Phase flux linkage estimation of external rotor switched reluctance motor with NARX neural network

Aydemir M., Okumus H. İ.

ELECTRICAL ENGINEERING, cilt.1, sa.1, ss.1-11, 2023 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 1 Sayı: 1
  • Basım Tarihi: 2023
  • Doi Numarası: 10.1007/s00202-022-01726-x
  • Dergi Adı: ELECTRICAL ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex, INSPEC, DIALNET
  • Sayfa Sayıları: ss.1-11
  • Anahtar Kelimeler: NARX NN (nonlinear autoregressive network with external input neural network), Artificial neural network (ANN), Switched reluctance motor (SRM), Flux linkage estimation, DESIGN
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

In this study, the stator phase flux linkage of a switched reluctance motor was estimated using a nonlinear autoregressive network with external input (NARX) neural network (NN). The application of artificial neural network (ANN) technique for phase flux linkage estimation yielded satisfactory results and the online phase resistance variation compensation requirement, which is an important problem in classical flux linkage calculation methodology, was eliminated. Using the NARX neural network model, which was trained with a set of experimental data, the phase flux linkage value was estimated by the NN from the phase current measurements during real-time operation. The stator phase current values measured under different speed and load conditions together with offline calculated flux linkage values composed the training dataset. The training data preparation methodology was explained in section 2. It was followed by an explanation of the NN training process and the training results were given. The NN based phase flux linkage estimation technique was compared with other techniques such as finite element analysis, torque balance method, voltage balance method in section 7 experimentally.