针对非线性系统Volterra泛函级数模型，结合混沌优化策略和种群多样性控制思想，提出了一种改进粒子群算法，并应用于Volterra模型参数的辨识，将非线性系统的辨识问题转化为高维参数空间上的优化问题。利用混沌序列增加初始种群的多样性，通过构建动态子群以进行协作寻优，且各子群采用不同的参数自适应调整策略，并定义算法收敛性测度以对精英粒子进行合理的混沌变异，避免了算法早熟收敛，提高了算法的寻优速度和寻优精度。仿真实验中，将该方法与基于标准粒子群算法、遗传算法、量子粒子群算法的Volterra模型参数辨识方法相比较，验证了该辨识方法的有效性和鲁棒性。
Abstract
By combining Particle Swarm Optimization (PSO) with chaotic optimization strategy and control thought of population diversity, an Improved Particle Swarm Optimization Algorithm (IPSO) is proposed for parameters identification of nonlinear Volterra series model. The basic idea of the method is that the problem of nonlinear system identification is changed into an optimization problem in high-dimensional parameter space. The chaotic optimization strategy is employed to increase the diversity of the initial population. By building dynamic subgroups, the best value is found relying on the collaboration of sub groups. The different adaptive adjustment strategy about the control parameters of IPSO is used in subgroups. The convergence measure of IPSO is defined to finish the chaotic mutation operation reasonably. So the premature convergence is avoided and the speed and accuracy of IPSO is improved. In the simulation experiment, the proposed method was compared with Volterra model identification method based on standard Particle Swarm Optimization algorithm, Genetic Algorithm, Quantum-behaved Particle Swarm Optimization, the effectiveness and robustness of the proposed method are verified by simulation results.
关键词
改进粒子群算法 /
非线性系统 /
Volterra级数 /
系统辨识
{{custom_keyword}} /
Key words
Improved Particle Swarm Optimization Algorithm (IPSO) /
nonlinear system /
Volterra series /
system identification
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 曹建福，韩崇昭，方洋旺. 非线性系统理论及应用[M]. 西安：西安交通大学出版社，2006.
[2] 张家良，曹建福，高峰. 大型装备传动系统非线性频谱特征提取与故障诊断[J]. 控制与决策，2012，27（1）：135- 138.
ZHANG Jia-liang，CAO Jian-fu，GAO Feng. Feature extraction and fault diagnosis of large-scale equipment transmission system based on nonlinear frequency spectrum[J]. Control and Decision，2012，27（1）：135- 138
[3] 张华君. 基于递推批量最小二乘的Volterra 级数辨识方法[J]. 小型微型计算机, 2004, 25（12）: 2282- 2285.
Zhang Hua-jun. Volterra Series Identif ication Method Based on Recursive Least Square Algorithm[J]. Mini-Micro Systems, 2004, 25（12）: 2282- 2285.
[4] 孔祥玉，韩崇昭，马红光，等. 基于Volterra 级数的全解耦RLS自适应辨识算法[J]. 系统仿真学报，2004,16（4）：807-809.
KONG Xiang-yu，HAN Chong-zhao，MA Hong-guang，et al. Fully Decoupled RLS Adaptive Identification Algorithm Based on Volterra Series[J]. Journal of System Simulation，2004,16（4）：807-809.
[5] 孔祥玉，韩崇昭，马红光，等. 一种总体最小二乘算法及在Volterra滤波器中的应用[J]. 西安交通大学学报，2004,38（4）：339-342.
Kong Xiang-yu，Han Chong-zhao，Ma Hong-guang，et al. Total Least Square Algorithm and Its Application to Volterra Filter[J]. Journal of XI’AN Jiao Tong University，2004,38（4）：339-342.
[6] 唐浩，屈梁生，温广瑞. 基于Volterra级数的转子故障诊断研究[J]. 中国机械工程，2009，20 (4):447-454.
