基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机

交通运输系统工程与信息 ›› 2006, Vol. 6 ›› Issue (6): 113-119 .

基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机

吕刚，范瑜

北京交通大学电气工程学院，北京100044

收稿日期:2006-05-21 修回日期:1900-01-01 出版日期:2006-12-20 发布日期:2006-12-20

Permanent-Magnet Linear Synchronous Motor Model Using NDEKF Neural Network on Hession

LV Gang，FAN Yu

Electric Engineering Institute，Beijing Jiaotong University，Beijing 100044，China

Received:2006-05-21 Revised:1900-01-01 Online:2006-12-20 Published:2006-12-20

摘要/Abstract

摘要： 永磁直线同步电动机模型的建立对研究其稳态特性、动态特性和控制策略都是非常重要的。首先，将带外部输入的非线性自回归模型展成多项式形式，然后在此基础上用残差分析法导出真实的阶次所满足的条件。为了克服神经网络结构依靠人工试凑的不足，使用基于Hession矩阵的修剪法来优化其结构。考虑到BP算法的一些固有缺点，使用NDEKF(基于节点的解耦扩展Kalman滤波器算法)来训练网络。实验证明，网络的输出结果与试验样机(永磁直线同步电动机驱动的垂直运输系统)的实际输出十分接近；同时将NDEKF与BP算法进行对比，NDEKF算法具有收敛较快、泛化能力强、不易陷入局部极小等特点。

关键词: 神经网络, 永磁直线同步电动机垂直运输系统, 辨识, Hession矩阵, NDEKF

Abstract:

The modeling of permanent-magnet linear synchronous motor is very important to the control and the static and dynamic characters analysis of the system．First，the nonlinear autoregressive with exogenous inputs model is expanded into the polynomial function，then the condition which true ranks satisfy is presented by using residual signal analysis．In order to overcome the shortages that the design of network structure is depended on one’s own personal experience，Hession-based network pruning is used to get the optimization network structures. Some shortages of BP(back-propagation algorithm) are considered，so NDEKF((node-decoupled extend/Kalman filter) is applied to train networks．The experiment results show that the hybrid neural networks of the nonlinear autoregressive with exogenous inputs can identified object’s(a vertical transport system driven by permanent-magnet linear synchronous motor) ranks precisely，and the output of networks is very close to the experimental result．In the experiments，the perform ance of NDEKF is often superior to that of BP，while requiring significantly fewer presentations of training data than BP and less over training time than that of BP．

Key words: neural networks, permanent-magnet linear synchronous motor, identification, Hession, NDEKF

吕刚,范瑜. 基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机基于Hession矩阵优化的NDEKF神经网络辨识永磁直线同步电机[J]. 交通运输系统工程与信息, 2006, 6(6): 113-119 .

LV Gang,FAN Yu. Permanent-Magnet Linear Synchronous Motor Model Using NDEKF Neural Network on Hession[J]. Journal of Transportation Systems Engineering and Information Technology, 2006, 6(6): 113-119 .

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract17297.shtml

http://www.tseit.org.cn/CN/Y2006/V6/I6/113

参考文献

[1] Chong，Kil To，Pados，Alexander G．Comparison of the traditionaland the neural networks approaches in a stochastic nonlinear system identification[J]．Proceedings of the American Control Conference，1997，2：1074—1075．
[2] Amini F，Chen H M，Qi G Z，Yang J C S Generalized neural network based model for structural dynamic identification，analytical and experimental studies[J]．Intelligent Information Systems，1997，ITS’97．Proceedings，8-10 Dec．1997，138-142．
[3] Mastoroeostas，Paris A，Theocharis，John B．A recurrent fuzzy-neural model for dynamic system identification[J]． IEEE Transactions on Systems，Man，and Cybemetics，Part B：Cybernetics，April，2002，32(2)：176-190．
[4] Yardizadeh A，Khorasani K．Identification of a turbogenerator system using adaptive time delay neural networks[J]．IEEE Conference on Control Applications-Proceedings，1998，1：502-506．
[5] Yamamoto，Yoshihiro，Nikiforuk，Peter N．Learning algorithm for recurrent neural networks and its application to nonlinear identification[J]．Proceedings of the IEEE International Symposium on Computer-Aided Control System Design，1999：551-556．
[6] Yu Wen，Poznynk，Alexander S，Sanchez，Edgar N．Dynamic multilayer neural networksfor nonlinear system online identification[J]．IEEE International Symposium on Intelligent Control-Proceedings. 2000：25-30.
[7] Selmic Rastko R，Lewis，Frank L．Multimodel neural networks identification and failure detection of nonlinear systems[J]．Proceedings of the IEEE Conference on Decision and Control，2001，4：3128-3133．
[8] Ibnkahla．Mohamed．Nonlinear system identification using neural networks trained with natural gradient descent[J]．Eurasip Journal on Applied Signal Processing，2003，12(1)：1229-1237．
[9] Nouri，Khaled，Dhaouadi，Rached，Braiek，Benhadj Naceur．Identification of a nonlinear dynamic systems using recurrent multilayer neural networks[J]．Proceedings of the IEEE Intemational Conference on Systems，Man and Cybernetics，2002，5：306-3l0．
[10] Alessandri A，Sanguineti，M，Maggiore M．Optimized feedforward neural networks for on-line identification of nonlinearmodels[J]．Proceedings of the IEEE Conference on Decision and Control，20, 2, 2：1751-1756．
[11] Patra Jagdish C，Pal，Ranendra N，Chatterji，B N，Panda，Ganapati．Identification of nonlinear dynamic systems using functional link artificial neural networks[J]．IEEE Transactions on Systems，Man ，and Cybernetics，Part B：Cybernetics，1999，29(2)：254-262．

