基于SALR网络的双率采样非线性系统鲁棒辨识

doi:10.16182/j.issn1004731x.joss.24-0557

摘要/Abstract

摘要：

针对具有复杂非线性特性且测量输出含有异常值的双率采样非线性系统，提出了一种基于自连接相邻反馈循环储备池(self-join adjacent-feedback loop reservoir，SALR)网络的鲁棒辨识算法。使用SALR网络对目标系统的非线性特性进行描述，并在网络的储备池中注入小波神经元，以增强网络的记忆能力和非线性描述能力，将非线性系统辨识问题转化为网络输出权值矩阵的辨识问题；采用Huber损失构造准则函数，引入误差阈值提高随机梯度辨识算法对异常值的鲁棒性。为解决双率采样引起的输出数据缺失问题，在输出权值矩阵的递推辨识过程中，引入辅助模型辨识思想和交互估计理论，用网络的估计输出值代替不可测输出，并使用鲸鱼优化算法优化网络的超参数，进一步提高辨识精度。仿真结果验证了算法的有效性。

关键词: 双率采样, 鲁棒辨识, 非线性系统, 辅助模型, 循环神经网络, 鲸鱼优化算法

Abstract:

A robust identification algorithm based on the self-join adjacent-feedback loop reservoir (SALR) network was proposed for dual-rate sampled nonlinear systems with complex nonlinear characteristics and measurement outputs containing outliers. The SALR network was applied to describe the nonlinear characteristics of the target system, and wavelet neurons were injected into the reservoir to enhance its memory and nonlinear description capabilities. The identification problem of the nonlinear system was transformed into the identification problem of the network's output weight matrix. The Huber loss function was used to construct the criterion function, and an error threshold was introduced to improve the robustness of the stochastic gradient identification algorithm against outliers. To solve the problem of output data loss caused by dual-rate sampling, the concepts of auxiliary model identification and interaction estimation theory were introduced into the recursive identification process of the output weights, where the estimated outputs of the network were used to replace the unmeasured outputs. Moreover, the whale optimization algorithm was adopted to optimize the network's hyperparameters, further enhancing the identification accuracy. Numerical simulation results validate the effectiveness of the proposed algorithm.

Key words: dual-rate sampling, robust identification, nonlinear system, auxiliary model, recurrent neural network, whale optimization algorithm

中图分类号:

TP273

蒋文彬,曹余庆,谢莉等 . 基于SALR网络的双率采样非线性系统鲁棒辨识[J]. 系统仿真学报, 2025, 37(10): 2652-2661.

Jiang Wenbin,Cao Yuqing,Xie Li,et al . Robust Identification of Dual-rate Sampled Nonlinear Systems Based on SALR Network[J]. Journal of System Simulation, 2025, 37(10): 2652-2661.

图/表 10

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

参考文献 26

[1]	余伟, 梁恒辉, 罗映. 一种改进差分进化算法的分数阶系统辨识研究[J]. 系统仿真学报, 2021, 33(5): 1157-1166.
	Yu Wei, Liang Henghui, Luo Ying. An Improved Differential Evolution Algorithm for Fractional Order System Identification[J]. Journal of System Simulation, 2021, 33(5): 1157-1166.
[2]	Li Bo, Liang Wei, Yang Shengmei, et al. Automatic Identification of Modal Parameters for High Arch Dams Based on SSI Incorporating SSA and K-means Algorithm[J]. Applied Soft Computing, 2023, 138: 110201.
[3]	靳其兵, 杨瑞赓, 王珠, 等. 一类统一非线性特性的Hammerstein模型辨识方法研究[J]. 系统仿真学报, 2014, 26(12): 2887-2891.
	Jin Qibing, Yang Ruigeng, Wang Zhu, et al. Description of Unified Nonlinear Characteristics Hammerstein Model and Its Identification Method Research[J]. Journal of System Simulation, 2014, 26(12): 2887-2891.
[4]	Liu Yanjun, Ding Feng, Shi Yang. An Efficient Hierarchical Identification Method for General Dual-rate Sampled-data Systems[J]. Automatica, 2014, 50(3): 962-970.
[5]	陶洪峰, 刘艳, 杨慧中. 输出时滞双率采样系统的鲁棒迭代学习控制[J]. 控制与决策, 2017, 32(9): 1707-1713.
	Tao Hongfeng, Liu Yan, Yang Huizhong. Robust Iterative Learning Control for a Dual-rate Sampling System with Output Delay[J]. Control and Decision, 2017, 32(9): 1707-1713.
[6]	杨天皓, 李健, 贾瑶, 等. 虚拟未建模动态补偿驱动的双率自适应控制[J]. 自动化学报, 2018, 44(2): 299-310.
	Yang Tianhao, Li Jian, Jia Yao, et al. Dual-rate Adaptive Control Driven by Virtual Unmodeled Dynamics Compensation in Industrial Heat Exchange Process[J]. Acta Automatica Sinica, 2018, 44(2): 299-310.
[7]	王宏伟, 夏浩. 非均匀多采样率非线性系统的模糊辨识[J]. 控制与决策, 2015, 30(9): 1646-1652.
	Wang Hongwei, Xia Hao. Fuzzy Identification for Non-uniformly Multi-rate Sampled Nonlinear Systems[J]. Control and Decision, 2015, 30(9): 1646-1652.
[8]	Zong Tiancheng, Li Junhong, Lu Guoping. Parameter Identification of Dual-rate Hammerstein-volterra Nonlinear Systems by the Hybrid Particle Swarm-gradient Algorithm Based on the Auxiliary Model[J]. Engineering Applications of Artificial Intelligence, 2023, 117, Part A: 105526.
[9]	刘艳君, 尤俊瑶, 丁锋. 基于辅助模型正交匹配追踪的多输入系统迭代辨识算法[J]. 控制与决策, 2019, 34(4): 787-792.
	Liu Yanjun, You Junyao, Ding Feng. Iterative Identification for Multiple-input Systems Based on Auxiliary Model-orthogonal Matching Pursuit[J]. Control and Decision, 2019, 34(4): 787-792.
[10]	Quaranta Giuseppe, Lacarbonara Walter, Masri S F. A Review on Computational Intelligence for Identification of Nonlinear Dynamical Systems[J]. Nonlinear Dynamics, 2020, 99(2): 1709-1761.
[11]	Baldacchino T, Anderson S R, Kadirkamanathan V. Computational System Identification for Bayesian NARMAX Modelling[J]. Automatica, 2013, 49(9): 2641-2651.
[12]	郑天, 李峰, 贺乃宝, 等. 基于组合式信号源的非线性系统辨识[J]. 系统仿真学报, 2022, 34(11): 2377-2385.
	Zheng Tian, Li Feng, He Naibao, et al. Nonlinear System Identification Based on Combined Signal Sources[J]. Journal of System Simulation, 2022, 34(11): 2377-2385.
[13]	宋伟, 韩佳虎, 李峰, 等. 有色噪声干扰下Hammerstein非线性模型辨识[J]. 陕西科技大学学报, 2023, 41(5): 189-194, 202.
	Song Wei, Han Jiahu, Li Feng, et al. Identification of Hammerstein Nonlinear Model with Colored Noise[J]. Journal of Shaanxi University of Science & Technology, 2023, 41(5): 189-194, 202.
[14]	李峰, 郑天, 宋伟. 基于Hammerstein模型的风力发电系统建模与辨识[J]. 系统仿真学报, 2023, 35(7): 1517-1525.
	Li Feng, Zheng Tian, Song Wei. Modeling and Identification of Wind Power Generation System Based on Hammerstein Model[J]. Journal of System Simulation, 2023, 35(7): 1517-1525.
[15]	徐珊玲, 李俊红, 刘梦茹, 等. Wiener系统的混沌引力搜索迭代辨识[J]. 系统仿真学报, 2021, 33(9): 2138-2146.
	Xu Shanling, Li Junhong, Liu Mengru, et al. Chaotic Gravitational Search Iterative Identification for Wiener Systems[J]. Journal of System Simulation, 2021, 33(9): 2138-2146.
[16]	Li Junhong, Zhang Jiali. Maximum Likelihood Identification of Dual-rate Hammerstein Output-error Moving Average System[J]. IET Control Theory & Applications, 2020, 14(8): 1089-1101.
[17]	Chen Jing, Li Juan, Liu Yanjun. Gradient Iterative Algorithm for Dual-rate Nonlinear Systems Based on a Novel Particle Filter[J]. Journal of the Franklin Institute, 2017, 354(11): 4425-4437.
[18]	Yang Xianqiang, Yang Xuebo. Local Identification of LPV Dual-rate System with Random Measurement Delays[J]. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1499-1507.
[19]	潘雅璞, 谢莉, 杨慧中. 基于KRLS的非均匀采样非线性系统辨识[J]. 控制与决策, 2021, 36(12): 3049-3055.
	Pan Yapu, Xie Li, Yang Huizhong. Identification of Non-uniformly Sampled Nonlinear Systems Based on KRLS[J]. Control and Decision, 2021, 36(12): 3049-3055.
[20]	Han Honggui, Sun Chenxuan, Wu Xiaolong, et al. Training Fuzzy Neural Network via Multiobjective Optimization for Nonlinear Systems Identification[J]. IEEE Transactions on Fuzzy Systems, 2022, 30(9): 3574-3588.
[21]	Xie Li, Pan Yapu, Tao Hongfeng, et al. Identification of Dual-rate Sampled Nonlinear Systems Based on the Cycle Reservoir with Regular Jumps Network[J]. International Journal of Robust and Nonlinear Control, 2023, 33(6): 3551-3565.
[22]	Guo Yu, Wang Fei, Chen Badong, et al. Robust Echo State Networks Based on Correntropy Induced Loss Function[J]. Neurocomputing, 2017, 267: 295-303.
[23]	Filipovic Vojislav Z. Outlier Robust Identification of Dual-rate Hammerstein Models in the Presence of Unmodeled Dynamics[J]. International Journal of Robust and Nonlinear Control, 2022, 32(3): 1162-1179.
[24]	Zhang Minghui, Wang Baozhu, Zhou Yatong, et al. Prediction of Chaotic Time Series Based on SALR Model with Its Application on Heating Load Prediction[J]. Arabian Journal for Science and Engineering, 2021, 46(9): 8171-8187.
[25]	Cui Hongyan, Feng Chen, Chai Yuan, et al. Effect of Hybrid Circle Reservoir Injected with Wavelet-neurons on Performance of Echo State Network[J]. Neural Networks, 2014, 57: 141-151.
[26]	Mirjalili S, Lewis A. The Whale Optimization Algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67.