Speed sensorless direct torque control of brushless direct current motors with adaptive filter based MRAS

Canberk Tuzcu; Engin Cemal Mengüç; Rıdvan Demir; Remzi İnan

Makale detayları

Speed sensorless direct torque control of brushless direct current motors with adaptive filter based MRAS

Yazarlar:

Canberk Tuzcu | Engin Cemal Mengüç | Rıdvan Demir | Remzi İnan

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Elektrik Makineleri ve Sürücüler

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 2, pp.680-687

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345097

pISSN:

2564-6605

Anahtar kelimeler:

Özet:

 This study introduces model reference adaptive system MRAS estimators based on least mean square LMS, least mean kurtosis LMK, and least mean fourth LMF approaches for estimating the rotor speed required in a permanent magnet brushless direct current motor SMFDAM drive. The proposed MRAS estimators directly estimate the rotor speed by minimizing the error term between the measured stator currents, serving as the reference model, and the stator currents generated at the adaptive models output. Additionally, since the weight vectors, which encapsulate rotor speed, are updated adaptively at each sampling step in all three estimators, the necessity for a fixed-gain proportional-integral controller commonly employed in conventional MRAS frameworks is removed. The performances of the proposed estimators are assessed through a direct torque control DTC-based SMFDAM drive under demanding operating scenarios. Simulation results show that the performances of the proposed estimators are alternative to each other. In particular, while the LMF-based MRAS structure performs slightly better than the other MRAS structures in speed estimation, the LMS and LMK-based MRAS structures provide better performances in 3-phase stator current estimation.

 This study introduces model reference adaptive system MRAS estimators based on least mean square LMS, least mean kurtosis LMK, and least mean fourth LMF approaches for estimating the rotor speed required in a permanent magnet brushless direct current motor SMFDAM drive. The proposed MRAS estimators directly estimate the rotor speed by minimizing the error term between the measured stator currents, serving as the reference model, and the stator currents generated at the adaptive models output. Additionally, since the weight vectors, which encapsulate rotor speed, are updated adaptively at each sampling step in all three estimators, the necessity for a fixed-gain proportional-integral controller commonly employed in conventional MRAS frameworks is removed. The performances of the proposed estimators are assessed through a direct torque control DTC-based SMFDAM drive under demanding operating scenarios. Simulation results show that the performances of the proposed estimators are alternative to each other. In particular, while the LMF-based MRAS structure performs slightly better than the other MRAS structures in speed estimation, the LMS and LMK-based MRAS structures provide better performances in 3-phase stator current estimation.

 This study introduces model reference adaptive system MRAS estimators based on least mean square LMS, least mean kurtosis LMK, and least mean fourth LMF approaches for estimating the rotor speed required in a permanent magnet brushless direct current motor SMFDAM drive. The proposed MRAS estimators directly estimate the rotor speed by minimizing the error term between the measured stator currents, serving as the reference model, and the stator currents generated at the adaptive models output. Additionally, since the weight vectors, which encapsulate rotor speed, are updated adaptively at each sampling step in all three estimators, the necessity for a fixed-gain proportional-integral controller commonly employed in conventional MRAS frameworks is removed. The performances of the proposed estimators are assessed through a direct torque control DTC-based SMFDAM drive under demanding operating scenarios. Simulation results show that the performances of the proposed estimators are alternative to each other. In particular, while the LMF-based MRAS structure performs slightly better than the other MRAS structures in speed estimation, the LMS and LMK-based MRAS structures provide better performances in 3-phase stator current estimation.

 Bu çalışma, kalıcı mıknatıslı fırçasız doğru akım motoru SMFDAM sürücüsünde gereken rotor hızını tahmin etmek için en küçük ortalama kare Least mean square, LMS, en küçük ortalama kurtosis LMK ve en küçük ortalama dördüncü Least mean fourth, LMF yaklaşımlarına dayalı model referanslı adaptif sistem Model reference adaptive system, MRAS tahmin edicilerini tanıtmaktadır. Önerilen MRAS tahmin edicileri, referans modeli olarak hizmet eden ölçülen stator akımları ile adaptif modelin çıkışında üretilen stator akımları arasındaki hata terimini minimize ederek rotor hızını doğrudan kestirmektedir. Ayrıca, rotor hızını kapsayan ağırlık vektörleri üç tahmin edicide de her örnekleme adımında adaptif olarak güncellendiğinden, geleneksel MRAS çerçevelerinde yaygın olarak kullanılan sabit kazançlı orantılı-integral bir denetleyiciye olan ihtiyacı ortadan kaldırmaktadır. Önerilen tahmin edicilerin başarımları, zorlu çalışma senaryoları altında doğrudan moment kontrolü DMK tabanlı bir SMFDAM sürücüsü aracılığıyla değerlendirilmiştir. Benzetim sonuçları, önerilen kestiricilerin başarımlarının birbirlerine alternatif olduğunu göstermiştir. Özellikle, hız kestiriminde LMF tabanlı MRAS yapısı diğer MRAS yapılarından bir miktar daha iyi başarım sağlarken, 3-faz stator akım kestiriminde LMS ve LMK tabanlı MRAS yapıları daha iyi başarımlar sağlamıştır.

Makale detayları

Speed sensorless direct torque control of brushless direct current motors with adaptive filter based MRAS

Yazarlar:

Canberk Tuzcu | Engin Cemal Mengüç | Rıdvan Demir | Remzi İnan

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Elektrik Makineleri ve Sürücüler

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 2, pp.680-687

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345097

pISSN:

2564-6605

PMBLDC

DTC

Least mean square

Least mean kurtosis

Least mean fourth

Model reference adaptive system.SMFDAM

DMK

En küçük ortalama kare

En küçük ortalama kurtosis

En küçük ortalama dördüncü

Model referans adaptif sistem.

Sayı

Makale detayları

Yazarlar: Canberk Tuzcu | Engin Cemal Mengüç | Rıdvan Demir | Remzi İnan

Yayıncı: Nigde Omer Halisdemir University

Ana Başlıklar: Elektrik Makineleri ve Sürücüler

Dergi: Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 2, pp.680-687

Belge Tipi: Makale

Yayın Tarihi: 4/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001345097

pISSN: 2564-6605

Sayı

Yazarlar:

Canberk Tuzcu | Engin Cemal Mengüç | Rıdvan Demir | Remzi İnan

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Elektrik Makineleri ve Sürücüler

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 2, pp.680-687

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345097

pISSN:

2564-6605