| Autor: |
Kim, Hyeon-Jun, Baek, Soo-Whang |
| Předmět: |
|
| Zdroj: |
Microsystem Technologies; Aug2023, Vol. 29 Issue 8, p1165-1179, 15p |
| Abstrakt: |
This study proposes an optimal design using artificial neural network models to improve the efficiency of a 15 kW-class interior permanent magnet synchronous motor (IPMSM) used in micro electric vehicles (EVs). The proposed method improves the efficiency of the initial IPMSM model by combining the number of poles and slots as well as the shape design of the rotor. The number of poles used and the combination of slots were determined according to the purpose of the micro-EV motor, the windings of the stator were changed from distributed winding to concentrated winding, and the dimensions of the rotor were selected to maximize efficiency. We realize the optimal design was performed with using a model with having a structure of a concentrated winding stator 18-slot rotor12-poles in the IPMSM having with a distributed winding stator 48-slot rotor 8-pole structure as an initial model. The objective function and constraints were determined for the optimal design and as a metamodel, we applied recurrent neural network and long short-term memory models. The optimal metamodel was selected by comparing the root-mean-square error test results according to the efficiency and cogging torque characteristics to evaluate its accuracy. We validate the optimal design result derived based on the proposed shape-optimization method through finite element analysis. An optimal model of the 15 kW-class IPMSM prototype used in micro-EVs was designed for validation. The efficiency performance improved by 1.6%, as validated through a motor dynamo test. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink