Recently, the control of multiphase electric drives has been a hot research topic due to the advantages of multiphase machines, namely the reduced phase ratings, improved fault tolerance and lesser torque harmonics. Finite control set model predictive control (FCS-MPC) is one of the most promising high performance control strategies due to its good dynamic behaviour and flexibility in the definition of control objectives. Although several FCS-MPC strategies have already been proposed for multiphase drives, a comparative study that assembles all these strategies in a single reference is still missing. Hence, this paper aims to provide an overview and a critical comparison of all available FCS-MPC techniques for electric drives based on six-phase machines, focusing mainly on predictive current control (PCC) and predictive torque control (PTC) strategies. The performance of an asymmetrical six-phase permanent magnet synchronous machine is compared side-by-side for a total of thirteen PCC and five PTC strategies, with the aid of simulation and experimental results. Finally, in order to determine the best and the worst performing control strategies, each strategy is evaluated according to distinct features, such as ease of implementation, minimization of current harmonics, tuning requirements, computational burden, among others.

Multiphase machines, associated with state-of-the-art control strategies like finite control set model predictive control (FCS-MPC) are suitable for applications where reliability and disturbance-free operation are key factors. Although FCS-MPC provides good transient performance and straightforward design, the performance of the drive system is affected by model inaccuracies due to parameter mismatch and unmodeled dynamics, such as deadtime effects in the power converters and back-electromotive force harmonics. In this context, this article proposes a novel and robust disturbance observer based predictive current control strategy for six-phase permanent magnet synchronous machine drives that minimizes the steady-state errors due to parameter mismatch errors, while simultaneously reducing the amplitude of the fifth- and seventh-order current harmonics due to unmodeled dynamics, which are non-negligible in six-phase machines. Simulation and experimental results demonstrate the very good performance of the proposed strategy even when considering significant errors in the system parameters and different values of the deadtime in the power converters.

The most recent finite control set model predictive control (FCS-MPC) strategies for six-phase electric drives avoid the excitation of the secondary subspace by using virtual vectors to minimize the current ripple. However, these strategies can only control the fundamental current components, which adjust the machine flux and torque and leave the secondary current components unregulated, which should be zero during regular operation in order to reduce the losses in the machine. Due to this limitation and also due to the low impedance of the secondary subspace in machines with distributed windings, the stator currents might contain non-negligible low-frequency harmonics caused by small asymmetries in the machine, non-linear dynamics of the semiconductors in the power converters and back-EMF harmonics. This article proposes a novel multistage predictive current control strategy based on virtual vectors (MVV-PCC) for six-phase drives based on permanent magnet synchronous machines (PMSMs), which contains a fundamental and a secondary control stages that are responsible for tracking the reference currents in both subspaces separately. The obtained experimental results validate the effectiveness of the proposed strategy in the minimization of the current harmonics and the reduction of the torque ripple for different operating conditions of the six-phase drive.

The severe torque ripple normally occurring in the DFIG-dc system can cause premature failure of mechanical components and shorten the life of the drive train. This paper addresses the torque ripple issue by proposing a predictive direct torque control strategy which delivers at the same time torque ripple suppression and minimization of losses. The existing control algorithms for torque ripple mitigation are mostly based on resonant controllers and repetitive control forcing the compensation signal either through the current chain or directly into the rotor voltage commands. All these techniques lead to structures with multiple controllers whose tuning is not straightforward. Furthermore, they are very sensitive to the operating frequency, making optimized operation with variable frequency highly challenging. Conversely, the proposed algorithm predicts directly the best rotor voltage space vector to minimize torque ripple and track a prescribed rotor flux amplitude to minimize losses, with no current control chain. As confirmed by simulations and experiments, the strategy allows large stator frequency variations as required by the optimal flux command for minimum losses, whilst ensuring effective torque ripple compensation.

Even though multiphase machines have inherent fault-tolerant capabilities due to their additional degrees of freedom, fault diagnosis still plays an important role to continuously monitor these machines and, thus, prevent critical failures. High-resistance connection (HRC) is one of the most common electric faults in electrical machines and is caused by damaged power connections. These faults create a current unbalance, increasing both the machine losses and torque pulsations, and can eventually lead to open-phase faults (OPFs). This article proposes an online fault diagnostic method to detect HRCs and OPFs in six-phase permanent magnet machine drives. The proposed method is intended for application in electric drives controlled by a predictive current control strategy and does not require the installation of extra hardware. Two variants of the fault diagnostic algorithm are presented in this work: one based on the analysis of the secondary components of the reference current errors and the other variant based on the prediction current errors. Experimental results demonstrate the effectiveness and sensitivity of the proposed diagnostic method for the detection of these types of fault.

Owing to the low equivalent impedance of six‐phase machines in the secondary subspace, in which currents do not contribute to torque production, low‐order current harmonics with high amplitude are likely to appear in the stator currents, increasing the harmonic distortion and the losses in the machine. In the case of permanent magnet synchronous machines (PMSMs), the main causes of these harmonics are machine asymmetries, non‐linearities of the power converters and back‐electromotive force harmonics. The existing predictive control strategies based on virtual vectors for six‐phase machines only regulate the currents mapped into the fundamental subspace, responsible for flux/torque production and leave the currents mapped into the secondary subspace uncontrolled. This study proposes a novel bi‐subspace predictive control strategy for six‐phase PMSMs with optimal amplitude virtual vectors, which is able to effectively regulate the currents in both subspaces simultaneously, leading to a significant minimisation of the current harmonic distortion. Moreover, the proposed control strategy also allows operating the six‐phase PMSM with a current unbalance between the two sets of windings. Both simulation and experimental results confirm the much better performance of the proposed strategy in comparison with other state‐of‐the‐art predictive control strategies found in the literature.