Multiphase electrical machines are inherently fault tolerant due to the higher number of phases. An important step in achieving a fault tolerant control is the ability to detect and identify the fault. In variable phase-pole drives, which are a class of multiphase machines that change the number of pole pairs during real-time operation, there are further unexplored possibilities. Based on the harmonic plane decomposition theory for variable phase-pole machines, a fault detection and identification algorithm is proposed, which analyzes the spatial harmonics of the current distribution. The fault detection is fast, operation independent, non-invasive, parameter-insensitive, and computationally simple. Experimental tests validate the proposed method.

In the future, faster-electrified means of transportation are a necessity for more sustainable technologies. The use of Linear Induction Machines (LIM) poses an important propulsion option for high-speed transportation, particularly in hyperloop applications. Such types of high-power linear machines are complex to design and needs development without compromising too much on efficiency. The reason for the low efficiency of LIMs is high rotor losses and large airgap. Furthermore, a need to estimate the efficiency map during the runtime is necessary as high losses under constrained environments of pressure and temperature leads to significant changes in the total efficiency. Currently, LIM manufacturers do not usually provide efficiency maps with respect to an entire range of torque and speed. Thus, a design methodology and temperature-dependent efficiency maps are proposed for LIM. In accordance with the design requirements of the hyperloop pod in its entirety, a maximum speed of approximately 150km/h is achieved.