Tang Hao，Qu Liang-sheng，Wen Guang-rui. Fault Diagnosis for Rotor System Based on Volterra Series[J]. China Mechanical Engineering，2009，20 (4):447-454.
[7] Hazem M. Abbas，Mohamed M. Bayoumi. Volterra system identification using adaptive genetic algorithms[J]. Applied Soft Computing，5(2004)： 75–86.
[8] 李志农，唐高松，肖尧先，等. 基于自适应蚁群优化的Volterra核辨识算法研究[J]. 振动与冲击，2011，30 (10):35-38.
LI Zhi-nong，TANG Gao-song，XIAO Yao-xian，et al. Volterra series identification method based on adaptive ant colony optimization[J]. Journal of Vibration and Hock，2011，30 (10):35-38.
[9] 李志农，蒋静，陈金刚，等. 基于量子粒子群优化的Volterra核辨识算法研究[J]. 振动与冲击，2013，32 (3)：60-63，74.
LI Zhi-nong，JIANG Jing，CHEN Jin-gang，et al. Volterra series identification method based on quantum particle swarm optimization[J]. Journal of Vibration and Hock，2013，32 (3)：60-63，74.
[10] 李志农，蒋静，冯辅周，等. 基于量子粒子群优化Volterra时域核辨识的隐Markov模型识别方法[J]. 仪器仪表学报，2011，32 (12):2693-2698.
LI Zhi-nong，JIANG Jing，Feng Fu-zhou，et al. Hidden Markov model recognition method based on Volterra kernel identified with particle swarm optimization [J]. Chinese Journal of Scientific Instrument，2011，32 (12):2693-2698.
[11] Kennedy J, Eberhart R C. Particle swarm optimization[J]. Proceedings of ICNN’95- IEEE International Conference on Neural Networks. IEEE, Piscataway, NJ, USA, 1995: 1942-1948.
[12] Eberhart R，Kennedy J.A new optimizer using particle swarm theory[C]//Proc of 6th International Symposium on Micro Machine and Human Science，Nagoya，Japan.Piscataway NJ：IEEE Service Center，1995：39-43.
[13] 毛鹏轩，肖扬. 粒子群算法在稳定时滞系统设计中的应用[J]. 计算机应用研究，2012，29（1）：214-216．
MAO Peng-xuan，XIAO Yang. Application of PSO algorithm for stable time-delay system design[J]. Application Research of Computers，2012，29（1）：214-216．
[14] 崔志华，曾建潮. 微粒群优化算法[M]. 北京：科学出版社，2011.
[15] 张浩，张铁男，沈继红等. Tent 混沌粒子群算法及其在结构优化决策中的应用[J]. 控制与决策，2008，23（8）：857-862．
ZHANG Hao，ZHANG Tie-nan，SHEN Ji-hong，et al. Research on decision-makings of structure optimization based on improved Tent PSO[J]. Control and Decision，2008，23（8）：857-862．
[16] 魏玉琴，戴永寿，张亚南等. 基于Tent 映射的自适应混沌嵌入式粒子群算法[J]. 计算机工程与应用，2013，49（10）：45-49．
WEI Yu-qin，DAI Yong-shou，ZHAN Ya-nan，et al. Adaptive chaotic embedded Particle Swarm Optimization algorithm based on Tent mapping[J]. Computer Engineering and Applications，2013，49（10）：45-49．
[17] Cui Z H，Cai X J, Zeng J C, et al. Particle swarm optimization with FUSS and RWS for high dimensional functions. Applied Mathematics and Computation, 2008,205（1）：98-108.
[18] Arumugam M S，Rao M V C. On the improved performances of the particle swarm optimization algorithms with adaptive parameters, cross-over operators and root mean square(RMS) variants for computing optimal control of a class ov hybrid system. Applied Soft Computing, 2008,8（1）：324-336.
[19] Bajpai P，Singh S N. Fuzzy adaptive particle swarm optimization for bidding strategy in uniform price spot market. IEEE Transactions on Power System, 2007,22（4）：2152-2160.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1761 KB)
