A state observer is a system that models a real system in order to provide an estimate of the internal state of the system. The design techniques and comparison of four different types of state observers are presented in this paper. The considered observers include Luenberger observer, Kalman observer, unknown input observer and sliding mode observer. The application of these observers to a Multiple Input Multiple Output (MIMO) DC servo motor model and the performance of observers is assessed. In order to evaluate the effectiveness of these schemes, the simulated results on the position of DC servo motor in terms of residuals including white noise disturbance and additive faults are compared.
Wind energy has achieved prominence in renewable energy production. There- fore, it is necessary to develop a diagnosis system and fault-tolerant control to protect the system and to prevent unscheduled shutdowns. The presented study aims to provide an experimental analysis of a speed sensor fault by hybrid active fault-tolerant control (AFTC) for a wind energy conversion system (WECS) based on a permanent magnet synchronous generator (PMSG). The hybrid AFTC switches between a traditional controller based on proportional integral (PI) controllers under normal conditions and a robust backstepping controller system without a speed sensor to avoid any deterioration caused by the sensor fault. A sliding mode observer is used to estimate the PMSG rotor position. The proposed controller architecture can be designed for performance and robustness separately. Finally, the proposed methodwas successfully tested in an experimental set up using a dSPACE 1104 platform. In this experimental system, the wind turbine with a generator connection via a mechanical gear is emulated by a PMSM engine with controled speed through a voltage inverter. The obtained experimental results show clearly that the proposed method is able to guarantee service production continuity for the WECS in adequate transition.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Power converters by their nature present hybrid behavior since they contains switched circuits. Such circuits can be described by a set of discrete states with associated continuous dynamics. In this paper a new hybrid model for a 3-cells power converter is proposed. This hybrid model incorporates both the continuous and discrete states allowing better understanding of the system operating modes and properties. Of particular interests the new concept of hybrid Z(TN) observability is used to deal with the observability of the capacitors voltages and a discussion on the achievable observation dynamics is given. A sliding mode observer strategy is derived to estimate the flying voltages of the converter. Furthermore, it is derived that under some specific control sequence, the observation errors are asymptotically stable. Finally, experimental results are presented in order to illustrate the performance of the proposed approach.
The paper studies the fault identification problem for linear control systems under the unmatched disturbances. A novel approach to the construction of a sliding mode observer is proposed for systems that do not satisfy common conditions required for fault estimation, in particular matching condition, minimum phase condition, and detectability condition. The suggested approach is based on the reduced order model of the original system. This allows to reduce complexity of sliding mode observer and relax the limitations imposed on the original system.
5
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The rotor speed and position control of a switched reluctance motor (SRM) needs an encoder or resolver to obtain information about the rotor position. These sensors may deteriorate in harsh environments, and increase the size and cost of the drive. This paper presents an evaluation of the sliding-mode observer to estimate a switched reluctance motor speed and rotor position. A new concept of observer based on a simple but quite accurate model of SRM has been proposed. A non-linear model of SRM is used and a computer simulation is developed based on data from an experimental motor system. The implication of discrete-time implementation with various sample rates was examined.
PL
Dla poprawnej pracy układu sterowania przełączalnego silnika reluktancyjnego (ang. Switched Reluctance Motor SRM) wymagany jest resolwer lub cyfrowy przetwornik położenia wału silnika. Jednak zastosowanie takiego przetwornika zwiększa rozmiary i koszt napędu, a poprawna praca przetwornika w obecności zaburzeń elektromagnetycznych może być utrudniona. W artykule przestawiono możliwość zastosowania obserwatora z ruchem ślizgowym do odtwarzania prędkości i położenia wirnika dla przełączalnego silnika reluktancyjnego. Przedstawiona została nowa koncepcja obserwatora, korzystająca z prostego lecz wystarczająco dokładnego modelu SRM. Wykorzystany został nieliniowy model SRM, a symulacje zostały przeprowadzone na bazie danych zebranych w układzie laboratoryjnym. W pracy przeanalizowano właściwości dyskretnej realizacji obserwatora z różnymi czasami próbkowania.
