Improvement of the power to weight ratio for an induction traction motor using design of experiment on neural network


DEMİR U.

Electrical Engineering, cilt.103, ss.2267-2284, 2021 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 103
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1007/s00202-020-01204-2
  • Dergi Adı: Electrical Engineering
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex, INSPEC, DIALNET
  • Sayfa Sayıları: ss.2267-2284
  • Anahtar Kelimeler: Traction motor dynamics, Power to weight ratio, Electric vehicle, Induction motor, Design of experiment, Neural networks, OPTIMIZATION
  • Marmara Üniversitesi Adresli: Evet

Özet

© 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH, DE part of Springer Nature.This paper proposes the traction motor analysis and weight reduction for an electric golf car with 1 + 5 passengers, because the power to weight ratio is the critical parameter for the electric vehicle in terms of the vehicle performance. Firstly, the induction motor (IM) already used in the electric golf car is tear-downed and modelled in a simulation environment, and the model is verified by the test results given in the datasheet to prove the accuracy of the model used in the design. Secondly, the vehicle dynamics of the electric golf car are investigated and the traction requirements are determined in the simulation environment. The design parameters of the IM model are investigated by using Taguchi’s design of experiment (DoE) method, and the critical design parameters are specified. Then, a neural network (NN) to predict better design parameters is trained to operate with the DoE model according to the prioritization of framework. Finally, the study is completed by comparing the obtained results of the NN-predicted IM model, DoE best-case IM model and the original IM model in terms of the battery consumption, power to weight ratio, efficiency and vehicle traction requirements.