Neural networks applied for induction motor speed sensorless estimation

Fodor, D.; Ionescu, Felicia; Floricau, D.; Six, J.P.; Delarue, Philippe; Diana, Donatella; Griva, G.

doi:10.1109/isie.1995.496623

Cited by 16 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several review and comparison papers are available on sensorless control techniques [4][5][6][7]. The commonly used methods for speed estimation are Model Reference Adaptive System (MRAS) [8][9][10][11], Neural Networks [12][13][14][15][16][17][18][19], Extended Kalman Filter (EKF) [20][21][22][23][24][25] and Nonlinear Observer [26][27][28][29]. Techniques for high performance induction motor control implementation requin; precise knowledge of machine parameters, and the rotor resistance being the most important of them.…”

Section: Research Problemmentioning

confidence: 99%

Commande de haute performance sans capteur d'une machine asynchronen /

Thongam¹

2006

View full text Add to dashboard Cite

Induction machine, when driven by a field oriented controller behaves like a separately excited dc machine, where torque and flux are naturally decoupled and are controlled independently; and this control strategy allows high performance to be achieved from it.However, conventional vector controlled induction motor drive has the disadvantage of requiring speed sensor which increases the cost and complexity of the drive system.Moreover, drive performance is affected by rotor resistance, whose unknown variation during the operation causes incorrect decoupling of flux and torque which leads to deterioration of drive performance. This thesis focusses on the development of a high performance sensorless induction motor drive. Control of the drive is done in rotor flux coordinates which allows natural decoupling of flux and torque producing components of stator current space vector. Sensorless control of the drive system is achieved by developing rotor flux and speed estimation algorithms using only the measurable stator terminal quantities: the current and voltage. The state and parameter estimation problem is looked into as a question of proper mathematical modeling of the machine in an appropriate reference frame, which along with proper choice of states, enables us to develop an estimation algorithm capable of accurate estimation.In this work a two pronged approach is adopted to achieve the research objective. First approach is based on the fact that dynamics of rotor speed is much slower than that of electrical states of the machine. Hence, the time derivative of rotor speed can be conveniently equated to zero in the machine model used for implementing the estimation

show abstract

Section: Research Problemmentioning

confidence: 99%

Commande de haute performance sans capteur d'une machine asynchronen /

Thongam¹

2006

View full text Add to dashboard Cite

show abstract

“…For example, in [82] a three-layer NN was used with offline training. This technique gives a fairly good estimate of speed and is robust to parameter variation.…”

Section: Neural Network As the Adaptation Mechanismmentioning

confidence: 99%

Modèle temps réel d'un estimateur de vitesse à base de réseaux de neurones pour une machine asynchrome /

Haghgoeian¹

2005

View full text Add to dashboard Cite

“…Speed sensorless control technique first appeared in (Abbondante & Brennen, 1975). The commonly used methods for speed estimation are Model Reference Adaptive System (MRAS) (Schauder, 1992;Tajima & Hori, 1993;Choy et al, 1996), Neural Networks (Simoes & Bose, 1995;Fodor et al, 1995;Ben-Brahim & Kudor, 1995;Kim et al, 2001;Toqeer & Bayindir, 2003;Haghgoeian et al, 2005), Extended Kalman Filter (EKF) (Kim et al,1994, Comnac et al, 2001Ma & Gui, 2002;Du et al, 1995;Thongam & Thoudam, 2004) and Nonlinear Observer (Bodson et al, 1995;Liu et al, 2001, Pappano et al, 1998. The aim of this chapter is to provide with a brief overview of high performance sensorless induction motor drive.…”

Section: Introductionmentioning

confidence: 99%

Electric Machines and Drives

Filizadeh¹

2013

View full text Add to dashboard Cite

Miroslav Chomat attained his M.Sc. degree in Electrical Power Engineering from the Czech Technical University in Prague in 1988 and his Ph.D. degree in Electrical Engineering from the Institute of Electrical Engineering of the Academy of Sciences of the Czech Republic in 1994. He is currently a senior researcher and the head of the Department of Electrical Machines, Drives, and Power Electronics at the Institute of Thermomechanics AS CR, v.v.i. He is also an assistant professor at the Department of Electric Drives and Traction of the Czech Technical University in Prague. Between 1999 and 2000, he spent one year at the University of Wisconsin-Madison, USA working as a honorary research fellow. His research interests include numerical modeling and control of electrical machines, drives, power units and electronic converters.

show abstract

Neural networks applied for induction motor speed sensorless estimation

Cited by 16 publications

References 3 publications

Commande de haute performance sans capteur d'une machine asynchronen /

Commande de haute performance sans capteur d'une machine asynchronen /

Modèle temps réel d'un estimateur de vitesse à base de réseaux de neurones pour une machine asynchrome /

Electric Machines and Drives

Contact Info

Product

Resources

About