Tracking and removing modulated sinusoidal components: A solution based on the kurtosis and the Extended Kalman Filter

Dion, Jean-Luc; Stéphan, Cyrille; Chevallier, Gaël; Festjens, Hugo

doi:10.1016/j.ymssp.2013.04.001

Cited by 25 publications

(16 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the experiments use a sampling period T s ¼ 0.002 s and the trapezoidal numerical integration method, which allows evaluating the integrals of the variables χ, Φ c and Φ k shown in (27), (28) and (29), respectively. The laboratory prototype (see Fig.…”

Section: Resultsmentioning

confidence: 99%

“…The relative accelerations € x im in (13) are used for constructing matrix € u m that appears in the regressor components Φ c , Φ k shown in (28) and (29), respectively. On the other hand, define the vector €…”

Section: Model Parametrizationmentioning

confidence: 99%

“…References [2][3][4][5][6][7][8] present adaptive observers for systems whose models need to be written in special canonical forms obtained through static or dynamic transformations. In the particular case of civil structures, several techniques that estimate their parameters and state composed of displacements and velocities, and that do not resort to canonical transformations have been proposed in [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]. Observer/Kalman filter identification (OKID) approaches are presented in [9][10][11] and they compute the Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/ymssp Markov parameters of a Kalman filter observer.…”

Section: Introductionmentioning

confidence: 99%

“…Wang et al [27] propose an adaptive observer for structures whose design employs the Hilbert transform. Moreover, Dion et al [28] present an adaptive observer that estimates the natural frequencies of structures and that is based on the Optimized Spectral Kurtosis and the Extended Kalman Filter (EKF); this adaptive observer tracks and removes sinusoidal components, which pollute the acceleration measurements and come from engines in operation and from spurious harmonic components of the electric power supply. Finally, references [29][30][31] propose Sequential Monte Carlo (SMC) methods that are robust against measurement noise, have gained popularity due to their superiority over the EKF, can simultaneously estimate the state and parameters of buildings, and require a discrete-time model of this system; with the SMC methods, the state and parameter identification consist in the computation of a non-Gaussian probability density function pðx k ; θjD k Þ, where x k is the state, θ contains the parameters of the structure, and D k is a set that contains the measurements available up to time instant k. It is worth mentioning that most of the references consider the presence of measurement noise, but none of them take into account offsets in the acceleration measurements.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Simultaneous parameter and state estimation of shear buildings

Concha

Álvarez-Icaza

Garrido

2016

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Section: Model Parametrizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Simultaneous parameter and state estimation of shear buildings

Concha

Álvarez-Icaza

Garrido

2016

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

“…Model-based (e.g. Kalman filter) ap-proaches demonstrate good accuracy [47][48][49], yet a patent limitation is that all harmonics have to be modeled -even those which are not of interestso as to produce a Gaussian residual. An important body of literature has also addressed the problem by using tailored time-frequency decomposition such as the Gabor transform and wavelet transforms [50][51][52][53][54][55].…”

Section: Introductionmentioning

confidence: 99%

A multi-order probabilistic approach for Instantaneous Angular Speed tracking debriefing of the CMMNO׳14 diagnosis contest

Leclère

André²,

Antoni

2016

Mechanical Systems and Signal Processing

113

View full text Add to dashboard Cite

The aim of this work is to propose a novel approach for the estimation of the Instantaneous Angular Speed (IAS) of rotating machines from vibration measurements. This work is originated from the organisation, by the authors of this paper, of a contest during the conference CMMNO 2014, that was held in Lyon, december 2014. One purpose of the contest was to extract the IAS of a wind turbine from a gearbox accelerometer signal. The analysis of contestant contributions led to the observation that the main source of error in this exercise was the wrong association of one selected and tracked harmonic component with one mechanical periodic phenomenon, this association being assumed as an a priori hypothesis by all the methods used by the contestants. The approach proposed in this work does not need this kind of a priori assumption. A majority (but not necessarily all) periodical mechanical events are considered from a preliminary analysis of the kinematics of the machine (harmonics of shaft rotation speeds, meshing frequencies ...). The IAS is then determined from probability density functions that are constructed from instantaneous spectra of the signal. The efficiency and robustness of the proposed approach are illustrated in the frame of the CMMNO 2014 contest case.

show abstract

A multi‐tone central divided difference frequency tracker with adaptive process noise covariance tuning

Brumana

Piroddi

2020

Adaptive Control & Signal

View full text Add to dashboard Cite

SummaryThe problem of real‐time frequency estimation of nonstationary multi‐harmonic signals is important in many applications. In this paper, we propose a novel multi‐frequency tracker based on a state‐space representation of the signal with Cartesian filters and the second‐order central divided difference filter (CDDF), which improves the performance of the extended Kalman filter (EKF) by using Stirling's interpolation method to approximate the mean and covariance of the state vector. A crucial element of the method is the adaptive scaling of the process noise covariance matrix appearing in the filter equations, as a function of the innovation sequence, which tunes the accuracy‐reactivity trade‐off of the filter. The proposed solution is evaluated against two approaches from the literature, namely the factorized adaptive notch filter (FANF) and the extended Kalman filter frequency tracker (EKFFT). Several experiments emphasize the estimation accuracy of the proposed method as well as the improved robustness with respect to initial errors and input signal complexity. The presented method appears to be particularly efficient with rapidly varying frequencies, thanks to the update mechanism that adjusts the filter parameters based on the amplitude of the estimation error.

show abstract

Tracking and removing modulated sinusoidal components: A solution based on the kurtosis and the Extended Kalman Filter

Cited by 25 publications

References 36 publications

Simultaneous parameter and state estimation of shear buildings

Simultaneous parameter and state estimation of shear buildings

A multi-order probabilistic approach for Instantaneous Angular Speed tracking debriefing of the CMMNO׳14 diagnosis contest

A multi‐tone central divided difference frequency tracker with adaptive process noise covariance tuning

Contact Info

Product

Resources

About