Fault-tolerant control of a five-phase (5Ph) permanent magnet (PM) motor has been recently widely studied, however decoupled modeling under faulty conditions is discussed in less details. In this paper, decoupled model of the 5Ph PM motor under single-phase open fault is investigated and based on the proposed model field oriented control (FOC) is applied to the motor. The proposed model is based on the concept of preserving values of fundamental magnetic motive force (MMF) and back electromotive force (EMF) under single-phase open fault, the same as in healthy case, along with keeping equal torque equation at fundamental rotating space. Nevertheless, the output torque still presents some noises due to third harmonics of air-gap flux and system uncertainties, and to cope with those disturbances, sliding mode control (SMC) is proposed in the speed loop to improve the speed performance. The proposed SMC contains a PI controller and a chattering term, and thus it can be easily tuned. The decoupled model is verified by transient finite element analysis finite element analysis (FEA) and further experiment results are presented to confirm the effectiveness of the proposed control strategy.

This paper investigates torque ripples cancellation of a five-phase (5Ph) permanent magnet (PM) motor under open fault of two phases, where failure phases are either adjacent or nonadjacent. In recent works, torque ripples cancellation is performed at a stationary frame and postfault current solutions for constant output torque are complicated. To implement field-oriented control (FOC) and carrier-based pulse width modulation, reduced-order Clarke and Park transformations are proposed. With the coordinate transformation, a 5Ph motor model under unbalanced conditions has been decoupled. Along with this, the fundamental torque at the d-q frame also remains the same as the healthy case. However, third harmonic of winding density distribution results in torque ripples. To reduce dependence on mathematical model of torque ripples, sliding mode control is employed to calculate q -axis current reference while keeping constant the rotor speed. Decoupled models corresponding to two types of faults, which are adjacent and nonadjacent open fault, are confirmed by finite element analysis, and effectiveness of proposed torque ripples cancellation method is verified on the 5Ph drive in the laboratory.

This letter presents a fast open circuit diagnostic method for modular multilevel DC/DC converter (MMDDC), which is applied in super capacitor energy storage system. The proposed method is derived based on the analysis of submodule output characteristics under both normal and fault conditions. For MMDDC, different types of open circuit fault are reflected by the output voltage of submodule in certain operating modes, thus a simple and effective diagnostic method by the aid of a hardware circuit and a fast fault location algorithm is proposed. The hardware circuit is responsible for detecting submodule output voltage and its output result is transmitted to a processor to accomplish the final fault diagnosis along with drive signals. The proposed approach avoids complex mathematical operations and minimizes fault diagnostic time; moreover, it can achieve fault diagnosis under a wide super capacitor operating voltage range. Finally, the proposed diagnostic method is verified experimentally based on a laboratory-scale MMDDC energy storage prototype.

This paper presents an active state-of-charge (SOC) balancing control strategy for modular super capacitor energy storage system (ESS). The strategy has a master-slave structure, including a common DC bus voltage loop and individual slave current loop for each sub-module. The common voltage loop ensures the ability of the system to stabilize DC bus voltage cannot be affected by SOC balancing control, and provides a same original current reference for all sub-modules. SOC information is introduced into current loops to achieve SOC balancing control. The highest SOC is taken as a reference for low SOC modules, thus, the low SOC modules can regulate the sub-module operation average current according to the reference and their own SOC. In order to minimize system current deviation caused by the introduction of SOC, a current correction method is proposed to improve system current regulate performance. The proposed strategy can achieve SOC balancing control during ESS charging and discharging process dynamically, in addition, the modularity of the system is improved and it has favorable stability. At last, a modular ESS prototype consisting of three half bridge sub- modules is tested in the laboratory, and the experimental results verify the effectiveness of the proposed control strategy.

In this paper, a simple digital hysteresis current control (DHCC) with quasi-constant switching frequency (SF) for the current control of a single-phase grid-connected inverter (SPGI) is proposed. The hysteresis band is cancelled, and the SF is fixed by online adjusting the sampling period. The principle of the DHCC with constant sampling period is analyzed, and the distribution of the switching period (SP) corresponding to the phase angle of the grid is derived. The rule of online adjusting the sampling period to fix SF is then determined according to the former theoretical analysis results. Furthermore, the effect of variation of the filter inductor on the SP is analyzed. An online estimation of the filter inductor based on the linear model of the current increment within one switching cycle is developed to eliminate this kind of effect. The proposed scheme shows the features of simple control, easy implementation in a digital way, robust to the filter inductor, and excellent steady and dynamic performance. The detailed simulation and experimental results verify its accuracy and feasibility.

In this brief, a new two-phase stationary-frame-based enhanced phase-locked loop (TPSF-EPLL) is proposed for detecting the amplitude, phase angle, and frequency of a three-phase grid. The differential equations of the proposed TPSF-EPLL are originally derived based on minimizing the cost function with a gradient descent algorithm. Moreover, the system structure, the design guidelines of each parameter in both frequency and amplitude loops, and the stability analysis of the TPSF-EPLL are presented. The main advantage of the TPSF-EPLL is that it requires less calculation since transformation into a rotating reference frame is not involved, and all the calculation is performed in a two-phase stationary frame. Experiment results of steady state and dynamic performance of TPSF-EPLL are presented and compared with some classical PLLs to verify the validity of the theoretical analysis and its advantages.