Pith. sign in

REVIEW 4 major objections 4 minor 31 references

Interturn Fault Detection in IPMSMs: Two Adaptive Observer-based Solutions

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A change of variables turns the faulty motor's current equations into a known linear system, so the short-circuit current amplitude is measured directly — exponentially fast with one observer, in finite time with a second.

desk verdict Clever transformation makes known-parameter fault detection work nicely, but the adaptive extension is unsupported as written. read the letter →

arxiv 2505.23125 v1 pith:DPKUIXNZ submitted 2025-05-29 eess.SY cs.SY

Romeo Ortega , Alexey Bobtsov , Leyan Fang , Oscar Texis-Loaiza , Johannes Schiffer This is my paper
classification eess.SYcs.SY MSC 93B5393C40
keywords inter-turnshort-circuitfaultpermanentmagnetsynchronousmotorlinearobserverGPEBOfiniteconvergencetimeintervalexcitationparameterestimationseverityfactor
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that an inter-turn short-circuit fault in a permanent magnet synchronous motor can be detected by reconstructing an auxiliary state whose distance from the measured currents equals the fault current amplitude, rather than by estimating an unknown disturbance term. The auxiliary state satisfies a linear time-varying differential equation whose coefficients are all known, so a simple linear observer recovers it exponentially fast, at a rate set by the motor's resistance and inductances. A second, generalized parameter estimation based observer recovers the same state in finite time, with the convergence time equal to the window of an interval-excitation condition on a filtered regressor. For isotropic motors, the paper adds adaptive versions that estimate resistance and inductance online even under the fault, using a least-squares estimator built for a nonlinear regression equation. If the claims hold, the fault severity factor $(2\eta/3)|i_f(t)|$ becomes a quantity read directly from measured signals, with no high-gain injection and no stochastic assumptions.

What carries the argument

The load-bearing object is the auxiliary state $i_{\alpha\beta}$ of equation (2): the measured $dq$ current plus a fault-dependent correction proportional to $(2\eta/3)\,i_f$ rotated by the rotor angle $\theta$. It converts the fault-detection problem from estimating an unknown disturbance into simulating a known linear time-varying system, whose mismatch with the measured current is exactly the fault amplitude. The finite-time variant is carried by the GPEBO mechanism: the observer error obeys the same LTV dynamics, so it equals $\Phi(t)\Theta$ with $\Phi$ the state-transition matrix and $\Theta$ the unknown initial error; after filtering with $F(p) = \lambda/(p+\lambda)$, the fault equations become the linear regression $y = \Psi^\top\Theta$, and a least-squares estimator with forgetting factor reconstructs $\Theta$ at exactly the interval-excitation time $T_c$. The adaptive extension is carried by the LS+DREM estimator applied to the nonlinear, monotonizable regression $Y = \xi^\top G(\mu)$ derived for isotropic motors.

What would settle it

Run the simulation of Proposition 3 with initial rotor position $\theta_0 = 0$ or $\theta_0 = \pi$: the derivation of the nonlinear regression (19) divides by $\eta_1 = \sin\theta_0$, so the LS+DREM estimates of $R_s$ and $L$ will not converge, contradicting the claim that convergence holds for all initial conditions. Separately, record $\Psi(t)$ during a run of Proposition 2 and compute $\int_0^{T_c}\Psi(s)\Psi^\top(s)\,ds$; if no $T_c$ and $\kappa > 0$ satisfy the interval-excitation inequality (15), the finite-time identity (16) will not hold at the claimed time.

