析的超声人脸识别算法研究[J]. 计算机工程与设计, 2013, 34(8): 2867-2871.
WANG Jingyao, XU Yong, CAO Benxi. Ultrasonic Face Recognition Algorithm Based on Principal Component Analysis[J]. Computer Engineering and Design, 2013, 34(8): 2867-2871.
[12]古莹奎, 杨子茜, 朱繁泷. 基于主成分分析的齿轮故障特征融合分析[J]. 中国机械工程, 2015, 26(11): 1532-1537.
GU Yingkui, YANG Zixi, ZHU Fanlong. Gearbox Fault Feature Fusion Based on Principal Component Analysis[J]. China Mechanical Engineering, 2015, 26(11): 1532-1537.
[13]古莹奎, 承姿辛, 朱繁泷. 基于主成分分析和支持向量机的滚动轴承故障特征融合分析[J]. 中国机械工程, 2015, 26(20): 2778-2783.
GU Yingkui, CHENG Zixin, ZHU Fanlong. Rolling Bearing Fault Fusion Based on PCA and SVM [J]. China Mechanical Engineering, 2015, 26(20): 2778-2783.
[14]NGUYEN V H, GOLINVAL J C. Fault Detection Based on Kernel Principal Component Analysis[J]. Engineering Structures, 2010, 32(11): 3683-3691.
[15]梁胜杰, 张志华, 崔立林. 主成分分析法和核主成分分析法在机械噪声数据降维中的应用比较[J]. 中国机械工程, 2011, 22(1): 80-83.
LIANG Shengjie, ZHANG Zhihua, CUI Lilin. Comparison between PCA and KPCA Method in Dimensional Reduction of Mechanical Noise Data[J]. China Mechanical Engineering, 2011, 22(1): 80-83.
[16]赵京, 李立明, 尚红,等. 基于主成分分析法的机械臂运动灵活性性能综合评价[J]. 机械工程学报, 2014, 50(13): 9-15.
ZHAO Jing, LI Liming, SHANG Hong, et al. Comprehensive Evaluation of Robotic Global Kinematic Dexterity Performance Based on Principal Component Analysis[J]. Journal of Mechanical Engineering, 2014, 50(13): 9-15.
[17]赵京, 李立明. 基于主成分分析法和核主成分分析法的机器人全域性能综合评价[J]. 北京工业大学学报, 2014, 40(12): 1763-1769.
ZHAO Jing, LI Liming. Comprehensive Evaluation of Robotic Global Performance Based on Principal Component Analysis and Kernel Principal Component Analysis[J]. Journal of Beijing University of Technology, 2014, 40(12): 1763-1769.
[18]孙志娟, 赵京, 赵辛. PCA与KPCA在并联机构综合性能评价中的应用比较[J]. 制造业自动化, 2014, 36(2): 66-71.
SUN Zhijuan, ZHAO Jing, ZHAO Xin. Comparision between PCA and KPCA Method in Comprehensive Performance Evaluation of Parallel Mechanism[J]. Manufacturing Automation, 2014, 36(2): 66-71.
[19]王文博, 陈秀芝. 多指标综合评价中主成分分析和因子分析方法的比较[J]. 统计与信息论坛, 2006, 21(5): 19-22.
WANG Wenbo, CHEN Xiuzhi. Comparison of Principal Component Analysis with Factor Analysis in Comprehensive Multi-indicators Scoring[J]. Statistics and Information Forum. 2006, 21(5): 19-22.
[20]秦寿康. 综合评价原理与应用[M]. 北京:电子工业出版社, 2003:135-142.
QIN Shoukang. The Theory and Application of Comprehensive Assessment[M]. Beijing: Publishing House of Electronics Industry, 2003:135-142.
[21]林海明, 杜子芳. 主成分分析综合评价应该注意的问题[J]. 统计研究, 2013, 30(8): 25-31.
LIN Haiming, DU Zifang. Some Problem in Comprehensive Evaluation in the Principal Component Analysis[J]. Statistical Research, 2013, 30(8): 25-31.