Frequency Domain Versus Time Domain Methods in System Identification – Revisited

Ljung, Lennart

doi:10.1007/11664550_15

Cited by 10 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We recommend the reader the books of Ljung (1999) and Söderström and Stoica (1989) for a systematic study of time domain system identification. The book of Pintelon and Schoukens (2012) gives a comprehensive introduction to frequency domain identification.…”

Section: Cross-referencesmentioning

confidence: 99%

Frequency Domain System Identification

Schoukens

2014

Encyclopedia of Systems and Control

View full text Add to dashboard Cite

In this chapter we give an introduction to frequency domain system identification. We start from the identification work loop in Ljung Contribution: System Identification --An Overview, Fig. 4, and we discuss the impact of selecting the time or frequency domain approach on each of the choices that are in this loop. Although there is a full theoretical equivalence between the time and frequency domain identification approach, it turns out that, from practical point of view, there can be a natural preference for one of both domains.

show abstract

Section: Cross-referencesmentioning

confidence: 99%

Frequency Domain System Identification

Schoukens

2014

Encyclopedia of Systems and Control

View full text Add to dashboard Cite

show abstract

“…Then, applying (CV4), (CV5), (DFT1), (DFT3) and Lemma 27, we have that the likelihood function for the cases (F1) and (F2) are given as in Corollaries 13 and 14. This procedure has been the usual way to derive the likelihood function described in [18,33,21,34,19,1,42], where the term α is considered as an extra parameter to be estimated. However, this procedure is not useful to derive the likelihood function for the case when α is considered as a random variable that depends on the initial state x 0 , the input signal u t and the noise w t (F3).…”

Section: Remark 16mentioning

confidence: 99%

“…Maximum Likelihood (ML) estimation methods have become a popular approach to dynamic system identification [10,40,18]. Different approaches have been proposed in the time and frequency domains [19,17,21,33,34]. A commonly occurring question is how time-and frequency-domain versions of ML estimation are related.…”

Section: Introductionmentioning

confidence: 99%

“…In the current paper we will see that the apparent differences are a result of inconsistent formulations rather than fundamental issues between the use of time or frequency domain data. In particular, we establish in this paper that the domain chosen to describe the available data (i.e., time or frequency) does not change the result of the estimation problem (see also [39,19]). Instead, it is the choice of the likelihood function, i.e., which parameters are to be estimated and what data is assumed available, that leads to perceived differences in the estimation problems.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the equivalence of time and frequency domain maximum likelihood estimation

et al. 2010

View full text Add to dashboard Cite

Maximum likelihood estimation has a rich history. It has been successfully applied to many problems including dynamical system identification. Different approaches have been proposed in the time and in the frequency domains. In this paper we discuss the relationship between these approaches and we establish conditions under which the different formulations are equivalent for finite length data. A key point in this context is how initial (and final) conditions are considered and how they are introduced in the likelihood function.

show abstract

“…A literatura contém um uso frequente de TD-4SID, onde diversas melhorias no algoritmo já foram formuladas (Overschee, 1996a;Ljung, 2007). Atualmente, o Predictor-Based Subspace Identification (PBSID)é uma robusta ferramenta de identificação, capaz de estimar tanto sistemas de tempo discreto como de tempo contínuo (Hajizadeh, 2017;Hoek, 2017).…”

Section: Introductionunclassified

Metodologia de ponderação em algoritmos baseados em subespaços para modelos de tempo contínuo com comportamento ressonante

Rodrigues

Oliveira

2020

Anais Do Congresso Brasileiro De Automática 2020

View full text Add to dashboard Cite

Este artigo apresenta uma nova metodologia de ponderação para algoritmos de identificação de sistemas baseados em subespaços com especial interesse em modelar sistemas ressonantes a partir de dados de resposta em frequência. Nos métodos de identificação baseados em subespaços, a projeção ortogonal das amostras de resposta em frequência revela informações intrínsecas sobre o sistema, que são posteriormente usadas para calcular os parâmetros do modelo. A novidade apresentada nesse trabalho é a adição de uma matriz de ponderação aos dados, com o objetivo de obter uma aproximação precisa entre a resposta em frequência do modelo e os dados de medição. Dois estudos de caso, baseado em dados de medição, são apresentados para mostrar as propriedades da metodologia descrita. No primeiro, o algoritmo é usado para estimar um modelo matemático para uma estrutura de viga flexível, que apresenta alta característica de ressonância. Os resultados são comparados com um modelo estimado usando a função do Matlab, o N4SID. O algoritmo proposto mostrou um desempenho superior com base no índice RMSE. No segundo estudo de caso, o método é utilizado para estimar um modelo baseado em dados de admitância de um transformador de potencial que opera na Usina Hidrelétrica de Jirau, no Brasil.

show abstract

Frequency Domain Versus Time Domain Methods in System Identification – Revisited

Cited by 10 publications

References 17 publications

Frequency Domain System Identification

Frequency Domain System Identification

On the equivalence of time and frequency domain maximum likelihood estimation

Metodologia de ponderação em algoritmos baseados em subespaços para modelos de tempo contínuo com comportamento ressonante

Contact Info

Product

Resources

About