Watch

Extended reading notes

Core claim

The central discovery is the change of variables in Lemma 1: define $i_{\alpha\beta} = [i_d + (2\eta/3)\cos\theta\, i_f,\; i_q - (2\eta/3)\sin\theta\, i_f]^\top$. The faulty motor's $dq$ equations, which contain the unknown fault voltages $e_d, e_q$, transform under this definition into the linear time-varying system $Q \,di_{\alpha\beta}/dt = A(t)i_{\alpha\beta} + b(t)$ with $A(t)$ and $b(t)$ entirely known — the fault terms are absorbed into the transformation rather than treated as disturbances. Because $|i_{\alpha\beta} - i_{dq}| = (2\eta/3)|i_f|$, any reconstruction of $i_{\alpha\beta}$ yields the short-circuit current amplitude directly. Proposition 1 proves that a copy of the LTV system achieves this exponentially fast with rate $\rho = R_s\min\{1/L_q, 1/L_d\}$. Proposition 2 proves that the same reconstruction holds in finite time: writing the observer error as $\varepsilon_{\alpha\beta}(t) = \Phi(t)\Theta$ with $\Theta$ an unknown constant, filtering the signals through $F(p) = \lambda/(p+\lambda)$, and estimating $\Theta$ with a least-squares estimator under an interval-excitation condition gives $|\hat i_{\alpha\beta} - i_{dq}| = (2\eta/3)|i_f|$ for all $t \ge T_c$. Proposition 3 extends the design to the adaptive case for isotropic motors, estimating $R_s$ and $L$ during the fault through a monotonizable nonlinear regression equation solved with the LS+DREM estimator.

Load-bearing premise

The adaptive estimator divides by $\sin\theta_0$, the sine of the initial rotor angle, so the parameter estimates and the adaptive observers built on them silently fail when the initial rotor position is a multiple of $\pi$; and the finite-time observer converges only if the filtered regressor satisfies an interval-excitation condition that the paper assumes but does not verify for the operating point it simulates.

Editorial extensions

If this is right

  • The fault severity factor $(2\eta/3)|i_f(t)|$ becomes an explicit output of the observers, replacing the indirect disturbance or noise reconstructions used by high-gain and Kalman-filter designs.
  • The linear observer detects the fault exponentially fast with no tuning freedom; its convergence rate $\rho = R_s\min\{1/L_q, 1/L_d\}$ is fixed by the motor, which may be too slow in practice.
  • The GPEBO variant achieves finite convergence in exactly the interval-excitation time window $T_c$, independent of motor parameters, while staying robust to additive noise and able to track time-varying parameters thanks to its forgetting factor.
  • For isotropic PMSMs the adaptive versions estimate $R_s$ and $L$ online during the fault — the only reported estimation scheme for the faulty motor — so the observers remain implementable when the motor parameters are uncertain.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the interval-excitation condition is generic for PI-controlled drives — suggested by the simulations but not proven there — the same construction transfers to other rotating machines such as induction motors or wound-field synchronous machines, whose fault models share a skew-symmetric LTV structure.
  • A testable modification suggested by the derivation: re-parameterize the regression so that the division is by $\max\{|\sin\theta_0|, |\cos\theta_0|\}$ instead of $\sin\theta_0$; removing that division would eliminate the known failure mode of the adaptive scheme.
  • Because the fault amplitude is an explicit observer output, the design could feed a fixed-threshold alarm; the estimator's claimed noise robustness would then make false-alarm and missed-detection rates a testable function of the gain $\gamma$ and the forgetting factor.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper presents two observers for detecting inter-turn short-circuit faults (ITSCFs) in PMSMs by reconstructing an auxiliary state iαβ whose distance from the measured currents equals a scaled amplitude of the fault current. The first observer is a simple linear copy of the auxiliary dynamics and converges exponentially with a rate determined by the motor parameters. The second uses the generalized parameter estimation based observer (GPEBO) technique: it represents the observer error as Φ(t)Θ, filters the dynamics to obtain a linear regression in the constant vector Θ, and applies a least-squares estimator with forgetting factor to achieve finite-time convergence under an interval excitation assumption. For isotropic PMSMs, the authors derive a nonlinear regression for the resistance and inductance and propose an LS+DREM estimator to render the observers adaptive. Simulation results compare the two known-parameter observers on an isotropic PMSM with an ITSCF.

