改进粒子群优化Takagi-Sugeno模糊径向基函数神经网络的非线性系统建模

doi:10.11772/j.issn.1001-9081.2014.05.1341

计算机应用 ›› 2014, Vol. 34 ›› Issue (5): 1341-1344.DOI: 10.11772/j.issn.1001-9081.2014.05.1341

改进粒子群优化Takagi-Sugeno模糊径向基函数神经网络的非线性系统建模

李丽娜¹,甘晓晔²,徐攀峰¹,马俊¹

1. 辽宁大学物理学院，沈阳 110036
2. 辽宁科技学院机械工程学院，辽宁本溪 117004

收稿日期:2013-11-04 修回日期:2013-12-31 出版日期:2014-05-01 发布日期:2014-05-30
通讯作者: 李丽娜
作者简介:李丽娜（1973-），女（满），辽宁本溪人，副教授，博士，主要研究方向：自动化测量与控制技术；甘晓晔（1963-），女，辽宁本溪人，教授，硕士，主要研究方向：信号分析与处理技术；徐攀峰（1978-），女，辽宁鞍山人，实验师，主要研究方向：检测系统与传感器技术；马俊（1988-），女，山东济南人，硕士研究生，主要研究方向：虚拟仪器及自动化检测技术。
基金资助:
辽宁省自然科学基金项目;辽宁省教育厅科学技术研究项目

Nonlinear system modeling based on Takagi-Sugeno fuzzy radial basis function neural network optimized by improved particle swarm optimization

LI Lina¹,GAN Xiaoye²,XU Panfeng¹,MA Jun¹

1. College of Physics, Liaoning University, Shenyang Liaoning 110036, China;
2. College of Mechanical Engineering, Liaoning Institute of Science and Technology, Benxi Liaoning 117004, China

Received:2013-11-04 Revised:2013-12-31 Online:2014-05-01 Published:2014-05-30
Contact: LI Lina

摘要/Abstract

摘要：

针对复杂非线性系统建模的难点问题，提出了一种基于改进的粒子群优化算法（PSO)优化的T-S模糊径向基函数（RBF）神经网络的新型系统建模算法。该算法将T-S模糊模型良好的可解释性及RBF神经网络的自学习能力相结合，构成T-S模糊RBF神经网络用于系统建模，并采用动态调整惯性权重的改进的PSO算法结合递推最小二乘算法实现网络参数的优化调整。首先，利用所提算法进行了非线性多维函数的逼近仿真，仿真结果均方差（MSE）为0.00017，绝对值误差不大于0.04，逼近精度较高；又将该算法用于建立动态流量软测量模型，并进行了相关的实验研究，动态流量测量结果平均绝对误差小于0.15L/min，相对误差为1.97%，基本满足测量要求，并优于已有算法。上述仿真及实验研究结果表明，所提算法对于复杂非线性系统具有较高的建模精度和良好的自适应性。

Abstract:

For the difficulty of complex non-linear system modeling, a new system modeling algorithm based on the Takagi-Sugeno (T-S) Fuzzy Radial Basis Function (RBF) neural network optimized by improved Particle Swarm Optimization (PSO) algorithm was proposed. In this algorithm, the good interpretability of T-S fuzzy model and the self-learning ability of RBF neural network were combined together to form a T-S fuzzy RBF neural network for system modeling, and the network parameters were optimized by the improved PSO algorithm with dynamic adjustment of the inertia weight combined with recursive least square method. Firstly, the proposed algorithm was used to do the approximation simulation of a non-linear multi-dimensional function, the Mean Square Error (MSE) of the approximation model was 0.00017, the absolute error was not greater than 0.04, which shows higher approximation precision; the proposed algorithm was also used to build a dynamic flow soft measurement model and to finish related experimental study, the average absolute error of the dynamic flow measurement results was less than 0.15L/min, the relative error is 1.97%, these results meet measurement requirements well and are better than the results of the existing algorithms. The above simulation results and experimental results show that the proposed algorithm is of high modeling precision and good adaptability for complex non-linear system.

中图分类号:

TP216

李丽娜甘晓晔徐攀峰马俊. 改进粒子群优化Takagi-Sugeno模糊径向基函数神经网络的非线性系统建模[J]. 计算机应用, 2014, 34(5): 1341-1344.

