基于结构自适应滤波方法的非线性系统辨识

doi:10.11772/j.issn.1001-9081.2019111996

计算机应用 ›› 2020, Vol. 40 ›› Issue (8): 2319-2326.DOI: 10.11772/j.issn.1001-9081.2019111996

基于结构自适应滤波方法的非线性系统辨识

冯子凯, 陈立家, 刘名果, 袁蒙恩

河南大学物理与电子学院, 河南开封 475000

收稿日期:2019-11-25 修回日期:2020-04-09 出版日期:2020-08-10 发布日期:2020-05-19
通讯作者: 冯子凯(1994-),男,河南安阳人,硕士研究生,主要研究方向:演化算法、信号处理、人工智能;chenlijia_just@163.com
作者简介:陈立家(1979-),男,河南开封人,副教授,博士,主要研究方向:数字信号处理、演化算法、人工智能;刘名果(1984-),男,河南巩义人,副教授,博士,主要研究方向:数字信号处理、人工智能;袁蒙恩(1992-),女,河南商丘人,硕士研究生,主要研究方向:演化算法、信号处理、人工智能。
基金资助:
河南省自然科学基金资助项目（15A510018）；河南大学种子基金资助项目（ZZJJ20140037）。

Nonlinear systems identification based on structural adaptive filtering method

FENG Zikai, CHEN Lijia, LIU Mingguo, YUAN Meng’en

School of Physics and Electronics, Henan University, Kaifeng Henan 475000, China

Received:2019-11-25 Revised:2020-04-09 Online:2020-08-10 Published:2020-05-19
Supported by:
This work is partially supported by the Natural Science Foundation of Henan Province (15A510018), the Seed Fund of Henan University (ZZJJ20140037).

摘要/Abstract

摘要： 针对非线性系统辨识中定结构参数辨识局限性高和辨识率低的问题，将结构自适应引入辨识的优化，提出一种基于子系统的结构自适应滤波（SSAF）方法。该方法的模型由若干子系统级联而成，每一个子系统均为线性-非线性混合结构。子系统的线性部分是一个一阶或二阶可选的无限脉冲响应滤波器（IIR），非线性部分则是一个静态的非线性函数。初始化中，子系统的参数随机产生，生成的若干子系统按照设定的连接规则进行随机连接，而不含反馈的连接机制确保了非线性系统的有效性。采用一种自适应多精英引导的复合差分进化（AMECoDEs）算法用于自适应模型循环优化，直至找到最优的结构和参数，即全局最优。实验结果表明，SSAF方法在非线性测试函数以及真实数据集上的表现优异，辨识率高且收敛性好，与聚焦时滞递归神经网络（FTLRNN）相比，它所用参数的个数仅为FTLRNN的1/10，且适应值精度提高了7%，验证了所提方法的有效性。

关键词: 结构自适应滤波, 非线性系统辨识, 子系统, 线性-非线性

Abstract: In order to solve the problems of high identification limitation and low identification rate in nonlinear system identification with fixed structure and parameters, a Subsystem-based Structural Adaptive Filtering (SSAF) method for nonlinear system identification was proposed with introducing structural adaptation into the optimization of identification. Multiple subsystems with linear-nonlinear hybrid structure were cascaded to form the model for this method. The linear part is a 1-order or 2-order Infinite Impulse Response (IIR) digital filter with uncertain parameters, and the nonlinear part is a static nonlinear function. In the initial stage, the parameters of the subsystems were randomly generated, and the generated subsystems were connected randomly according to the set connection rules, and the effectiveness of the nonlinear system was guaranteed by the connection mechanism with no feedback branches. An Adaptive Multiple-Elites-guided Composite Differential Evolution with a shift mechanism(AMECoDEs) algorithm was used for loop optimization of the adaptive model until the optimal structure and parameters were found, that is, the global optimal. The simulation results show that AMECoDEs performs well on nonlinear test functions and real data sets with high identification rate and good convergence rate. Compared with the Focused Time Lagged Recurrent Neural Network (FTLRNN), the number of parameters used in SSAF is reduced to 1/10, and the accuracy of fitness is improved by 7%, which proves the effectiveness of the proposed method.

