This paper describes the use of the extended Park's vector approach (EPVA) for diagnosing the occurrence of stator winding faults in operating three-phase synchronous and asynchronous motors. The major theoretical principles related with the EPVA are presented and it is shown how stator winding faults can be effectively diagnosed by the use of this noninvasive approach. Experimental results, obtained in the laboratory, corroborate that these faults can be detected, in the EPVA signature, by the identification of a spectral component at twice the fundamental supply frequency. Onsite tests, conducted in a power generation plant and in a cement mill, demonstrate the effectiveness of the EPVA in the detection of these faults in large industrial motors, rated up to 5 MW.

This paper addresses the subject of stator fault diagnosis in direct torque control (DTC) induction motor drives. The particularities of the behavior of a DTC drive, when in the presence of inter-turn short circuits, are exploited, thus allowing the proposal of two different approaches for the diagnosis of this type of fault. One of the proposed diagnostic techniques is the well known motor current spectrum analysis. It is demonstrated that in a DTC induction motor drive, with stator inter-turn short circuits, the action of the torque and flux controllers introduce a strong third harmonic in the motor supply currents, which can be used to detect this type of fault. The other diagnostic technique is based on the multiple reference frames theory, recently proposed by the authors for the diagnosis of stator faults in line-connected motors. Here, its use is exploited in the diagnosis of stator faults in DTC induction motor drives. Simulation and experimental results demonstrate the effectiveness of the two proposed techniques for the diagnosis of stator faults but emphasize the fact that the multiple reference frames technique is the preferred one when the residual asymmetries of the motor are not negligible.

This paper introduces a new approach, based on the spectral analysis of the motor current Park's Vector modulus, for detecting the occurrence of rotor cage faults in operating three-phase induction machines. The major theoretical principles of the Extended Park's Vector Approach are described. Both simulation and laboratory test results demonstrate how the presence of rotor cage faults can be effectively detected by the application of this new approach. These results are also correlated with the ones obtained by the conventional motor current spectral analysis and by the classic Park's Vector Approach.

This paper proposes the use of the multiple reference frames theory for the diagnosis of stator faults in three-phase induction motors. The development of a simplified mathematical motor model allowed the establishment of the equivalent circuits of the motor, in d-q-0 axes, in the presence of stator interturn short circuits. The use of the stationary reference frame, clockwise and counterclockwise synchronous reference frames, allows the extraction and manipulation of the information contained in the motor supply currents in a way that the effects introduced by the fault are easily isolated and measured. A severity factor is defined and the simulation and experimental results presented demonstrate its independence in relation to the working conditions of the motor, such as the load level and unbalances in the voltage supply system. Although the technique is here introduced for the diagnosis of stator faults, it is possible to extend its use for the diagnosis of other asymmetries such as broken rotor bars and air-gap eccentricity.

In this paper, new techniques are presented to improve the performance of Optimal Switching Vector Model Predictive Control. A well-defined analytic technique is proposed for the real-time adjustment of dynamic weighting factors, as well as a new objective function formulation. These techniques provide a dynamic controller behavior, automatically adapted to distinct operating points. This ensures better steady-state performance in different conditions while improving simultaneously the dynamic response. The proposed controller presents a lower computational load than the conventional formulation and its design requires less extensive testing. General guidelines are presented to guide the user in the controller design. Simulation and experimental results are presented, demonstrating the advantages and superior performance of the proposed controller.