Model structure selection for switched NARX system identification: A randomized approach

Bianchi, Federico; Breschi, Valentina; Piga, Dario; Piroddi, Luigi

doi:10.1016/j.automatica.2020.109415

Cited by 13 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The identification of jump input‐output models is addressed in Reference 15 by placing a penalty on mode transitions, in order to control the switching frequency, and retrieving the switching signal by way of dynamic programming. This work has been further extended to jump Box‐Jenkins model in Reference 16 and to jump polynomial nonlinear ARX models in Reference 17. The penalty term plays a similar role to the mode‐switching constraints mentioned previously.…”

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

A distributed randomized method for the identification of switched ARX systems

Yu,

Bianchi,

Piroddi

2024

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

SummaryThe identification of switched systems amounts to a mixed integer nonlinear optimization problem, where the continuous variables are associated to the model parameterizations of the different modes, and the discrete ones are related to the switching signal (each data sample is assigned to a mode, and switching occurs when the mode assignment changes over time). In the batch form of the identification problem, the combinatorial complexity increases exponentially with the size of the training set, which makes the precise identification of the switching signal the most challenging task in the identification problem. To tackle this complexity we propose a distributed optimization approach, based on the solution of multiple instances of a much simpler problem, where switching can occur only at specific time instants (different for each subproblem), and an information sharing mechanism that preserves likely switching times to improve the local solutions. We employ an adapted version of a previously developed randomized algorithm to solve the individual subproblems. Another important feature of the proposed method is an a posteriori heuristic correction method, that is applied to further refine the switching locations based on the estimated local models before the information sharing phase. The performance of the proposed algorithm is analyzed and compared with other methods on synthetic datasets.

show abstract

Section: Introductionmentioning

confidence: 97%

“…** In this article,

{n}_y

and

{n}_u

are assumed to be known as an a priori information. Otherwise, a model structure selection has to be added to the identification process, as done for example, in References 19 or 17.…”

mentioning

confidence: 99%

A distributed randomized method for the identification of switched ARX systems

Yu,

Bianchi,

Piroddi

2024

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

show abstract

“…It does not rely on detailed knowledge of the system's internals but uses sufficient input-output data to fit the model, which can then be used to predict the system's output under different input data. Common black-box models include the Volterra model [5,6], neural network model [7][8][9], nonlinear autoregressive network (NARX) [10], nonlinear autoregressive moving average (NARMAX) [11], and state-space model (SSM) [12,13]. Although black-box models do not provide physical insights into the system, their flexibility and reliance on data make them highly suitable for capturing and modeling complex nonlinear behaviors; thus, they are widely used in the identification and modeling of nonlinear systems.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear System identification of Air turbine rocket engine based on Polynomial nonlinear state space model

Liang,

Zhao,

Zhao

et al. 2024

Preprint

View full text Add to dashboard Cite

The Air Turbine Rocket (ATR) engine is a promising combined cycle propulsion engine. This paper uses the polynomial nonlinear state-space (PNLSS) model to model and identify the nonlinear system of the ATR engine. A method of multiple uncorrelated step signals is proposed as the excitation signals for the nonlinear system. The adaptive Nelder-Mead simplex (ANMS) algorithm is used as the nonlinear least squares optimization algorithm to solve the PNLSS model parameters. The identification results show that the multiple uncorrelated step signals have good excitation effects on the steady-state operating points and the large-scale dynamic processes of the nonlinear system. Under the same initial values, the ANMS algorithm has obvious advantages over the Levenberg-Marquardt (L-M) algorithm and the standard Nelder-Mead simplex (SNMS) algorithm in terms of optimization effect and convergence speed. The PNLSS model shows higher fitting accuracy and prediction ability than the linear state space (LSS) model for the operating points and the wide-range dynamic processes of the ATR engine. This study provides a new method for excitation signal design and parameter identification for nonlinear systems and lays a foundation for the design of nonlinear controllers.

show abstract

“…Switching models are widely used in engineering practices [19,20]. Such models have several modes with different dynamical properties, and the modes are associated with various operating conditions [21,22]. e difficulty in switching system identification is that the times of the operating points (model identities) may be unknown.…”

Section: Introductionmentioning

confidence: 99%

Flexible Least Squares Algorithm for Switching Models

2022

Complexity

View full text Add to dashboard Cite

The self-organizing model and expectation-maximization method are two traditional identification methods for switching models. They interactively update the parameters and model identities based on offline algorithms. In this paper, we propose a flexible recursive least squares algorithm which constructs the cost function based on two kinds of errors: the neighboring two-parameter estimation errors and the output estimation errors. Such an algorithm has several advantages over the two traditional identification algorithms: it (1) can estimate the parameters of all the sub-models without prior knowledge of the model identities; (2) has less computational efforts; and (3) can update the parameters with newly arrived data. The convergence properties and simulation examples are provided to illustrate the efficiency of the algorithm.

show abstract

Model structure selection for switched NARX system identification: A randomized approach

Cited by 13 publications

References 37 publications

A distributed randomized method for the identification of switched ARX systems

A distributed randomized method for the identification of switched ARX systems

Nonlinear System identification of Air turbine rocket engine based on Polynomial nonlinear state space model

Flexible Least Squares Algorithm for Switching Models

Contact Info

Product

Resources

About