A sliding-mode-based control scheme with integral action for combined speed and direct thrust force control of a linear permanent-magnet synchronous motor is proposed. A nonlinear state-space model for the combined dynamics of speed and thrust force as system states is utilized for the synthesis of the sliding-mode control law. Direct integral action is also included in the control law to eliminate the steady-state error in the speed tracking. The sensorless speed estimation is performed by using an adaptive flux observer with a modified dual boundary layer sliding-mode component. Lyapunov stability analysis to prove the global asymptotic stabilities of both the controller and observer is provided. The effectiveness of the proposed method is validated experimentally and demonstrates excellent transient and steady-state speed control performance.

A combined speed and direct thrust force control scheme in the stator flux reference frame is proposed for a linear permanent magnet synchronous motor. First order stator flux dynamics, and a 2nd order non-linear state space model for the combined dynamics of speed and thrust force as system states are presented . These dynamic models are then utilized for the two sliding mode control laws for flux and combined speed and thrust force control. These sliding mode control laws produce the command voltages in the stator flux reference frame which are then applied to the machine by a space vector pulse width modulated inverter. Importantly, integral action is also directly included in the control laws by using a modified first order plus integral sliding condition. The integral action eliminates the steady-state error and provides additional restoring effort in the stator flux and speed tracking combined with thrust force control. Lyapunov stability analysis proves the stability of the proposed control scheme. The effectiveness of the proposed method is validated by extensive experimentation on a prototype linear machine. Practical results demonstrate excellent transient and steady-state speed control performance of the proposed scheme when compared to the state of the art method.

Torque density of multiphase PMSMs can be enhanced by using current harmonic injection (CHI) techniques that utilize potential torque-producing nonfundamental spatial harmonics of the stator winding function (SWF) to produce an additional average torque. This possibility is ignored in the existing multiphase winding design methods that only aim at maximizing the fundamental spatial SWF harmonic, thus neglecting the remaining torque-producing spatial harmonics. These methods are also restricted by the number of machine phases. Motivated by these drawbacks in the existing methods, this paper proposes a novel heuristic algorithm for designing symmetrical multiphase concentrated windings with maximized average torque by considering all the nonfundamental torque-producing spatial SWF harmonics. The proposed approach is valid for any number of phases. Accordingly, an index is proposed to evaluate the multiphase winding performance in terms of its torque density under the CHI technique. The proposed method is formulated by using analytical treatment of the SWF, based on which, a systematic approach for calculating harmonic winding factors and evaluating the average torque is synthesized. The proposed approach is validated through case studies and experimental tests.

This paper proposes a zero sequence current control strategy that allows continuous operation of the three-phase dual-inverter drive using an open-end winding PMSM machine when freewheel diodes (or semiconductor switches) in either inverter fail open-circuit. With an identified failure present, the proposed control strategy injects zero sequence current to bias the phase currents of the faulted phase(s) such that they are unipolar thereby eliminating current flow in the device(s) that have failed. The first mode injects the minimum required zero-sequence current to bias the faulted phase to be unipolar. Hence the injected zero sequence current varies over a fundamental cycle. This reduces the total rms currents in the phases yielding higher torque in fault conditions. The second mode injects a zero sequence current that has a constant value over a fundamental cycle to bias all the currents appropriately. Additional thermal and current stresses result from both techniques and these are assessed and compared. A detailed electromagnetic torque model is also developed to demonstrate the zero sequence current does not affect the average torque output. Experimental results are used to compare the advantages of the techniques and show the improvement in the maximum available torque under fault conditions