ESTIMATION OF OVERLOAD STAGES IN INDUCTION MOTORS USING BY DISCRIMINANT ANALYSIS (DA)


Creative Commons License

Noğay H. S.

6. ULUSLARARASI MÜHENDİSLİK, MİMARLIK VE TASARIM KONGRESİ, İstanbul, Turkey, 17 - 18 December 2020, pp.685-692

  • Publication Type: Conference Paper / Full Text
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.685-692
  • Kayseri University Affiliated: Yes

Abstract

In order to decide at which load level and with which operating parameters the induction motors fed by the drive will operate with the most optimal efficiency and loss, it is necessary to observe how the structural and supply parameters of the induction motor change at which load level. As the load level increases in an induction motor, the internal and external temperature of the motor increases depending on the load current and the winding structure. This increase is also directly related to the increase in load current resulting from the increase of the load level. The load current increments at different inverter switching frequencies and different winding structures in induction motors powered by a driver, differ characteristically. Being able to detect these differences is extremely important to rotary electrical machine designers. In this study, a prediction model is designed to determine the different load levels (light, half, full, half excess and full overload) of a three-phase cage induction motor fed by a PWM inverter, using discriminant analysis (DA) method, depending on the operating conditions of the induction motor. The designed DA prediction model was trained with the data set obtained by experimental applications and tested with ten-fold cross validation technique. In addition to the highly satisfactory accuracy rates achieved in the study, it was concluded that DA prediction models can be used very effectively in the design of rotary electrical machines.