Method of condition monitoring of rotor windings of the medium voltage induction motors of power plant auxiliaries in start-up modes

Abstract

Background. A promising method of monitoring rotor winding condition of induction electric motors ensuring higher reliability of their operation is spectral analysis of the external magnetic field signal. Modern methods of external magnetic field analysis use only the spectrum of this signal in the steady-state operation mode, which does not allow effective monitoring on a number of medium voltage electric motors of power plant auxiliaries operating with long start-ups (for example, machines with variable load, with a double squirrel cage, etc.). It is possible to evaluate the rotor winding condition of such machines in start-up modes, which means that an important task is to study the frequency-time spectra of the external magnetic field of the above-mentioned electric motors and to develop a method for monitoring the state of their rotor windings in the startup mode. Materials and methods. In the study, the authors used two-dimensional models of two different types of motors representing drives of the mechanisms of draft systems at power stations. Both models are developed in the ANSYS software package. The calculation of the external magnetic field was carried out using the finite element method. The modelling was done for cases with or without defects. Signal processing was performed using the Fourier window transform method in the MatLab software package. Results. Frequency-time spectra of the external magnetic field were built in the start-up mode on models of the investigated induction motors with the following defects: with dynamic eccentricity, with one broken bar of the rotor winding and without the indicated defects. The research results allowed us to develop a method for monitoring the rotor winding state of medium voltage electric motors of power plant auxiliaries operating with long start-ups. The method is based on detecting lower side frequencies of the fictitious rotor winding harmonics in the frequency-time spectrum during the startup. Their specific feature is «reflection» in the spectrum from the time axis. Conclusions. The developed method of monitoring rotor winding condition of the above-mentioned induction electric motors allows increasing the reliability of their operation by timely detection of faults in the rotor windings.