Neuro fuzzy modeling of axial field machines behavior

• Autores: A. Al-Badi, K. El‐Metwally, A. Gastli
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 26, Nº 2 (Selected papers from the 9th Workshop on Optimization and Inverse Problems in Electromagnetism, Sorr), 2007, págs. 407-417
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – This paper aims to study modeling of the nonlinear behavior of the Torus machine back EMF using an adaptive networks fuzzy inference system (ANFIS). The model can be used to study the steady‐state as well as the dynamic performances of the machine operating as a motor or as a generator.

    Design/methodology/approach – Using the universal approximation capability of fuzzy systems the authors designed an ANFIS network to model the nonlinear behavior of the back EMF of the Torus motor. The ANFIS is trained using an actual set machine measurements data to generate the motor back EMF for different operating conditions.

    Findings – Simulation results of the ANFIS model of the Torus motor at different loads proved the ability of the algorithm to effectively model the complex electromagnetic behavior of the machine. Such efficient modeling can directly help in improving and optimizing the Torus motor drive system design.

    Originality/value – It demonstrates that ANFIS can model the nonlinear behavior of the back EMF of the Torus motor with excellent accuracy.