Influence of the magnetic model accuracy on the optimal design of a car alternator

• J. Cros ^[1] ; L. Radaorozandry ^[1] ; J. Figueroa ^[1] ; P. Viarouge ^[1]
  1. [1] Université Laval
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 27, Nº 1, 2008, págs. 196-204
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – The machine design with optimization method using analytical models is efficient to evaluate a large number of variables because these models are faster to solve. Nevertheless, the validation of the final optimal solution by FE simulation often shows that some specification constraints are not verified. To solve the problem, it is possible to apply a hybrid approach for the design method while combining analytical methods and 3D FE simulations to compensate analytical model errors. The paper addresses this.

    Design/methodology/approach – Each intermediate optimal solution is evaluated by FE simulation to quantify the analytical model errors. Correction coefficients are derived from this evaluation and another optimization process is performed. With this method, the convergence of the hybrid optimal design process is obtained with a limited number of FE simulations.

    Findings – This study shows that it is possible to compensate errors of analytical models with a limited number of 3D field calculations during a global optimization design process. The 3D FE software validates the optimal solution but this solution is also a function of the sensitivity of analytical models that is not improved by the correction method.

    Practical implications – This error compensation of analytical models using FE simulations can be applied for the design of a wide range of electromagnetic devices with optimization methods.

    Originality/value – This paper presents a correction method that guaranteed the validity of the solution after the optimization process when analyzed with a FE software.