Permanent magnet machine design trade-offs to achieve sensorless control at high load

The purpose of this paper is to introduce a new design optimization technique for a surface mounted permanent magnet (SMPM) machine to increase sensorless performance at high loadings by compromising with torque capability.

An SMPM parametric machine model was created and analysed by finite element analysis (FEA) software by means of the Matlab environment. Eight geometric parameters of the machine were optimized using genetic algorithms (GAs). The outer volume of the machine, namely copper loss per volume, was kept constant. In order to prevent sensorless performance loss at high loading, an optimization process was realized using two loading stages: maximum torque with minimum ripple at nominal load and maximum self-sensing capability at twice load. In order to show the effectiveness of the proposed technique, the obtained results were compared with the classical one-stage optimization realized for each loading condition separately.

With the proposed technique, fairly good performance results of the optimization were obtained when compared with the one-stage optimizations. Using the proposed technique, sensorless performance of the motor was highly increased by compromising torque capability for high loading. Additionally, this paper shows that the self-sensing properties of a SMPM machine should be considered at the design stage of the machine.

In related literature, design optimization studies for the sensorless capability of SMPM motor are very few. By increasing optimization performance, new proposed technique provides to achieve good result at high load for sensorless performance compromising torque capability.