De Cecco, Eric, Marchand, Claude, and Besbes, Mondher
Subjects
*COMPUTER software, *MATHEMATICAL optimization, *MAGNETICS, *FINITE element method, *NUMERICAL analysis, *SYNCHRONOUS electric motors
Abstract
The aim of this paper is to describe a software tool for conception and optimisation of permanent magnet synchronous machines associated to their converters, using a coupled model for magnetic field and electrical circuit equations. In this paper, the emphasis is on the specifications of the tool: modularity, adaptability and evolutive capacity. [ABSTRACT FROM AUTHOR]
*MATHEMATICAL optimization, *FINITE element method, *NUMERICAL analysis, *MAGNETOSTATICS, *ELECTROMAGNETISM
Abstract
In the paper the method of gradient balance is revisited from microeconomics and proposed in electromagnetics as an alternative formulation of a design optimisation problem, characterized by two objective functions; this formulation exploits the design space directly. Accordingly, a numerical procedure to approximate a non-dominated solution was implemented; after validating the numerical procedure by means of an analytical benchmark, a field-based case study was developed. [ABSTRACT FROM AUTHOR]
Permanent magnet motors have been widely used. In recent years, from the point of view of energy saving, it is required to improve the efficiency of permanent magnet motors by applying an optimal design method. In this paper, we propose an adjusting technique of the supply voltage of permanent magnet motor when carrying out the optimal design by adopting voltage driven finite element method and genetic algorithm. In calculation for the optimal design, it is difficult to keep a desirable balance of the supply voltage and the induced electromotive force. By using the proposed method, the supply voltage is appropriately decided to generate the rated torque. Owing to adopting this adjusting technique, the computation time for the optimal design can be reduced and the solution can escape from local minima because the needless computations can be drastically reduced. [ABSTRACT FROM AUTHOR]