Significance. The known-parameter part is a clean and potentially useful contribution: Lemma 1's change of variables turns the faulty model into a known linear time-varying system, Proposition 1's Lyapunov analysis is straightforward, and the GPEBO derivation leading to the linear regression equation (14) is internally consistent. The explicit reconstruction of the fault current amplitude, rather than indirect information, is a genuine advantage over high-gain and Kalman-filter approaches. However, the adaptive extension advertised in the title and abstract is not supported as printed: Proposition 3 divides by sin θ0 without excluding singular initial angles, and the monotonicity proof in Appendix D is erroneous. These issues are repairable, but they currently prevent the paper from fully substantiating the adaptive claims.

major comments (4)
  1. [V, Eq. (19)] The derivation of the nonlinear regression (19) divides by η1 = sin θ0 without any assumption excluding θ0 ∈ πZ. Since θ0 is the initial rotor position, a fixed physical quantity determined by the motor at the start of estimation, the adaptive estimator is undefined for θ0 = 0 or π. The convergence claim in Proposition 3 therefore fails for these initial conditions, and the paper should either add an explicit regularity assumption or show how the singularity can be removed by a coordinate transformation.
  2. [Appendix D] The claim that P = [I3 03×2] yields P∇G(μ) = I3 is false. For G(μ) as defined in Section V, ∇G(μ) has rows [1,0,0], [μ3,0,μ1], [0,1,0], [0,μ3,μ2], [0,0,1], so P∇G(μ) equals [[1,0,0],[μ3,0,μ1],[0,1,0]], not I3, and its symmetrization is indefinite. Hence the strong-monotonicity condition required for the convergence theorem of Proposition 5 is not established. The result appears repairable by choosing P as the selector of rows 1, 3, and 5 of the Jacobian, but the printed proof does not support Proposition 3, P2.
  3. [IV, Proposition 2, F3] The finite-time convergence claim relies on the interval excitation condition (15) for the regressor Ψ(t). The paper does not verify this condition for the operating point used in the Section VI simulations, nor does it state conditions on speed, load, or fault severity under which (15) holds. As a result, the simulated finite-time convergence of the GPEBO is not rigorously connected to the theory. The authors should either verify the IE condition numerically for the simulated trajectory or provide sufficient conditions in terms of the motor operating regime.
  4. [VI and Abstract] The title and abstract advertise 'adaptive observer-based solutions', but the adaptive version based on Proposition 3 is neither simulated nor experimentally validated: the simulations in Section VI use known values of Rs and L, and the concluding remarks list implementation of the adaptive versions as future research. Since Proposition 3 is the only support for the adaptive claim and is not exercised in the paper, the presentation should either include adaptive simulations or explicitly limit the claims to the known-parameter observers.
minor comments (4)
  1. [II-A, Eq. (1b)] Equation (1b) contains a typo: the coefficient of diq/dt should be 1/Lq, not 1/Ld; otherwise it is inconsistent with the matrix Q = diag{Ld, Lq} used in Lemma 1 and Proposition 1.
  2. [VI] The abbreviation 'OLO' used in Figs. 2 and 3 is not defined; it presumably stands for 'open-loop observer', which should be stated either in the text or in the caption.
  3. [Appendix C] The finite-convergence-time property of the standard LS estimator is attributed to the submitted manuscript [21]; since this reference is not yet available, the authors should either give a self-contained proof or cite a published source for this result.
  4. [VI] Figs. 5 and 6 show the GPEBO parameter Θ(t) converging, but they do not show the estimates of Rs and L from Proposition 3; the simulation section would benefit from clarifying that the adaptive parameter estimator is not exercised in the presented results.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular reduction found: the fault-detection observers are derived in-paper from the fault model, and the self-cited GPEBO/DREM/FCT lemmas are supporting mathematical tools, not inputs that make the conclusion equal to its assumptions.

