工程科学学报，第39卷，第4期：611618,2017年4月 Chinese Journal of Engineering,Vol.39,No.4:611-618,April 2017 D0:10.13374/j.issn2095-9389.2017.04.017:htp:/journals..ustb.edu.cm 函数型数据分析与优化极限学习机结合的弹药传输 机械臂参数辨识 赵抢抢，侯保林四 南京理工大学机械工程学院，南京210094 ☒通信作者，E-mail:houbl(@njust.cdu.cn 摘要为实现弹药传输机械臂中不可测参数的辨识，建立了机械臂的虚拟样机，并将其作为样本数据的来源：考虑到样本 数据的连续性和平滑特性，使用函数型数据分析和函数型主成分分析对样本数据进行了特征提取，并利用提取的特征参数和 待辨识参数作为训练样本对极限学习机(ELM)进行了训练.为提高极限学习机的辨识精度和泛化能力，利用粒子群算法对 极限学习机的输入层与隐含层的连接权值和隐含层节点的阈值进行了优化.最后，分别利用仿真数据与测试数据对此方法 进行了验证，仿真数据的辨识结果表明，优化后的极限学习机具有更高的辨识精度和泛化能力：同时，通过对比将测试数据的 辨识结果代入模型中进行仿真得到的支臂角速度与测试角速度，验证了此方法的可行性和有效性. 关键词参数辨识：函数型数据分析：极限学习机：粒子群优化：弹药传输机械臂 分类号TP241:TH113 Parameter identification of a shell transfer arm using FDA and optimized ELM ZHAO Qiang-qiang,HOU Bao-in School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China Corresponding author,E-mail:houbl@njust.edu.cn ABSTRACT To identify the unmeasurable parameters of a shell transfer arm,a virtual prototype of the shell transfer arm was built, and the built virtual prototype is regard as the source of the sample data.Considering the continuity and smoothness properties of the sample data,features of the curves were extracted by functional data analysis and functional principal component analysis,and the fea- tures and unknown parameters were used to train the extreme leaming machine (ELM).At the meantime,the weight connecting the input layer and hidden layer and the threshold of the hidden nodes were optimized by particle swarm optimization (PSO)to improve the identification accuracy and generalization performance of ELM.At last,the presented method was verified by simulation data and test data.The identification results of the simulation data show that the optimized ELM has higher identification accuracy and better generalization performance.Also,the presented method is proved to be feasible and effective by comparing the real angular velocity and the angular velocity from the virtual prototype with respect to the test data identification results. KEY WORDS parameter identification:functional data analysis;extreme learning machine;particle swarm optimization:shell transfer arm 弹药传输机械臂是大口径火炮自动装填系统的一 膛后返回原位”.因系统复杂且工作环境恶劣，弹药 个重要部件，主要用于接收弹仓内被推弹器推送出来 传输机械臂始终存在定位精度超差的情况，严重降低 的弹丸，再将该弹丸传送至输弹线上由输弹机输弹入 了整个弹药自动装填系统的可靠性，亟待进行改进设 收稿日期：201607-29 基金项目：国家自然科学基金资助项目(51175266)：国家高技术研究发展计划资助项目(6132490102)工程科学学报，第 39 卷，第 4 期: 611--618，2017 年 4 月 Chinese Journal of Engineering，Vol． 39，No． 4: 611--618，April 2017 DOI: 10． 13374 /j． issn2095--9389． 2017． 04． 017; http: / /journals． ustb． edu． cn 函数型数据分析与优化极限学习机结合的弹药传输 机械臂参数辨识 赵抢抢，侯保林 南京理工大学机械工程学院，南京 210094 通信作者，E-mail: houbl@ njust． edu． cn 摘 要 为实现弹药传输机械臂中不可测参数的辨识，建立了机械臂的虚拟样机，并将其作为样本数据的来源; 考虑到样本 数据的连续性和平滑特性，使用函数型数据分析和函数型主成分分析对样本数据进行了特征提取，并利用提取的特征参数和 待辨识参数作为训练样本对极限学习机( ELM) 进行了训练． 为提高极限学习机的辨识精度和泛化能力，利用粒子群算法对 极限学习机的输入层与隐含层的连接权值和隐含层节点的阈值进行了优化． 最后，分别利用仿真数据与测试数据对此方法 进行了验证，仿真数据的辨识结果表明，优化后的极限学习机具有更高的辨识精度和泛化能力; 同时，通过对比将测试数据的 辨识结果代入模型中进行仿真得到的支臂角速度与测试角速度，验证了此方法的可行性和有效性． 关键词 参数辨识; 函数型数据分析; 极限学习机; 粒子群优化; 弹药传输机械臂 分类号 TP241; TH113 Parameter identification of a shell transfer arm using FDA and optimized ELM ZHAO Qiang-qiang，HOU Bao-lin School of Mechanical Engineering，Nanjing University of Science and Technology，Nanjing 210094，China Corresponding author，E-mail: houbl@ njust． edu． cn ABSTＲACT To identify the unmeasurable parameters of a shell transfer arm，a virtual prototype of the shell transfer arm was built， and the built virtual prototype is regard as the source of the sample data． Considering the continuity and smoothness properties of the sample data，features of the curves were extracted by functional data analysis and functional principal component analysis，and the fea￾tures and unknown parameters were used to train the extreme learning machine ( ELM) ． At the meantime，the weight connecting the input layer and hidden layer and the threshold of the hidden nodes were optimized by particle swarm optimization ( PSO) to improve the identification accuracy and generalization performance of ELM． At last，the presented method was verified by simulation data and test data． The identification results of the simulation data show that the optimized ELM has higher identification accuracy and better generalization performance． Also，the presented method is proved to be feasible and effective by comparing the real angular velocity and the angular velocity from the virtual prototype with respect to the test data identification results． KEY WOＲDS parameter identification; functional data analysis; extreme learning machine; particle swarm optimization; shell transfer arm 收稿日期: 2016--07--29 基金项目: 国家自然科学基金资助项目( 51175266) ; 国家高技术研究发展计划资助项目( 6132490102) 弹药传输机械臂是大口径火炮自动装填系统的一 个重要部件，主要用于接收弹仓内被推弹器推送出来 的弹丸，再将该弹丸传送至输弹线上由输弹机输弹入 膛后返回原位［1］． 因系统复杂且工作环境恶劣，弹药 传输机械臂始终存在定位精度超差的情况，严重降低 了整个弹药自动装填系统的可靠性，亟待进行改进设