In this report, the feasibility of predicting mechanical properties using magnetic parameters measurements and artificial neural network (ANN) will be presented. The yield and ultimate tensile strength are predicted by means of two back-propagation neural networks on the basis of hysteresis loop parameter measurements and sample thickness. Inductive measurements are carried out with exciting and pickup coils attached to the magnetizing yoke. Therefore, remanence (B r ), coercive force (H c ), the hysteresis loss (W h ), maximum relative differential permeability ( max r μ ) and harmonic components of the field and flux density extracted and used as input parameters for training neural network. The individual influence of several input parameters are shown and compared with metallurgical phenomena. The ANN, shown good performance and the results are in agreement with the experimental results. The developed model can be used as an on-line non-destructive evaluation technique.
Abstract-In this paper, a novel technique for improving the torque characteristics of the Interior Permanent Magnet Synchronous Motor is proposed using rotor shape optimization. The main objective is to decrease the torque ripple while increasing average torque. The improvement process is performed for the maximum torque-angle operating point, and then studies are carried for other currents and angles. Defining a multi-element grid on rotor surface regions in which each element could be either iron or air, the best practical rotor surface topology could be obtained to improve the overall torque characteristics of IPMSM. The best motor performance is achieved using practical rotor shapes obtained from a cluster of points in average torque versus torque ripple plane. Finally, for torque ripple cancellation, two or three alternate rotor configurations with optimized average torque and out of phase torque pulsation have been selected. This selection will guarantee improved average torque while mitigating torque pulsation by a significant margin. Using this method, a rotor topology obtained in which torque ripple is reduced by 80% with slightly improved average torque.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.