Research On Rotor Fault Diagnosis Of AC Motors

Electric motors convert electrical energy into mechanical energy.They have been widely used in industrial and agricultural production as well as daily life.As the key equipment of the system,motors have a vital role in the operation of the entire system safely and reliably.Unscheduled shutdown due to motor faults may cause significant economic losses,even human casualties.Therefore,it is of great theoretical significance and economic benefits to detect the rotor faults for reasonable maintenance at an early stage.Rotor fault is one type of the most commonly faults in electrical motors.The rotor fault of induction motors has been widely studied in the past few decades and many methods have been proposed.However,most of the methods are proposed for the mains-fed induction motors.With the development of power electronics technology,the inverter and closed-loop control have been widely used.Due to the compensating effect of the control system,the rotor fault information propagates to the motor voltage and affects the signature in the motor.Therefore,the performances of the diagnosis methods developed for line-fed induction motors are doubtful for closed-loop controlled motors.However,scarce methods have been made to solve the problem.Furthermore,permanent magnet synchronous motors(PMSMs)are gradually applied in many applications due to the increased requirements of energy-saving.Magnet faults may occur because of electrical,mechanical,thermal stress in the long operating.At present,the diagnosis of irreversible demagnetization fault is still in its infancy.Against this background,some work has been carried out on the broken bar fault for closed-loop controlled induction motors and demagnetization fault for PMSMs.The main research of the thesis can be outlined as follows:Firstly,the phase model and sequence model of the induction motors have been derived.The fault characteristic components due to broken bar fault and demagnetization fault are elaborated by analyzing the magnetomotive force.This part provides theory support for the diagnosis methods in the following chapters.Secondly,the fault information of broken rotor bars propagates to the voltage when the induction motor is in a closed-loop control.A left sideband component emerges in the voltage spectrum,which represented the negative component in the rotor reference.The conventional sequence model has been improved with considerating the negative voltage component.A fault severity factor has been proposed from the improved model.The amplitudes and phases for calculating the factor are obtained by the reference frame theory.The diagnosis result is not sensitive to the control strategy and its parameters.Therefore,the method can be used in different control strategies.Thirdly,from another point of view,the zero-sequence components are proved independent of the control strategy.The relation between the zero sequence compenent and the broken rotor bar fault of the induction motor is analysed.With the help of the superposition model of the motor,two fault severity factors have been proposed.It has been proved theoretically that the factors are almost independent of the control strategy.The method is able to discriminate broken bar fault and load oscillations.The influence of unbalanced voltage supply on the diagnosis result is negligible.Finally,the effects of partial demagnetization fault on the PMSM are analysed in detail.Partial demagnetization fault induces circulating currents between the branches,which increase the losses of the motor.When the motor is at a low speed and no-load operation,the increased losses will be highlighted.Therefore,the increase losses can be used to detect partial demagnetization fault.The fault decreases the flux in d-axis and finally decreases the magnetic saturation of the core.The paper extracts the d-axis saturation information of the motor by applying a "DC + square wave" voltage and a "DC + sine wave" voltage in the d-axis.The d-axis saturation information can be used to detect demagnetization fault.The experimental results have proved that the d-axis saturation based methods can detect the demagnetization fault effectiveness and the results are sensitive to the fault.