LI Lina GAN Xiaoye XU Panfeng MA Jun. Nonlinear system modeling based on Takagi-Sugeno fuzzy radial basis function neural network optimized by improved particle swarm optimization[J]. Journal of Computer Applications, 2014, 34(5): 1341-1344.

参考文献

［1］WANG F, XING K, XU X. Study on modeling method of nonlinear dynamic system ［J］. Computer Engineering and Applications, 2009,45(10):7-10.(王峰，邢科义，徐小平.非线性动态系统建模方法研究［J］.计算机工程与应用，2009,45(10):7-10.
［2］TAKAGI T, SUGENO M. Fuzzy identification of systems and its applications to modeling and control ［J］. IEEE Transactions on Systems, Man and Cybernetics, 1985,SMC-15(1):116-132.
［3］DU H, ZHANG N. Application of evolving Takagi-Sugeno fuzzy model to nonlinear system identification ［J］. Applied Soft Computing, 2008,8(1):676-686.
［4］HAN M, SUN Y, FAN Y. An improved fuzzy neural network based on T-S model ［J］. Expert Systems with Applications, 2008,34(4):2905-2920.
［5］GAO S, GAO X, WANG J. Soft-sensor modeling of polymerizing process based on multi T-S fuzzy neural network optimized with harmony search algorithm ［J］. Chinese Journal of Scientific Instrument, 2011,32(3):495-500.(高淑芝,高宪文,王介生.基于和声搜索优化多T-S模糊神经网络的聚合釜过程软测量建模［J］.仪器仪表学报,2011,32(3):495-500.)
［6］ZHANG Y. Improved T-S fuzzy neural network applied in soft sensing of chemical industry ［J］. Journal of Electronic Measurement and Instrument, 2010,24(6):585-588.(张颖.改进的T-S模糊神经网络在化工软测量中的应用［J］.电子测量与仪器学报,2010,24(6):585-588.)
［7］LYU M, SUN H, TANG Y. Application of RBF neural network optimized by BFGS-GA in soft measurement system of dynamic flow ［J］. Journal of Chinese Computer Systems, 2009, 30(9):1831-1833. (吕梦雅,孙宏岩,唐勇.BFGS-GA优化的RBF网络在动态流量软测量中的应用［J］.小型微型计算机系统,2009,30(9):1831-1833.)
［8］LYU M, YAN J, TANG Y, et al. Optimal algorithm of RBF center to improve the real-time characteristic of soft measurement system of dynamic flow ［J］. Chinese Journal of Scientific Instrument, 2009, 30(12):2619-2623. (吕梦雅,闫晶,唐勇,等.提高动态流量软测量实时性的RBF 中心优化算法［J］.仪器仪表学报,2009,30(12):2619-2623.)
［9］BAI Q. Analysis of particle swarm optimization algorithm ［J］. Journal of Computer and Information Science, 2010,3(1):180-184.
［10］LONG W, LIANG X, DONG S, et al. Particle swarm optimization with dynamic change of inertia weights ［J］. Journal Of Computer Applications, 2009,29(8):2240-2242.(龙文,梁昔明,董淑华,等.动态调整惯性权重的粒子群优化算法［J］.计算机应用,2009,29(8):2240-2242.)
［11］GENG H, HUANG Y, GAO J, et al. A self-guided particle swarm optimization with independent dynamic inertia weights setting on each particle ［J］. Journal of Applied Mathematics and Information Sciences, 2013,7(2):545-552.
［12］JIANG W, LEI Y, ZHANG Q, et al. Research on dynamical flow measuring method based on neural network soft measurement ［J］. Journal of Fluid Power Transmission and Control, 2007,25(6):25-30.(姜万录,雷亚飞,张齐生,等.基于神经网络软测量的动态流量测量方法研究［J］.流体传动与控制,2007,25(6):25-30.)

[1]	孙婷, 马磊. 巡检机器人中指针式仪表示数的自动识别方法[J]. 计算机应用, 2019, 39(1): 287-291.
[2]	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.

改进粒子群优化Takagi-Sugeno模糊径向基函数神经网络的非线性系统建模

Nonlinear system modeling based on Takagi-Sugeno fuzzy radial basis function neural network optimized by improved particle swarm optimization

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics