Gradient-based recursive parameter estimation for a periodically nonuniformly sampled-data Hammerstein–Wiener system based on the key-term separation

The identification of the Hammerstein–Wiener (H-W) systems based on the nonuniform input–output dataset remains a challenging problem. This article studies the identification problem of a periodically nonuniformly sampled-data H-W system. In addition, the product terms of the parameters in the H-W system are inevitable. In order to solve the problem, the key-term separation is applied and two algorithms are proposed. One is the key-term-based forgetting factor stochastic gradient (KT-FFSG) algorithm based on the gradient search. The other is the key-term-based hierarchical forgetting factor stochastic gradient (KT-HFFSG) algorithm. Compared with the KT-FFSG algorithm, the KT-HFFSG algorithm gives more accurate estimates. The simulation results indicate that the proposed algorithms are effective.     

Identification of Hammerstein systems with asymmetric dead-zone nonlinearities using canonical parameterized model

A non-iterative identification method with parameterization of the unknown dead-zone is proposed for Hammerstein systems in presence of asymmetric dead-zone nonlinearities.The canonical parameterized model which is a single expression without segmentation is utilized to describe the dead-zone,based on which a universal-type parametric model can be established to approximate the entire system.This model can be established without separating the nonlinear part from the linear part.The dead-zone parameters and the coefficients in the linear transfer function can be estimated simultaneously according to the proposed algorithm.Numerical experiments are presented to illustrate the effectiveness of the proposed scheme.     

Variance analysis of identified linear MISO models having spatially correlated inputs,with application to parallel Hammerstein models

This contribution concerns variance analysis of linear multi-input single-output models when the inputs are temporally white but where different inputs may be correlated. An expression is provided for the variance of a linearly parametrized estimate of the frequency response function from one block, i.e. from one input to the output. In particular, this expression reveals that the variance increases in one block when the number of estimated parameters in another block is increased, but levels off when the number of parameters in the other block reaches the number of parameters in the block in question. It also quantifies exactly how correlation between inputs affects the resulting accuracy and a graphical representation is provided for this purpose. The results are applicable to parallel MISO Hammerstein models when the nonlinearities are known and generalize an existing variance expression for this type of model.     

超磁致伸缩作动器的率相关Hammerstein模型与H∞鲁棒跟踪控制

利用Hammerstein模型对超磁致伸缩作动器（Giant magnetostrictive actuators,GMA）的率相关迟滞非线性进行建模,分别以改进的 Prandtl-Ishlinskii（Modified Prandtl-Ishlinskii）模型和外因输入自回归模型（Autoregressive model with exogenous input,ARX）代表Hammerstein模型中的静态非线性部分和线性动态部分,并给出了模型的辨识方法. 此模型能在1～100Hz频率范围内较好地描述GMA的率相关迟滞非线性. 提出了带有逆补偿器和H∞鲁棒控制器的二自由度跟踪控制策略,实时跟踪控制实验结果证明了所提策略的有效性.     

Improvements of nonlinear dynamic modeling of hot-film MAF sensor

The operating procedure of engine requires that the mass air flow (MAF) sensor can measure the pulsation air flow accurately. Therefore the MAF sensor must possess fast response speed. It is necessary to study the dynamic characteristic of the MAF sensor. Both the experimental equipments and method used by authors previously are improved to obtain the sensor response data accurately in this paper. The static and dynamic separable modeling method is adopted to build the uniform nonlinear dynamic model of the hot-film MAF sensor in the Hammerstein and Wiener forms, which can describe the large/small flow-rate and positive/negative step responses. The performance indexes are calculated by the actual and model responses.     

一种频率依赖性迟滞的神经网络内模控制

针对压电陶瓷等智能材料存在的依赖输入频率的迟滞非线性问题,采用BP神经网络对迟滞非线性进行辨识,并通过内模控制方案来对其进行控制.在迟滞的建模上,构建了一种静态迟滞非线性环节串联一个对输入频率敏感的线性动态环节组成的Hammerstein模型.在此基础上,得出Hammerstein模型的逆模型,并通过构造的正、逆模型进行内模控制.实验结果说明,提出的建模方法与内模控制方案是有效的.     

The discrete collocation method for Fredholm–Hammerstein integral equations based on moving least squares method

The purpose of this paper is to investigate the discrete collocation method based on moving least squares (MLS) approximation for Fredholm–Hammerstein integral equations. The scheme utilizes the shape functions of the MLS approximation constructed on scattered points as a basis in the discrete collocation method. The proposed method is meshless, since it does not require any background mesh or domain elements. Error analysis of this method is also investigated. Some numerical examples are provided to illustrate the accuracy and computational efficiency of the method.     

聚丙烯反应器熔融指数预测控制算法

针对双环管聚丙烯反应器中的熔融指数难以在线测量的问题,采用机理建模方法建立熔融指数软测量模型。为了实现预测控制,将软测量模型转化为Hammerstein模型形式作为预测模型。并基于这个预测模型设计预测控制器,实现了聚合反应过程的熔融指数预测控制,仿真结果验证了该方法的有效性。     

带有稳定学习的递归神经网络动态偏最小二乘建模

针对目前非线性动态偏最小二乘(PLS)建模方法在拟合较强非线性化工过程时存在的问题, 提出一种基于稳定学习的递归神经网络动态PLS建模方法. 该算法将递归神经网络与Hammerstein模型相结合, 对外部PLS提取的特征向量进行内部建模, 具有逼近较强非线性化工过程的能力, 改善了模型的适用范围. 此外, 采用带有稳定学习的参数更新算法对模型参数进行在线修正, 改善了模型的预测精度和自适应能力. 将此方法应用于氧化铝生产过程铝酸钠溶液组分浓度建模实验, 仿真结果表明, 本方法是可行有效的.