M. Pulvirenti

Fault-Tolerant AC Multidrive System

This paper deals with an original control strategy for AC multidrive systems able to mitigate the effects of failures occurring on one or more drives. A key feature of the proposed technique is that fault tolerance capability is achieved by a suitable reconfiguration of the system in order to allow the healthy drives to provide additional paths for the currents of the faulty drives. Moreover, a modified control algorithm able to enforce the vector control in damaged drives has been implemented, by cooperatively managing some or all the drives of the system. Therefore, differently from previous techniques, the fault tolerance capability is achieved by exploiting the healthy drives, rather than activating back-up inverter legs. As a result, no additional high-frequency switching power devices and related drives circuitries are needed. In the following, two different scenarios will be analyzed, highlighting pros and cons of the proposed approach through simulations and experimental results.

A fault-tolerant power conversion topology for PMSG based Wind Power Systems

This work aims to investigate a novel fault tolerant topology for Permanent Magnet Synchronous Generator (PMSG) based Wind Power Systems (WPS), where the static power converter consists of two three-phase inverters connected in back to back configuration. The first inverter is used to control the PMSG while the other one is used to deliver power to the grid through an output filter and a line frequency transformer. Essentially, the fault tolerance is achieved by performing a cooperative control strategy when a faulty condition occurs in one of the two drives. Whenever a fault occurs in one of these two parts, the faulty elements are isolated, the faulty part is controlled in open phase mode and the resulting current flowing out the neutral point is redirected to the healthy part which is so exploited as a feedback current path. The proposed Fault Tolerant WPS (FTWPS) requires very limited modifications to the vector control performed in the PMSG as well as to the active and reactive power control implemented in the grid side, by keeping similar performance with respect to the healthy system. The theoretical analysis is supported by experimental tests, validating the effectiveness of the system configuration.

Current-Sharing Strategies for Fault-Tolerant AC Multidrives

Thanks to an inner redundancy, AC Multi-Drives can be made fault tolerant with minor modifications of their structure. However, some additional components are generally required in order to create emergency current paths in case of a fault. According to a recently presented approach the amount of additional components can be reduced by exploiting healthy drives to support faulty drives. In this case current and power sharing among drives assumes a key role. The paper deals with some new current sharing methodologies for AC multi-drive systems, where a suitable interaction among healthy and faulty drives is exploited, in order to cope with one or more faults. Specifically, three different current sharing control methods are presented and evaluated by means of simulations and experimental tests, taking into account the effects of operations in fault conditions both on healthy and faulty drives.

Space vector modulation technique for common mode currents reduction in six phase AC drives

The paper presents a space vector based modulation technique able to reduce Common Mode Currents (CMC) in six phase integrated AC motor drives. The proposed modulation strategy performs a suitable selection of the inverter states in order to achieve a considerable reduction of the Common Mode Voltage (CMV) variations, with a slightly reduction of the DC bus utilization compared to the standard Space Vector Modulation (SVM) approaches; drives operating with the proposed modulation method still maintain same performance in terms of dynamic control. The entire analysis has been performed on a six phase motor configurations, supplied by a single two level multiphase inverter and the effectiveness of the theoretical study has been verified by simulation results.

Secondary saliencies decoupling technique for self-sensing integrated multi-drives

The paper deals with the study and implementation of a secondary saliencies decoupling technique for self-sensing control algorithms used in multiphase motor drives and based on the injection of an additional high frequency excitation. The proposed approach can be applied to multiphase motors whose stator winding is composed of multiple three phase units suitably spatially shifted. The method exploits the spatial phase shift among the stator units to isolate undesired prominent saliency harmonics. Compared to the state of the art, this approach does not require multiple saliencies models or sophisticated filtering processes, but only the knowledge of the saliency harmonic order to be suppressed. Finite Element Analysis and Experimental results applied to a dual-three phase permanent magnet synchronous motor confirm the effectiveness of the method.