Key words: structure adaptive filtering, nonlinear systems identification, subsystem, linear-nonlinear

中图分类号:

TP273

冯子凯, 陈立家, 刘名果, 袁蒙恩. 基于结构自适应滤波方法的非线性系统辨识[J]. 计算机应用, 2020, 40(8): 2319-2326.

FENG Zikai, CHEN Lijia, LIU Mingguo, YUAN Meng’en. Nonlinear systems identification based on structural adaptive filtering method[J]. Journal of Computer Applications, 2020, 40(8): 2319-2326.

参考文献

[1] SOKOLOY V F. Problems of adaptive optimal control of discrete-time systems under bounded disturbance and linear performance indexes[J]. Automation and Remote Control, 2018, 79(6):1086-1099.
[2] NA J, CHEN A S, HERRMANN G, et al. Vehicle engine torque estimation via unknown input observer and adaptive parameter estimation[J]. IEEE Transactions on Vehicular Technology, 2018, 67(1):409-422.
[3] CHEN L, LIU M, DAI Z, et al. Evolution based structure design for low-pass IIR digital filter 10 with fault tolerance[J]. Engineering Letters, 2018, 26(3):340-347.
[4] SERSOUR L, DJAMAH T, BETTAYEB M. Nonlinear system identification of fractional Wiener models[J]. Nonlinear Dynamics, 2018, 92(4):1493-1505.
[5] KAZEMI M, AREFI M M. A fast iterative recursive least squares algorithm for Wiener model identification of highly nonlinear systems[J]. ISA Transactions, 2017, 67:382-388.
[6] GUO J, WANG L Y, YIN G, et al. Identification of Wiener systems with quantized inputs and binary-valued output observations[J]. Automatica, 2017, 78:280-286.
[7] TANG Y, QIAO L, GUAN X. Identification of Wiener model using step signals and particle swarm optimization[J]. Expert Systems with Applications, 2010, 37(4):3398-3404.
[8] 白晶,毛志忠,蒲铁成. 多变量Hammerstein-Winner模型的参数辨识[J]. 东北大学学报(自然科学版), 2018, 39(1):6-10. (BAI J, MAO Z Z, PU T C. Parameter identification of mutivatiate Hammerstein-Winner model[J]. Journal of Northeastern University (Natural Science), 2018, 39(1):6-10.)
[9] 李冬伍,任雪梅,吕晓华. 含间隙非线性的Wiener-Hammerstein系统复合补偿控制[J]. 控制理论与应用, 2016, 33(1):54-61. (LI D W, REN X M, LYU X H. Combined compensation control for Wiener-Hammerstein systems with backlash nonlinearities[J]. Control Theory and Applications, 2016, 33(1):54-61.)
[10] VANBEYLEN L. A fractional approach to identify Wiener-Hammerstein systems[J]. Automatica, 2014, 50(3):903-909.
[11] SCHOUKENS M, VANDERSTEEN G, ROLAIN Y, et al. Fast identification of Wiener-Hammerstein systems using discrete optimization[J]. Electronics Letters, 2014, 50(25):1942-1944.
[12] NAITALI A, GIRI F. Wiener-Hammerstein system identification-an evolutionary approach[J]. International Journal of Systems Science, 2016, 47(1):45-61.
[13] 何伟铭,宋小奇,甘屹,等. 传感器校正的优化灰色神经网络建模方法研究[J]. 仪器仪表学报, 2014, 35(3):504-512. (HE W M, SONG X Q, GAN Y, et al. Research on optimized grey neural network modeling method for sensor calibration[J]. Chinese Journal of Scientific Instrument, 2014, 35(3):504-512.)
[14] CARRILLO-SANTOS C A, SECK-TUOH-MORA J C, HERNÁNDEZ-ROMERO N, et al. Wavenet identification of dynamical systems by a modified PSO algorithm[J]. Engineering Applications of Artificial Intelligence, 2018, 73:1-9.
[15] GAO G, LIU F, SAN H, et al. Hybrid optimal kinematic parameter identification for an industrial robot based on BPNN-PSO[J]. Complexity, 2018, 2018:No.4258676.
[16] GUAN S, LI Z. Normalised spline adaptive filtering algorithm for nonlinear system identification[J]. Neural Processing Letters, 2017, 46(2):595-607.
[17] DUDUL S V. Identification of a liquid saturated steam heat exchanger using focused time lagged recurrent neural network model[J]. IETE Journal of Research, 2007, 53(1):69-82.
[18] CUI L, LIA G, ZHU Z, et al. Adaptive multiple-elites-guided composite differential evolution algorithm with a shift mechanism[J]. Information Sciences, 2018, 422:122-143.
[19] ZADEH M R, AMIN S, KHALILI D, et al. Daily outflow prediction by multi layer perceptron with logistic sigmoid and tangent sigmoid activation functions[J]. Water Resources Management, 2010, 24(11):2673-2688.
[20] SUBUDHI B, JENA D. A differential evolution based neural network approach to nonlinear system identification[J]. Applied Soft Computing, 2011, 11(1):861-871.
[21] KUMAR R, SRIVASTAVA S, GUPTA J R P, et al. Comparative study of neural networks for dynamic nonlinear systems identification[J]. Soft Computing, 2019, 23(1):101-114.