full rationale

The central detection chain is self-contained. Lemma 1 defines the change of variables iαβ (Eq. 2) and derives Q diαβ/dt = A iαβ + b (Eq. 3) directly from the faulty PMSM equations; Proposition 1 gives a complete Lyapunov proof of exponential convergence with no use of fault data. Proposition 2 derives the linear regression y = Ψ^T Θ (Eqs. 10, 14) in the text from (1), (2) and (13), and Θ is the observer initial-error vector, not a fitted fault quantity; the fault indicator 2η/3 |if| is obtained from the converged state via (4), so it is not an input to the estimator. The adaptive regression (19) is likewise obtained from the model, with the overparameterized map G(µ) explicitly stated. The paper does lean on the authors' earlier GPEBO [18], LS+DREM [20], and the submitted finite-convergence-time note [21] for general parameter-estimator facts, but these are external mathematical lemmas about LS estimators and are also attributed to [25] and [20]; they are not equivalent to the ITSCF detection conclusion, so the self-citation is not a circular reduction. Two non-circular weaknesses are worth flagging: Proposition 3 divides by η1 = sin θ0 to form (19), so the adaptive observer is undefined for initial rotor angles that are multiples of π, and the Appendix D claim that P∇G(µ) = I3 for P = [I3 0] is arithmetically false as printed—the Jacobian has rows [1,0,0], [μ3,0,μ1], [0,1,0], whose symmetrization is indefinite. These are correctness and assumption gaps, not cases where a prediction equals its input by construction. The paper also states in Section VII that the adaptive version has not been implemented or experimentally validated, consistent with a limited but non-circular contribution.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the faulty PMSM model from prior literature; no new physical entities are proposed. Tuning gains are simulation choices, not fitted to data. The load-bearing assumptions are the exactness of the fault model, interval excitation of the regressors, nonzero sin(theta0) for the adaptive extension, and correctness of the cited FCT and DREM results. One advertised proof step, the monotonicity of G, is incorrect as printed.

free parameters (2)
  • Observer and estimator tuning gains in Section VI = lambda=100, gamma=1, chi0=4, k=15, f11=1, f12=1/5, f21=1/4, f22=1/3, G=diag(6.25e5, 6.25e12)
    Chosen by hand for the simulation; not fitted to physical data. The stability and convergence proofs hold for any values in the allowed ranges, so these gains affect transient behavior but not the validity of the central claim.
  • mu3 = eta2/eta1 = cot(theta0) = Unknown; estimated online by LS+DREM in Proposition 3
    Overparameterized nuisance parameter in the NLRE (19). Its presence increases the estimator dimension and requires sin(theta0) != 0 for the division in Section V.
assumptions (6)
  • domain assumption The faulty PMSM is exactly described by the dq model (1) with constant parameters and a single short-circuit current if in phase A.
    The observer derivations in Lemma 1 and Propositions 1 to 3 inherit this model from [7], [26]. If additional unmodeled fault dynamics or parameter changes occur, the residual relation (4) is biased. No experimental validation is provided.
  • domain assumption The electrical rotor position theta, speed omega, voltages vdq, and currents idq are measured or available with negligible noise.
    Both observers require A(t) and b(t) known online. Section VI uses ideal simulation signals without measurement noise.
  • domain assumption The regressor Psi(t) in (10) satisfies the interval excitation condition (15).
    Invoked in Proposition 2 F3 for finite convergence time. The paper does not verify this for the PI-controlled PMSM operating point used in Section VI, though Lemma 2 states IE is equivalent to identifiability.
  • ad hoc to paper The NLRE map G is monotonizable with an appropriate matrix P.
    Appendix D claims P=[I3 0] gives P nabla G = I3, but direct computation gives a non-identity matrix. The monotonicity proof is incorrect as printed, though it may be repairable with a different P.
  • ad hoc to paper sin(theta0) is nonzero for the parameter estimator.
    In Section V, the NLRE (19) is obtained by dividing by eta1 = sin(theta0). Initial rotor positions with sin(theta0)=0 are excluded without comment.
  • standard math The finite-convergence-time least-squares result in Proposition 4 is correct as stated in [21], [25].
    The proof is not included in this paper and relies on a submitted self-cited manuscript [21] and a textbook [25].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Interturn Fault Detection in IPMSMs: Two Adaptive Observer-based Solutions." pith.science (2026). https://pith.science/paper/DPKUIXNZ

@misc{pith2026250523125,
  author       = {Pith},
  title        = {Pith review of: Interturn Fault Detection in IPMSMs: Two Adaptive Observer-based Solutions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DPKUIXNZ}},
  note         = {Machine review of arXiv:2505.23125}
}
read the original abstract

