Multi-Parameter Optimization of Stator Coreless Disc Motor Based on Orthogonal Response Surface Method

Abstract

In response to the structural optimization problem of PCB stator coreless disc motors, the orthogonal response surface method was used to optimize the motor structure, preliminarily determine the basic parameters of the motor, and conduct orthogonal experiments on the motor parameters based on the optimization design objectives. The optimization factors were the quantity of the magnetic pole of the motor rotor p, the ratio of the main/auxiliary pole sizes Rnd, the thickness Tm of the permanent magnet, and the air gap length δ. The motor torque Td, the amplitude of the magnetic density of the air gap Bδ, and the waveform distortion rate THD were used as optimization objectives. The motor parameters that cause the motor torque to reach a maximum cause the air gap magnetic density to reach a maximum and the waveform distortion rate to reach a minimum. Due to the use of PCB plates instead of motor stator cores in the PCB coreless disc motors, the service life of the PCB board during motor operation will be reduced with the temperature increase generated by the stator winding. To solve this problem, a response surface analysis of the motor was carried out to reduce the increase in the temperature of the stator windings during the operation of the motor. The PCB board and stator winding are the main factors affecting the motor’s temperature increase. Taking the thickness of the PCB board, the hole diameter on the board, and the uneven width of the stator winding as optimization factors, the motor parameters with the lowest increase in the temperature of the motor winding were obtained. A simulation analysis was conducted using Ansys/Maxwell software, and the results prove the feasibility of the optimization.

A Data-Driven Motor Optimization Method Based on Support Vector Regression—Multi-Objective, Multivariate, and with a Limited Sample Size

“…The sensitivity analysis test utilizes the orthogonal test method, which is an experimental design approach that examines various factors and levels [11,37]. Signal-to-noise ratio analysis and variance analysis are conducted without the need to cover the entire test space.…”

“…Model-driven modeling relies on fundamental electromagnetic principles to construct motor models [3][4][5][6], while data-driven methods leverage observed motor performance data and employ statistical and machine learning techniques to unveil system patterns and relationships [7][8][9]. Employing data-driven methodologies to establish mathematical models for motors offers advantages such as reduced computational complexity and robust generalization capabilities [5,10,11]. Moreover, data-driven techniques harness machine learning methodologies to better capture the intricacies of complex systems [12].…”