Abstract: Boost-type switch-mode rectifiers (SMR) commonly use a resistor or a magnetic current sensor to measure the instantaneous input or inductor current that is used as the feedback to the current controller. A novel current sensorless scheme is described that computes the inductor current from the measured inductor voltage in a single-phase boost-type SMR using an adaptive low pass filter. This calculation requires an estimate of the inductance and the equivalent series resistance of the inductor coil. Both these parameters are dependent on operating conditions and are updated continuously. This is done using an adaptive model of the inductor that computes these parameters of the inductor once in every half cycle of the input current. The adaptation scheme is robust against parameter variations. Simulation and experimental results confirm the effectiveness of the proposed technique which provides comparable performance to standard measured feedback current scheme both under steady-state and transient conditions.

Abstract: Faults and fault types in electric machines can be identified based on spectra associated with stator and field currents. Signals associated with such currents can be compensated for line variations, and spectral contributions for non-eccentric machines can be reduced or eliminated to permit distinguish static eccentricities from non-eccentric machine operation. One or more of a static eccentricity, a dynamic eccentricity, and a mixed eccentricity can be identified based on spectral component magnitudes at selected frequencies.

Abstract: Boost-type switch-mode rectifiers (SMR) commonly use a resistor or a magnetic current sensor to measure the instantaneous input or inductor current that is used as the feedback to the current controller. A novel current sensorless scheme is described that computes the inductor current from the measured inductor voltage in a single-phase boost-type SMR using an adaptive low pass filter. This calculation requires an estimate of the inductance and the equivalent series resistance of the inductor coil. Both these parameters are dependent on operating conditions and are updated continuously. This is done using an adaptive model of the inductor that computes these parameters of the inductor once in every half cycle of the input current. The adaptation scheme is robust against parameter variations. Simulation and experimental results confirm the effectiveness of the proposed technique which provides comparable performance to standard measured feedback current scheme both under steady-state and transient conditions.

Abstract: Faults and fault types in electric machines can be identified based on spectra associated with stator and field currents. Signals associated with such currents can be compensated for line variations, and spectral contributions for non-eccentric machines can be reduced or eliminated to permit distinguish static eccentricities from non-eccentric machine operation. One or more of a static eccentricity, a dynamic eccentricity, and a mixed eccentricity can be identified based on spectral component magnitudes at selected frequencies.

Abstract: A fault detection system using field current and rotor search coil voltage harmonics is provided. The technology is capable of detecting an inter-turn fault unambiguously and rapidly in a synchronous machine. The internal asymmetry of the machine that has adversely affected the earlier fault diagnosis results has, for the first time, been shown as an advantage.

Abstract: A fault detection system using field current and rotor search coil voltage harmonics is provided. The technology is capable of detecting an inter-turn fault unambiguously and rapidly in a synchronous machine. The internal asymmetry of the machine that has adversely affected the earlier fault diagnosis results has, for the first time, been shown as an advantage.