Minimal Power Loss Control for Doubly Salient Electromagnetic Machine Based on Optimal Currents Distribution

The traditional doubly salient electromagnetic machine (DSEM) system sets the field current to the rated value regardless of speed or load torque conditions, bringing in large power losses, especially under less rated power conditions. To overcome this shortcoming, this article proposes a minimal power loss control for DSEM based on optimal current distribution. First, based on the power loss calculation model, the quantified relationships between the power loss and field current are obtained under certain speed and load torque conditions. Then, to minimize the power loss, a distribution strategy of field current and armature current is put forward, the command value of field current is achieved according to system operating conditions, and an easy-implemented torque observer with high accuracy is designed for identifying the load torque, in which the magnetic saturation is taken into consideration. Further, to obtain the optimal field current value more easily rather than looking up the large memory-consumed 3-D table or directly solving the sixth-order equation, the current distribution strategies based on an ergodic algorithm and a back propagation (BP) neural network algorithm are proposed, which achieve online calculation of the optimal current and their performances are comparatively explored. With the decreased field current, less power loss as well as smaller cogging torque ripples are desired to be achieved. The simulation and experiments verify the correctness and feasibility of the proposed strategies under multiple operating conditions.