The paper is devoted to the problem of increasing the efficiency of underwater vehicles by using a fault diagnosis system for their thrusters which provides detection, isolation, and identification of minor faults. To address the problem, a two-stage method is proposed. At the first stage, a bank of diagnostic observers is designed to detect and isolate the emerging faults. Each observer in this bank is constructed to be sensitive to some set of faults and insensitive to others. At the second stage, additional observers working in sliding mode are synthesized in order to accurately estimate the error value in the signal obtained from the angular velocity sensor and to estimate deviations of the thruster parameters from their nominal values due to the faults. In contrast to the existing solutions, reduced-order (i.e., lower-dimensional) models of the original system are proposed as a basis to construct sliding mode observers. This approach permits reduction of the complexity of the obtained observers in comparison with the known methods, where full-order observers are constructed. The simulation results show the efficiency and high quality of all synthesized observers. In all cases considered, it was possible to detect typical faults, as well as estimate their values.
7
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W pracy przedstawiono nową metodę estymacji zastępczego napięcia sieci dla celów sterowania aktywnych filtrów mocy. Jakość algorytmów sterowania aktywnych filtrów mocy o stałej częstotliwości przełączeń zależy od dokładności modelu sieci zasilającej. Błąd estymacji zastępczego źródła napięcia reprezentującego sieć zasilającą może być traktowany jako zakłócenie. Zaprezentowany obserwator pracujący w trybie ślizgowym umożliwia śledzenie zmieniającego się napięcia. Zaprezentowano numeryczny przykład realizacji obserwatora ślizgowego współpracującego z filtrem mocy. W przedstawionym algorytmie sterowania struktura przełączana filtru jest zastąpiona przez model zawierający zmienną pojemność.
EN
This paper proposes a new method of equivalent source voltage estimation for power active filters. The performance of fixed switching frequency controllers is constrained to the accuracy of the plant model as well as the accuracy of the reference signal and disturbance predictions. The error of the equivalent source voltage estimation can be treated as a unknown disturbance. The proposed sliding mode observer enables one to track the changing of supplying system voltage. The numerical example of sliding mode observer is presented in the paper. The control algorithm of the presented active power filter is based on the time-varying model. Because the switched structure of active power filter is substituted for time-varying system the observer is quite simple. The computer simulation shows that the problem of low frequency component separation from sliding mode control signal is essential for the method effectiveness.
8
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule opisano wpływ zastosowania ciągłej aproksymacji funkcji znaku na pracę klasycznego ślizgowego obserwatora prędkości. Funkcje nasycenia oraz funkcja sigmoidalna użyte zostały w celu redukcji zjawiska chatteringu, który powstaje na skutek wykorzystania funkcji znaku. Szeroko przebadany został również wpływ stałej czasowej filtru. Zaprezentowane zostały także dwie proste metody adaptacji wzmocnienia obserwatora. Opisywane zagadnienia zostały zilustrowane przy pomocy badań symulacyjnych i eksperymentalnych.
EN
This paper deals with sign function continuous approximation forms influence on the performance of the classical Sliding Mode Observer (SMO). Saturation and sigmoid functions are used to reduce the chattering, introduced due to the sign function usage. The influence of the filter time constant is also being widely considered. Finally, two simple methods of the observer gain adaptation are presented and described. Simulation and experimental tests, obtained using a DSP-based system, are shown to illustrate presented issues.
