Real‐time identification of nonlinear multiple‐input–multiple‐output systems with unknown input time delay using Wiener model with Neuro‐Laguerre structure

Abstract

In this article, a real-time block-oriented identification method for nonlinear multiple-input-multiple-output systems with input time delay is proposed. The proposed method uses the Wiener structure, which consists of a linear dynamic block (LDB) followed by a nonlinear static block (NSB). The LDB is described by the Laguerre filter lattice, whereas the NSB is characterized using the neural networks. Due to the online adaptation of the parameters, the proposed method can cope with the changes in the system parameters. Moreover, the convergence and bounded modeling error are shown using the Lyapunov direct method. Four practical case studies show the effectiveness of the proposed algorithm in the open-loop and closed-loop identification scenarios. The proposed method is compared with the recently published methods in the literature in terms of the modeling accuracy, parameter initialization, and required information from the system. How to cite this article: Sadeghi M, Farrokhi M. Real-time identification of nonlinear multiple-inputmultiple-output systems with unknown input time delay using Wiener model with Neuro-Laguerre structure. Int J Adapt Control Signal Process. 2019;33:157-174. https://doi.

Dynamical Soft Sensors from Scarce and Irregularly Sampled Outputs Using Sparse Optimization Techniques

“…Laguerre filters are popular OBFs in the context of process industries due to their ability to approximate overdamped systems with high accuracy. 14,15 The selection of Laguerre parameters, namely, the Laguerre pole and order requires prior knowledge about the system dynamics in order to generate parsimonious models. This is especially challenging in the presence of complex linear dynamics involving different time scales and process delays.…”

“…A paradigm shift in the approach is obtained when one constructs an approximate linear representation by expanding the model on a fewer number of OBFs while still achieving high accuracy by choosing the filter parameters that are most relevant to the system. Laguerre filters are popular OBFs in the context of process industries due to their ability to approximate overdamped systems with high accuracy. , The selection of Laguerre parameters, namely, the Laguerre pole and order requires prior knowledge about the system dynamics in order to generate parsimonious models. This is especially challenging in the presence of complex linear dynamics involving different time scales and process delays.…”

Modeling and identification for practical nonlinear process using neural fuzzy network–based Hammerstein system

“…Recently, some novel representation strategies—such as Lipschitz uncertain state space system (Allahverdi et al, 2020), neural networks (Dong et al, 2021; Ma and Yang, 2021), fuzzy systems (Willian et al, 2021; Zangeneh et al, 2020), fuzzy neural network (Zhou et al, 2020), and block-oriented nonlinear systems (Castro et al, 2017; Degachi et al, 2020; Li et al, 2021; Mohsen and Mohammad, 2019; Saif et al, 2020)—have been proposed to improve or construct the nonlinear system. In the literature (Allahverdi et al, 2020), the issues of sensor fault detection and isolation for a class of Lipschitz uncertain nonlinear system were addressed.…”

Parameter identification for Hammerstein nonlinear system with polynomial and state space model

“…Over the last decades, a wide variety of effective modeling methodologies have been developed for approximating nonlinear systems, such as Volterra series, 4 neural networks, 5,6 support vector machines, 7 fuzzy logic systems, 8 and block-oriented nonlinear systems. 9–13 Among these developed modeling methods, the block-oriented systems have attracted a lot of interest owing to prominent modeling ability.…”