Inverter statistics for online detection of stator asymmetries in inverter-fed induction motors

A simulation model of an induction machine that enables the analysis of the influence of stator fault conditions on the machine control of inverter-fed drives is presented. Based on this model, the influence of the breakdown of the stator windings insulation on the behavior of the machine and especially on the current control scheme is shown. A new online method to detect such asymmetries caused, for example, by an interturn insulation failure in the stator windings is proposed and investigated. This new method utilizes the influence of stator asymmetries on the inverter current control scheme. By evaluating the statistical distribution of the different inverter switching states and switching times, asymmetries in the stator can be detected and isolated. The measurements required to implement this method are already available in modern inverter fed drives as they are used to realize the current control loop. Thus, no additional sensors are necessary. The practical realization of the fault detection algorithm is demonstrated in combination with a predictive single step current controller. Measurements performed on a drive test stand verify the applicability of the proposed online method to detect and isolate stator faults.     

Scheme to reconstruct phase current information of inverter fed ACdrives

In modem inverter fed AC drives the current control loop is usually realised by applying two current sensors in the phases of the machine. Different schemes have been presented in the past which enable a current control loop with a single current sensor in the DC link. At low fundamental frequencies they all need to modify the standard pulse-width modulation (PWM) scheme periodically in order to update the phase current values. Here, an advanced method is presented which improves the accuracy of the phase current reconstruction, thus reducing the necessary modification of the PWM     

Magnetising method for speed sensorless controlled induction motordrives

In speed-sensorless controlled standard induction machine drives a start without torque is only possible if the rotor speed of the demagnetised machines is known. A new method is proposed, which allows a fast and accurate identification of the rotor speed of demagnetised machines without a speed or position sensor     

Online-Erkennung von Wicklungsfehlern in Asynchronmaschinen

The operation of induction machines with fast switching power electronic devices puts additional stress on the stator windings which leads to an increased probability of machine faults. These faults can cause considerable damage and repair costs and — if not detected in an early stage — may end up in a total destruction of the machine. To reduce maintenance and repair costs many methods have been developed and presented in literature for an early detection of machine faults. This paper gives an overview of today’s detection techniques and describes algorithms and results of some methods with special emphasis to inverter fed operation.     

Current-controller with single DC link current measurement for inverter-fed AC machines based on an improved observer-structure

In modern industrial applications even low-cost drives have excellent dynamic behavior, which is achieved by field-oriented control combined with high-dynamic current regulation. Usually at least two phase-current sensors are necessary to realize such an operation. In this paper, a new current-control scheme is presented, which enables high-dynamic control of ac machine-line currents with improved accuracy by measuring only the current of the dc-link. The controller is based on three individual and adaptive phase-current observers, which guarantee very accurate phase-current estimates even at low-modulation indices. Thus, operation is possible even at low-speed without modifying the switching pattern of the controller and the necessary modification at zero-speed can be clearly reduced. Different realizations of the proposed scheme are given and a comparison is made to previously published schemes which are also based on a single dc-link current sensor. Measurements performed on an induction-machine drive at different points of operation show the applicability of the proposed structure. It is shown that the long-term accuracy is clearly increased and the maximum deviation of the observer estimate from the actual phase-current is reduced when using the proposed scheme.     

Transient simulation of lamination material properties inelectrical machines

All zero-speed-sensorless control schemes of electrical drives use a high-frequency excitation of the machine to determine local differences of the leakage inductances and to estimate the flux or the rotor position. One of these methods detects the local differences by evaluating the change of the stator current space phasor due to stator voltage test phasors. To get a deeper insight into the transient electric behavior of the machine, it is necessary to take into account also dynamic processes in the lamination material. To achieve this, a model has been developed to simulate the magnetization process in an induction motor fed by a voltage-source inverter. The proposed model calculates the current change due to voltage pulses taking into account the hysteresis as well as eddy currents. In order to limit the computation time, the model has to be kept simple. Measurements on an inverter-fed induction machine as well as on a stator lamination are presented. to verify the applicability of the model     

Möglichkeiten zum Einsatz von neuronalen Netzen zur mechanisch sensorlosen Regelung von umrichtergespeisten Asynchronmotoren

Problems in the realization of drives without mechanical sensor are discussed with special emphasis on the operation at low fundamental electrical frequency including zero frequency. One method that is based on spatial differences in the saturation of the induction machine is shown and a possible combination with a neural network is presented. The main task of the neural network is the identification of the magnetic fingerprint of the induction machine.