[1]	赵建东, 陈溱, 焦彦利, 张凯丽, 韩明敏. 重点营运车辆的异常驾驶行为识别研究[J]. 交通运输系统工程与信息, 2022, 22(1): 282-291.
[2]	杨建喜，郁超顺，李韧，杜利芳，蒋仕新，王笛. 基于多周期组件时空神经网络的路网通行速度预测[J]. 交通运输系统工程与信息, 2021, 21(3): 112-119.
[3]	陈秀锋，吴阅晨，邴其春，聂蕊蕊. 基于类锚虚拟线圈的多流向车流量检测算法[J]. 交通运输系统工程与信息, 2021, 21(2): 51-57.
[4]	高天航，徐力，靳廉洁，葛彪. 考虑航艏向与数据变化差异的船舶轨迹预测[J]. 交通运输系统工程与信息, 2021, 21(1): 90-94.
[5]	张政，陈艳艳，梁天闻. 基于出行特征的用地类型推断方法研究[J]. 交通运输系统工程与信息, 2020, 20(5): 29-35.
[6]	赵聪，朱逸凡，李兴华，杜豫川. 动态管理模式下路侧停车泊位占有率预测方法[J]. 交通运输系统工程与信息, 2020, 20(5): 107-113.
[7]	张文胜，郝孜奇，朱冀军，杜甜添，郝会民. 基于改进灰狼算法优化BP 神经网络的短时交通流预测模型[J]. 交通运输系统工程与信息, 2020, 20(2): 196-203.
[8]	王志建，李达标，崔夏. 基于LSTM 神经网络的降雨天旅行时间预测研究[J]. 交通运输系统工程与信息, 2020, 20(1): 137-144.
[9]	朱顺应，王淑钰，邹禾，王红. 公交信息的乘客光环效应[J]. 交通运输系统工程与信息, 2019, 19(3): 169-175.
[10]	牛宏侠，张肇鑫，宁正，陈光武. 铁路轨道异物完整性检测与跟踪算法研究[J]. 交通运输系统工程与信息, 2019, 19(1): 45-54.
[11]	刘爽，支晓宇，陈绍宽，兴妍. 基于车站滞留风险的应急驻车点选址研究[J]. 交通运输系统工程与信息, 2018, 18(4): 110-115.
[12]	段宗涛，张凯，杨云，倪园园，SAURAB Bajgain. 基于深度CNN-LSTM-ResNet组合模型的出租车需求预测[J]. 交通运输系统工程与信息, 2018, 18(4): 215-223.
[13]	唐艳丽，蒋超，郑伯红，李茜铭. 基于多源数据融合的城市出租车载客出行特征研究——以岳阳市为例[J]. 交通运输系统工程与信息, 2018, 18(2): 45-51.
[14]	李桂毅，胡明华. 考虑航段相关性的航路拥挤态势多模型融合动态预测方法[J]. 交通运输系统工程与信息, 2018, 18(1): 215-222.
[15]	丁点点，孙磊，陈松. 机器学习—动力学耦合车辆跟驰模型[J]. 交通运输系统工程与信息, 2017, 17(6): 33-39.