In this paper we address the problem of online detection of inter-turn short-circuit faults (ITSCFs) that occur in permanent magnet synchronous motors (PMSMs). We propose two solutions to this problem: (i) a very simple linear observer and (ii) a generalized parameter estimation based observer, that incorporates a high performance estimator -- with both observers detecting the short-circuit current and the fault intensity. Although the first solution guarantees the detection of the fault exponentially fast, the rate of convergence is fully determined by the motor parameters that, in some cases, may be too slow. The second observer, on the other hand, ensures finite convergence time under the weakest assumption of interval excitation. To make the observers adaptive, we develop a parameter estimator that, in the case of isotropic PMSMs, estimates on-line (exponentially fast) the resistance and inductance of the motor. It should be underscored that, in contrast with existing observers (including the widely popular Kalman filter) that provide indirect information of the fault current, our observers provide explicit one -- namely the amplitude of the fault current. The performance of both observers, in their linear and generalized parameter estimation-based versions, is illustrated with realistic simulation studies.

Figures

Figures reproduced from arXiv: 2505.23125 by the authors.

Figure 1
Figure 1. Inter-turn short circuit in phase A for a PMSM. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Comparison of the current iα and its estimate ˆiα using OLO and GPEBO In order to address the problem of online detection of ITSCFs in PMSMs, the open loop observer (OLO) and the GPEBO were considered. The OLO was designed following Proposition 1, while the GPEBO is designed according to Proposition 2, with the observer of iαβ given in (5), and the parameter estimator defined in (26) and (27). In the simulation, the… view at source ↗
Figure 4
Figure 4. The norm of the estimated fault current |idαβ(t) − idq(t)| and the norm of the actual fault current 2η 3 |if (t)|. 0 0.05 0.1 0.15 0.2 0.25 0 0.2 0.4 0.6 0.8 1 0 1 2 3 10-4 0 0.5 1 [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: The estimated value θˆ 1 and the actual one θ1 of GPEBO 0 0.05 0.1 0.15 0.2 0.25 0 0.5 1 1.5 2 0.01 0.015 0.02 0.025 0.03 1 1.05 1.1 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The estimated value θˆ 2 and the actual one θ2 of GPEBO [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages

  1. [21]

    Standard LS parameter estimators ensure finite convergence time for linear regression equations under an interval excitation assumption

    R. Ortega, J. G. Romero, S. Aranovskiy and G. Tao, “Standard LS parameter estimators ensure finite convergence time for linear regression equations under an interval excitation assumption”, Automatica, (submit- ted), 2025

    work page 2025

  2. [1]

    A brushless permanent magnet machine with integrated differential

    K. Atallah and J. Wang, “A brushless permanent magnet machine with integrated differential”, IEEE Trans. Magn. , vol. 47, no. 10, pp. 4246– 4249, 2011

    work page 2011

  3. [2]

    Detection of external rotor PMSM inter-turn short circuit fault using extended Kalman filter

    A. Belkhadir, R. Pusca, R. Romary, D. Belkhayat, and Y . Zidani, “Detection of external rotor PMSM inter-turn short circuit fault using extended Kalman filter”, in 2023 IEEE 14th International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives (SDEMPED), pp. 491–497, 2023

    work page 2023

  4. [3]

    On line interturn short-circuit fault diagnosis and nonlinear control of PMSM

    S. Bouslimani, S. Drid, L. Chrifi-Alaoui, and L. Delahoche, “On line interturn short-circuit fault diagnosis and nonlinear control of PMSM”, in 19th International Multi-Conference on Systems, Signals & Devices (SSD), pp. 1139-1143, May, 2022

    work page 2022

  5. [4]

    Inter-turn short circuit fault diagnosis of PMSM

    X. Chen, P. Qin, Y . Chen, J. Zhao, W. Li, Y . Mao and T. Zhao, “Inter-turn short circuit fault diagnosis of PMSM”, Electronics, vol. 11, no. 1576, 2022

    work page 2022

  6. [5]

    Robust adaptive sensorless control for permanent-magnet synchronous motors

    J. Choi, K. Nam, A. Bobtsov, A. Pyrkin, and R. Ortega, “Robust adaptive sensorless control for permanent-magnet synchronous motors”, IEEE Trans. Power Electron., vol. 32, no. 5, pp. 3989-3997, 2016

    work page 2016

  7. [6]

    Dissipativity of nonlinear systems of differential equations

    B. P. Demidovich, “Dissipativity of nonlinear systems of differential equations”, in Vestnik Moscow State University, Ser. Mat. Mekh., Part I-6, (1961) pp. 19-27; Part II-1, (1962), pp. 3-8, (in Russian)

    work page 1961

  8. [7]

    Interturn fault diagnosis strategy for interior permanent-magnet synchronous motor of electric vehicles based on digital signal processor

    B. Du, S. Wu, S. Han, and S. Cui, “Interturn fault diagnosis strategy for interior permanent-magnet synchronous motor of electric vehicles based on digital signal processor”, IEEE Trans. Ind. Electron. , vol. 63, no. 3, pp. 1694-1706, 2016

    work page 2016

Show all 31 references
  1. [8]

    Interturn Short-Circuit Fault Diagnosis of PMSM for Small Hydropower

    L. Fang, X. Zhang, S. Huang, J. Xie, Z. Long and F. Huang, “Interturn Short-Circuit Fault Diagnosis of PMSM for Small Hydropower”, in 2021 IEEE 4th International Electrical and Energy Conference (CIEEC) , pp. 1-5, 2021

    2021

  2. [9]

    Fliess and C

    M. Fliess and C. Join, Model-free control and intelligent PID controllers: towards a possible trivialization of nonlinear control?, in 15th IFAC Symp. System Identif., Saint-Malo, France, 2009

    2009

  3. [10]

    Fliess and C Join, Model-free control, Int

    M. Fliess and C Join, Model-free control, Int. J. of Control, vol. 86, no. 12, pp. 2228-2252, 2013

    2013

  4. [11]

    Goodwin and K

    G. Goodwin and K. Sin, Adaptive Filtering Prediction and Control , Prentice-Hall, 1984

    1984

  5. [12]

    A note on fixed-and discrete-time estimation via the DREM method

    M. Korotina, S. Aranovskiy, R. Ushirobira, D. Efimov and J. Wang, “A note on fixed-and discrete-time estimation via the DREM method”, IEEE Trans. Autom. Control, vol. 69, no. 7, pp. 4793-4797, 2024

    2024

  6. [13]

    Richness and excitation on an interval—with application to continuous-time adaptive control

    G. Kreisselmeier and G. Rietze-Augst, “Richness and excitation on an interval—with application to continuous-time adaptive control”, IEEE Trans. Autom. Control, vol. 35, no. 2, pp. 165-171, 1990

    1990

  7. [14]

    A computationally efficient design technique for electric-vehicle traction machines

    P. Lazari, J. Wang, and L. Chen, “A computationally efficient design technique for electric-vehicle traction machines”, IEEE Trans. Ind Appl. , vol. 50, no. 5, pp. 3203–3213, 2014

    2014

  8. [15]

    Interturn short-circuit fault diagnosis in PMSM with partitioned stator windings

    M. Mazzoletti, G. Bossio and C. De Angelo, “Interturn short-circuit fault diagnosis in PMSM with partitioned stator windings”, IET Elect. Power Appl., vol. 14, no. 12, pp. 2301-2311, 2020

    2020

  9. [16]

    Nam, AC Motor Control and Electric Vehicle Applications , CRC Press, 2010

    K. Nam, AC Motor Control and Electric Vehicle Applications , CRC Press, 2010

    2010

  10. [17]

    On modified parameter estimators for identification and adaptive control: a unified framework and some new schemes

    R. Ortega, V . Nikiforov and D. Gerasimov, “On modified parameter estimators for identification and adaptive control: a unified framework and some new schemes”, Annu. Rev. Control, vol. 50, pp. 278-293, 2020

    2020

  11. [18]

    Generalized parameter estimation-based observers: Application to power systems and chemical-biological reactors