This research paper proposes a novel design of an efficient combined sliding mode observer (SMO) for induction motor flux and speed estimation. The suggested sensorless technique employs the sliding mode’s second-order approach using a model reference adaptive system (MRAS). The second-order super-twisting control method is free-chattering, which lowers the chattering effect while preserving the same good features as the first-order sliding mode control (SMC). In addition, the conjunction with the MRAS as a separated speed estimator can raise the accuracy and make the observer immune to speed fluctuations, particularly for low-speed applications. Furthermore, in order to achieve effective decoupled flux–torque control, the super-twisting algorithm (STA) was combined with a non-linear feedback linearisation controller for the inner control loop construction. This strategy can boost the control system’s stability and robustness against external disturbances and modelling uncertainty. The performance analysis of the suggested methods has been carried out via simulation and experimental validation utilizing MATLAB/Simulink with the dSpace 1104 real-time interface.
The paper proposes a new dynamic model based on the LuGre model and an electrical equation to describe the hysteresis phenomenon for a magnetorheological (MR) damper. In addition, a sliding mode observer (SMO) is proposed to estimate unmeasurable states of the MR damper. The parameters of the MR damper are successfully identified by using the self-learning particle swarm optimization (SLPSO) algorithm. The contributions of this paper are: i) a new dynamic model based on the LuGre model and an electrical equation for an MR damper is successfully formulated to fit for the hysteresis behavior, ii) the exerted damping force can be practically adjusted by using input voltage for the dynamic model, iii) the SMO is proposed to estimate the internal states and current, and iv) the unknown parameters of the MR damper are successfully identified by using the SLPSO algorithm with a numerical experiment.
For fault detection of doubly-fed induction generator (DFIG), in this paper, a method of sliding mode observer (SMO) based on a new reaching law (NRL) is proposed. The SMO based on the NRL (NRL- SMO) theoretically eliminates system chatter caused by the reaching law and can be switched in time with system interference in terms of robustness and smoothness. In addition, the sliding mode control law is used as the index of fault detection. Firstly, this paper gives the NRL with the theoretically analyzes. Secondly, according to the mathematical model of DFIG, NRL-SMO is designed, and its analysis of stability and robustness are carried out. Then this paper describes how to choose the optimal parameters of the NRL-SMO. Finally, three common wind turbine system faults are given, which are DFIG inter-turn stator fault, grid voltage drop fault, and rotor current sensor fault. The simulation models of the DFIG under different faults is established. The simulation results prove that the superiority of the method of NRL-SMO in state tracking and the feasibility of fault detection.
In recent years there has been an increasing demand for electric vehicles due to their attractive features including low pollution and increase in efficiency. Electric vehicles use electric motors as primary motion elements and permanent magnet machines found a proven record of use in electric vehicles. Permanent magnet synchronous motor (PMSM) as electric propulsion in electric vehicles supersedes the performance compared to other motor types. However, in order to eliminate the cumbersome mechanical sensors used for feedback, sensorless control of motors has been proposed. This paper proposes the design of sliding mode observer (SMO) based on Lyapunov stability for sensorless control of PMSM. The designed observer is modeled with a simulated PMSM model to evaluate the tracking efficiency of the observer. Further, the SMO is coded using MATLAB/Xilinx block models to investigate the performance at real-time.
This paper considers the problem of fault-tolerant control (FTC) and fault reconstruction of actuator faults for linear parameter varying (LPV) descriptor systems with time delay. A polytopic sliding mode observer (PSMO) is synthesized to achieve simultaneous reconstruction of LPV polytopic descriptor system states and actuator faults. Exploiting the reconstructed actuator faults and state estimates, a fault-tolerant controller is designed to compensate the impact of actuator faults on system performance by stabilizing the closed-loop LPV delayed descriptor system. Besides, the controller and PSMO gains are obtained throughout the resolution of linear matrix inequalities (LMIs) using convex optimization techniques. The developed PSMO could force the output estimation error to converge to zero in a finite time when the actuators faults are bounded through the reinjection of the output estimation error via a nonlinear switching term. Simulation results applied to a given numerical system are presented to highlight the superiority and effectiveness of the proposed approach.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.