IM to IPM design transformation using neural network and DoE approach considering the efficiency and range extension of an electric vehicle

DEMİR U.

ELECTRICAL ENGINEERING, cilt.104, sa.2, ss.1141-1152, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 104 Sayı: 2
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1007/s00202-021-01378-3
  • Dergi Adı: ELECTRICAL ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex, INSPEC, DIALNET
  • Sayfa Sayıları: ss.1141-1152
  • Anahtar Kelimeler: Neural networks, Parameter fitting, Design of experiment, Electric vehicle, IPM motor, OPTIMIZATION, MOTOR
  • Marmara Üniversitesi Adresli: Evet

Özet

This paper is regarding to the design transformation methodology from induction motor (IM) to interior permanent magnet motor (IPM) to increase the efficiency of the traction motor and extend the range of the electric vehicle. Furthermore, the rotor structure of IPM is tried to be optimized without changing the stator and winding structure of the IM from the previous study to meet the vehicle requirements and maximize the motor efficiency in order to design the rotor structure of the IPM. Firstly, the design parameters, the vehicle requirements, and constraints are determined for the reference IM model. Secondly, the appropriate rotor geometry is chosen by considering the advantages and disadvantages of different rotor geometries for IPM. Then, the design parameters for the rotor are determined and investigated by using Taguchi's design of experiment (DoE) method considering the importance and priorities of the rotor design parameters of IPM. Thirdly, a neural network (NN) to predict better design parameters is trained to operate with the data of DoE according to the prioritization and normalization of framework. Finally, the reference IM and the predicted IPM models are evaluated in terms of the acceleration, the slope climbing, the driving performance (ECE R15), and battery consumption characteristics.