Abstract:This article proposes a multi-objective optimization method for the optimization design of a new dual-stator arc permanent magnet machine (DS-APMM) which can be applied on the direct-drive scanning systems with limited angular movement, such as radar, large telescope. The proposed optimization method integrates light gradient boosting machine (LightGBM) with differential evolution algorithm (DEA) to achieve optimal design objectives of high back electromotive force, low total harmonic distortion, high average torque, and low torque ripple at different rotor speeds. The machine topology and analytical model of DS-APMM are firstly presented to determine the structural parameters to be optimized. The sensitivity of each structural parameter to the optimization objectives is analyzed based on the SHAP (SHapley Additive exPlanations) value. Then, a finite-element analysis (FEA)-based DS-APMM model is developed to acquire sample data. Based on the acquired sample data, a machine learning algorithm, LightGBM, is introduced to establish surrogate model that can fit the function relationship between design objectives and structural parameters. Subsequently, an intelligent search algorithm named DEA is adopted to search optimal combination of the structural parameters and hence obtain optimal machine performances of DS-APMM. Finally, the electromagnetic characteristics of the initial model, middle model and optimal model of DS-APMM are compared and analyzed, both FEA and prototype experiments verify the feasibility and superiority of the proposed multi-objective optimization method.

Multi-Objective Optimization Design of PMECD by Multiple Population Genetic Algorithm

Simulation and Optimization of a Permanent Magnet for Small-sized MRI by Genetic Algorithm

Multi-field Coupling Model of Embedded Giant Magnetostrictive Components Optimization

Optimization Design of a Permanent Magnet Synchronous Generator for a Potential Energy Recovery System

A Hybrid Multi-Objective Optimization Method and Its Application to Electromagnetic Device Designs

Hybrid Multiobjective Optimization Algorithm for PM Motor Design

Multi-Objective Optimization Design of the External Rotor Permanent Magnet-Assisted Synchronous Reluctance Motor Based on the Composite Algorithm

Nonlinear Analytical Model-Embedded Optimal Design for ER-PMSM

Multi-Physical Field Optimization Analysis Of High-Speed Permanent Magnet Synchronous Motor Based On Nsga-Ii Algorithm

Kriging-Assisted Multiobjective Optimization of Embedded Magnetic Encoder in PM Synchronous Machines

Optimization for Giant Magnetostrictive Actuator Based on Genetic Algorithm

Design and Analysis of an Interior Permanent Magnet Synchronous Motor for a Traction Drive Using Multiobjective Optimization

Research on optimal design of wheel permanent-magnet motor based on the Tabu search

Multi-Objective Optimal Design of High-Speed Surface-Mounted Permanent Magnet Synchronous Motor for Magnetically Levitated Flywheel Energy Storage System

An Efficient Multiobjective Optimizer Based on Genetic Algorithm and Approximation Techniques for Electromagnetic Design

Multi-Objective-Layered Optimization of a Magnetic Planetary Gear for Hybrid Powertrain

Analysis and Research on No-Load Air Gap Magnetic Field and System Multi-Objective Optimization of Interior PM Motor

LightGBM Technique and Differential Evolution Algorithm-Based Multi-Objective Optimization Design of DS-APMM

Multiobjective Optimization of the PMSM With the Same Number of Poles and Slots Considering Dynamic Response and Torque Performance

A Multi-Objective Topology Optimization Methodology and Its Application to Electromagnetic Actuator Designs

Multi-Objective Optimization Design and Analysis of V-Shape Permanent Magnet Synchronous Motor