9 MULTI-OBJECTIVE OPTIMIZATION OF A BLDC

Many applications can not be realised with standard sized motors, so there is a need for a application specific motors. In most of the applications there are stringent conditions on the motor, viz. efficiency, cost, size etc. As a result of this the designer is faced with a task to design a motor that meets most these requirements. Thus the motor design becomes a typical case of multi-objective optimisation. When a multi-objective problem is treated, quite often objective conflict with each other and, unlike a single objective optimisation, the solution to this problem is not a single point, rather it has a family of solutions known as Pareto optimal set. Among these solutions, the designer selects the best compromise taking into account higher level information available. In this article we discuss the multi-objective optimisation of permanent magnet brushless DC (BLDC) motors using genetic algorithms. 1