基于结构自适应滤波方法的非线性系统辨识

Nonlinear systems identification based on structural adaptive filtering method

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	牛春彦, 夏克文, 张江楠, 贺紫平. 基于云量子花朵授粉的极限学习机算法[J]. 计算机应用, 2020, 40(6): 1627-1632.
[2]	冯帅栋, 陈立家, 刘名果. 基于随机结构的无乘法无限冲激响应数字滤波器设计方法[J]. 计算机应用, 2018, 38(9): 2621-2625.
[3]	钟丽, 刘彦, 余思洋, 谢中. 嵌入式系统芯片中SM2算法软硬件协同设计与实现[J]. 计算机应用, 2015, 35(5): 1412-1416.
[4]	王畅谢永华袁复兴. 基于结构化粒子模型的云可视化应用[J]. 计算机应用, 2013, 33(11): 2013-01.
[5]	刘云龙韩星海唐述宏. 基于趋近律方法的Delta算子滑模变结构控制系统[J]. 计算机应用, 2013, 33(08): 2397-2400.
[6]	陈修亮梁英杰郭福亮. 基于CUDA粒子系统的烟花仿真[J]. 计算机应用, 2013, 33(07): 2059-2062.
[7]	徐元金曾亮陈世龙. 一种真实感雪场景模拟方法[J]. 计算机应用, 2013, 33(05): 1428-1431.
[8]	杨森孟晨王成. 电子系统健康状态监测数据优化算法[J]. 计算机应用, 2012, 32(10): 2927-2930.
[9]	王运盛雷航. 面向综合模块化航电系统的驻留应用开发平台设计[J]. 计算机应用, 2012, 32(03): 861-863.
[10]	林镜华于东李鸿彬李凌南. 基于IP多媒体子系统的IPTV业务控制服务器集群负载均衡机制[J]. 计算机应用, 2010, 30(9): 2488-2493.
[11]	罗为君林亚平. XNA中基于素材管道的粒子系统设计与实现[J]. 计算机应用, 2010, 30(3): 589-592.
[12]	杨春毛轩易禾朱福全. 基于粒子系统的化学实验现象动态模拟[J]. 计算机应用, 2010, 30(1): 150-152.
[13]	文治中刘直芳李纲粱威. 基于GPU粒子系统的大规模场景高效雨雪实时模拟[J]. 计算机应用, 2010, 30(05): 1398-1401.
[14]	王剑锋万明秀陈焱. 基于新分布域和叠加纹理的飘雪模拟[J]. 计算机应用, 2009, 29(12): 3381-3384.
[15]	吴银霞陈雷霆何明耘. OpenGL中基于粒子系统雷达扫描实时模拟[J]. 计算机应用, 2009, 29(1): 258-260.