Neural-Network-Based Parameter Estimations of Induction Motors

Toliyat, H.A.; Włas, Mirosław; Krzemiriski, Z.

doi:10.1109/tie.2008.918615

Cited by 111 publications

(40 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several authors have proposed model optimization techniques [e.g. neural network, GA, particle swarm optimization (PSO)] for parameters identification of induction motor (Kampisios et al 2008;Karanayil et al 2009;Toliyat et al 2008;Chen et al 2007). However, most of these models were not proposed for high frequency behavior.…”

Section: Introductionmentioning

confidence: 99%

Modeling Induction Motor for Prediction of High frequency problems

Mohammadi-Rostam

Shahabi

2016

Iran. J. Sci. Technol. Trans. Electr. Eng.

View full text Add to dashboard Cite

In this paper a novel high-frequency model of Induction motor is presented for investigating induction motor behavior in a wide electromagnetic interference (EMI)-frequency range from 100 Hz to 30 MHz. This model is suitable for prediction of common mode current, EMI problems and terminal over-voltage analysis. The model parameters are estimated by asymptotical method using novel differential and common-mode impedance measurements. The values of parameters are limited in a specified range and then optimized using genetic algorithm (GA). A least mean square method is used with GA optimization to solve the identification problem. The proposed model can also be used to predict the high-frequency problems and design EMI filters to improve electromagnetic compatibility limitation in cable fed motor-drive system. In order to validate the proposed high-frequency motor model, novel common-mode and differential-mode measuring tests have been carried out on the 9.3 kW industrial induction motor using a network analyzer in a wide frequency range as described. Very good agreement was seen between the experimental and simulation results of the proposed model in both magnitude and phase.

show abstract

Section: Introductionmentioning

confidence: 99%

Modeling Induction Motor for Prediction of High frequency problems

Mohammadi-Rostam

Shahabi

2016

Iran. J. Sci. Technol. Trans. Electr. Eng.

View full text Add to dashboard Cite

show abstract

“…For instance, Salmasi and Najafabadi [9] have addressed an adaptive observer which is capable of concurrent estimation of stator currents and rotor fluxes with online adaptation of rotor and stator resistances. Toliyat et al [10] have developed artificial neural networks (ANNs) in closed loop observer for estimating rotor resistance and mutual inductance. There is also a stochastic approach that uses EKF in estimating the variables of IM, such as speed, torque, and flux [11]- [14].…”

Section: Introductionmentioning

confidence: 99%

Experimental Evaluation of Torque Performance of Voltage and Current Models using Measured Torque for Induction Motor Drives

Alsofyani¹,

Sutikno²,

Alamri³

et al. 2015

IJPEDS

View full text Add to dashboard Cite

In this paper, two kinds of observers are proposed to investigate torque estimation. The first one is based on a voltage model represented with a lowpass filter (LPF); which is normally used as a replacement for a pure integrator to avoid integration drift problem due to dc offset or measurement error. The second estimator used is an extended Kalman filter (EKF) as a current model, which puts into account all noise problems. Both estimation algorithms are investigated during the steady and transient states, tested under light load, and then compared with the measured mechanical torque. In all conditions, it will be shown that the torque estimation error for EKF has remained within narrower error band and yielded minimum torque ripples when compared to LPF estimation. This motivates the use of EKF observer in high performance control drives of induction machines for achieving improved torque response.

show abstract

“…The model-based techniques are designed to obtain the speed and other variable information using the measurement of voltage and current at IM terminals. Various approaches have been proposed, such as model reference adaptive system (MRAS) [1], sliding-mode observers [2][3], artificial intelligence techniques [4], and Kalman-filter observers [5]. It is well known that the modelbased estimation schemes have a constraint when the stator frequency approaches zero since the rotor-induced voltage becomes zero, and accordingly the control system turn out to be unobservable.…”

Section: Introductionmentioning

confidence: 99%

Using NSGA II multiobjective genetic algorithm for EKF-based estimation of speed and electrical torque in AC induction machines

Alsofyani

Idris

Jannati

et al. 2014

2014 IEEE 8th International Power Engineering and Optimization Conference (PEOCO2014)

View full text Add to dashboard Cite

High-performance AC drives require accurate speed and torque estimations to provide a proper system operation. The selection and quality of extended Kalman fitter (EKF) covariance elements have a considerable bearing on the effectiveness of motor drive. Many EKF-based optimization techniques involve only a single objective for the optimal estimation of speed without giving concern to the torque. This paper presents a new methodology for the selection of EKF filters that uses non-dominated sorting genetic algorithm-II (NSGA-II) developed for filter element selection in order to investigate the concurrent optimization of speed and torque. The proposed optimizing technique for EKF-based estimation scheme is used in the combination with the sensorless direct torque control of induction motor. The multi-optimal based-EKF is tested with three regions of Pareto front curve.

show abstract

Neural-Network-Based Parameter Estimations of Induction Motors

Cited by 111 publications

References 19 publications

Modeling Induction Motor for Prediction of High frequency problems

Modeling Induction Motor for Prediction of High frequency problems

Experimental Evaluation of Torque Performance of Voltage and Current Models using Measured Torque for Induction Motor Drives

Using NSGA II multiobjective genetic algorithm for EKF-based estimation of speed and electrical torque in AC induction machines

Contact Info

Product

Resources

About