Hall-effect sensor fault detection, identification an compensation in brushless DC drives

This paper investigates Hall-effect sensor faults in brushless dc drives and proposes a very effective methodology for their detection, identification and compensation. It is shown that these faults cause erroneous commutation, generally leading to unstable operation. By using a fault detection and identification technique proposed by the authors in a previous paper related to low cost field-oriented drives, [22], together with appropriate fault-compensated position and speed estimation algorithms, it is shown for the first time that proper operation is guaranteed for both single and double faults. Comparative experimental results are provided for operation with three state of the art, Hall-effect sensor based, estimation algorithms: the zeroth order algorithm, [19], the hybrid observer, [20], and the vector-tracking observer, [21]. It is verified that stable operation is achieved with three, two or only a single Hall-effect sensor functioning correctly.

Fault tolerant AC multi-drive system

The paper deals with an original control strategy for AC Multi-Drive (MD) systems able to mitigate failures on one or more drives, through a minimal system reconfiguration. In particular, fault tolerance capability is achieved by isolating the failed inverter legs and connecting together the neutral point of stator windings of all motors; moreover, control algorithms are modified in order to maintain vector control in faulted drives. Differently from other MD configurations, the proposed solution avoids the connection of the neutral point of the machines to the middle point of the DC bus, thus overcoming the drawbacks related to the DC bus ripple and DC bus voltage utilization. In the following, two different scenarios have been analyzed, and pros and cons of the proposed solution are highlighted through simulations and experimental results.

Fault tolerant capability of deadbeat — Direct torque and flux control for three-phase PMSM drives

This paper investigates the performance of a three-phase permanent magnet synchronous motor (PMSM) drive operating under a single fault, adopting a low cost and fault-tolerant control based on deadbeat-direct torque and flux control (DB-DTFC). Under faulty operation, this fault-tolerant DB-DTFC offers an independent regulation of the electromagnetic torque and the stator flux linkage by using the same torque line equation, stator flux linkage, and current observers adopted during healthy conditions, requiring very limited hardware reconfigurations. In particular, it has been demonstrated that in a fault situation, the same drive model equations adopted for the healthy electric drive can be exploited with very limited detrimental effects on the drive performance simply by applying a suitable reference frame transformation set in the torque and flux control structure. The proposed fault-tolerant DB-DTFC has been validated by experimental tests, confirming that the proposed fault-tolerant DB-DTFC ensures satisfactory faulty operations and drive stability, without significant degradation of parameter sensitivity, keeping limited the increment of computational efforts.

On-line stator resistance and permanent magnet flux linkage identification on open-end winding PMSM drives

This paper deals with the development and implementation of an on-line stator resistance and permanent magnet flux linkage identification approach devoted to three-phase and open-end winding permanent magnet synchronous motor drives. In particular, the stator resistance and the permanent magnet flux linkage variations are independently determined by exploiting a current vector control strategy, in which one of the phase currents is continuously maintained to zero while the others are suitably modified in order to establish the same rotating magnetomotive force. Moreover, other motor parameters can be evaluated after re-establishing the normal operation of the drive, under the same operating conditions. As will be demonstrated, neither additional sensors nor special tests are required in the proposed method; Motor electrical parameters can be “on-line” estimated in a wide operating range, avoiding any detrimental impact on the torque capability of the PMSM drive.

Over-voltage mitigation on SiC based motor drives through an open end winding configuration

An original overvoltage mitigation technique is presented in this paper for Sic based motor drives, exploiting an open-end winding motor configuration. According to the proposed solution over-voltage occurring at the terminals of the motor phases is mitigated by driving the two inverters through a modified switching pattern including a suitable dwell time. No passive RLC networks are required, thus avoiding additional costs and extra power losses. A comparison between IGBT and SiC MOSFET equipped motor drives is first provided, highlighting the difference in terms of power cable critical length and other factors affecting peak terminal over-voltages. The proposed approach is then theoretically introduced and its consistency is finally assessed by simulations and experimental results.