Title :
A novel sequential strategy for multivariable optimization in designing of PM motor
Author :
Liu, X. ; Fu, W.
Author_Institution :
Electr. & Electron. Eng., Hong Kong Ploytechnic Univ., Hong Kong, China
Abstract :
This paper presents a novel sequential strategy which is based on finite element method (FEM) and optimization methodology. It is proposed to deal with the multivariable designing of the permanent magnet (PM) motor. In this paper, the property variables are divided into groups by a key factor, each group only involves one or two variables. Then the optimization process will be obtained sequentially by the groups. After obtains the optimal result for one group of property variables, next group of variables will be employed in the optimization. This process will continue until the designing maximizes its power efficiency. This method distributes the multivariable optimization into single variable and bi-variables optimizations. It can reduce the computing complexity and save a lot of working time. In order to check the efficiency of this method, this project compares this new strategy with a normal optimal FEM method.
Keywords :
finite element analysis; particle swarm optimisation; permanent magnet motors; permanent magnets; bivariable optimization; finite element method; multivariable optimization; permanent magnet motor; power efficiency; sequential strategy; single variable optimization; Analytical models; Computational modeling; Finite element analysis; Numerical models; Optimization; Particle swarm optimization; Permanent magnet motors; Multivariable optimization; particle swarm optimization; permanent magnet motor; random search; sequential strategy;
Conference_Titel :
Magnetics Conference (INTERMAG), 2015 IEEE
Conference_Location :
Beijing
Print_ISBN :
978-1-4799-7321-7
DOI :
10.1109/INTMAG.2015.7156997