Sensorless Control of the Levitated Ball

Bobtsov, Alexey A.; Pyrkin, Anton A.; Vedyakov, Alexey A.; Sinetova, Madina

doi:10.1016/j.ifacol.2019.12.662

Cited by 4 publications

(8 citation statements)

References 23 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bobtsov et al 22 proposed the method of flux linkage

λ (t)

estimation using parameter estimation based observer (PEBO) approach 23 . The authors introduce a new signal

ψ (t)

with

ψ (0) = 0

and

\dot{ψ} (t) = - R i (t) + u (t) .

Integrating and combining () and () gives

λ (t) = ψ (t) + η,

where

η : = λ (0) - ψ (0)

.…”

Section: Examplesmentioning

confidence: 99%

“…Consider the model of the 1‐DOF MagLev system () and the dynamic extension (). The constant parameter

η

satisfies the following regression model:

y (t) = ϕ^{T} (t) Θ,

where

Θ = {[\begin{matrix} η & η^{2} & η^{3} & η^{4} & η^{5} & η^{6} \end{matrix}]}^{T} .

For the analytical representation of the signals

y (t) \in ℝ

and

ϕ (t) \in ℝ^{6}

, see Reference 22. Here we highlight that the signals

y (t)

and

ϕ (t)

are calculated from measured signals

i (t)

u (t)

, and known model parameters.…”

Section: Examplesmentioning

confidence: 99%

“…Step 6. Using (), we get the estimate of magnetic flux

\hat{λ}

from

\hat{η}

\hat{λ} = ψ + \hat{η} .

In the simulations, we compare the full state observer described in Reference 22 (labeled as DREM in figures) and the proposed one for the 1‐DOF MagLev model controlled by the sensorless version of the feedback linearization control (FLC) 29 . In the Simulink model for each algorithm, the flux estimate obtained with this algorithm is used in the controller instead of the real flux value.…”

Section: Examplesmentioning

confidence: 99%

“…The gradient estimator from Reference 22 uses the following gains

α = 35

γ = 100

, the DREM filter parameters:

κ_{1} = 1

κ_{2} = 2

κ_{3} = 10

κ_{4} = 20

κ_{5} = 5

κ_{6} = 7

v_{i} = κ_{i}

i = \overset{true‾}{1, 6}

. The same parameters are used for the proposed estimator.…”

Section: Examplesmentioning

confidence: 99%

“…For better transient performance, we update the gradient estimator parameters for each method. For DREM method, 22 the gains are

α = 100

γ = 400

. For proposed method, we choose the relatively high threshold value

ϵ = 1

and gain

γ = 1

.…”

Section: Examplesmentioning

confidence: 99%

See 4 more Smart Citations

Overparameterized model parameter recovering with finite‐time convergence

Ovcharov

Vedyakov

Kazak

et al. 2022

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

In this article, we consider the online parameter estimation problem for overparameterized linear regression models. Many estimation methods require at least the interval excitation condition, which is not always fulfilled for such models, especially in normal operation. To relax the required excitation level, we detect linearly dependent columns in the regressor, obtain the reduced model and estimate parameters in a finite time. On the last step, the parameters of the initial model are recovered. The proposed method efficiency is demonstrated on a simple overparameterized model and a magnetic levitation system.

show abstract

“…Bobtsov et al 22 proposed the method of flux linkage