    R. Ortega, A. Bobtsov, N. Nikolayev, J. Schiffer and D. Dochain, “Generalized parameter estimation-based observers: Application to power systems and chemical-biological reactors”, Automatica, vol. 129, pp. 109635, 2021

    2021

  12. [19]

    Parameter identification with finite-convergence time alertness preservation

    R. Ortega, A. Bobtsov and N. Nikolayev, “Parameter identification with finite-convergence time alertness preservation”, IEEE Contr. Syst. Lett ., vol. 6, pp. 205-210, 2022

    2022

  13. [20]

    A new least squares parameter estimator for nonlinear regression equations with relaxed excitation conditions and forgetting factor

    R. Ortega, J. G. Romero and S. Aranovskiy, “A new least squares parameter estimator for nonlinear regression equations with relaxed excitation conditions and forgetting factor”, Syst. Control Lett., vol. 169, pp. 105377, 2022

    2022

  14. [22]

    Rao and H

    C. Rao and H. Toutenburg, H, Linear Models: Least Squares and Alternatives, Springer Series in Statistics (3rd ed.). Berlin, Springer, 2008

    2008

  15. [23]

    Sastry, M

    S. Sastry, M. Bodson, Adaptive Control: Stability, Convergence and Robustness, Prentice-Hall, New Jersey, 1989

    1989

  16. [24]

    J.-J. E. Slotine and W. Li, Applied Nonlinear Control , Prentice-Hall, New Jersey, USA, 1991

    1991

  17. [25]

    Tao, Adaptive Control Design and Analysis , vol

    G. Tao, Adaptive Control Design and Analysis , vol. 37, John Wiley & Sons, New Jersey, 2003

    2003

  18. [26]

    A BL-Homogeneous observe for ITSCF detection in PMSMS

    O. Texis, E. Zurita and J. Schiffer, “A BL-Homogeneous observe for ITSCF detection in PMSMS”, in 2024 IEEE Conference on Control Technology and Applications (CCTA), Newcastle upon Tyne, UK, August 21-23, 2024

    2024

  19. [27]

    Identifiability implies robust, globally exponentially convergent on-line parameter esti- mation: Application to model reference adaptive control

    L. Wang, R. Ortega, A. Bobtsov, J. G. Romero and B. Yi, “Identifiability implies robust, globally exponentially convergent on-line parameter esti- mation: Application to model reference adaptive control”, Int. J. Control, vol. 97, no. 9, pp. 1967-1983, 2023

    1967

  20. [28]

    Inter-turn short circuit faults in permanent magnet synchronous machines: an extended review and comprehensive analysis

    M. Zafarani, E. Bostanci, Y . Qi, T. Goktas, and B. Akin, “Inter-turn short circuit faults in permanent magnet synchronous machines: an extended review and comprehensive analysis”, IEEE J. Emerg. Sel. Topics Power Electron., vol. 6, no. 4, pp. 2173-2191, 2018

    2018

  21. [29]

    Fault diagnosis of inter- turn short circuit of permanent magnet synchronous motor based on voltage residual vector

    Z. Zhang, Z. Yin, D. Yuan, and M. Yuan, “Fault diagnosis of inter- turn short circuit of permanent magnet synchronous motor based on voltage residual vector”, in 2022 IEEE International Power Electronics and Application Conference and Exposition (PEAC) , pp. 617–621, 2022

    2022

  22. [30]

    Inter-turn short circuit fault detection of a novel PMSM with fault tolerance

    L. Zhao, Y . Bai, X. Wang, and P. Zhao, “Inter-turn short circuit fault detection of a novel PMSM with fault tolerance”, in International Conference on Power and Renewable Energy , pp. 209-213, 2023

    2023

  23. [31]

    Online interturn short circuits fault monitoring for permanent magnet synchronous machines

    Y . Zuo, A. Darabi, C. Lai, and K. L. V . Iyer, “Online interturn short circuits fault monitoring for permanent magnet synchronous machines”, in IECON 2022–48th Annual Conference of the IEEE Industrial Electronics Society, pp. 1–6, 2022

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.