Multiple Discriminant Analysis and Neural-Network-Based Monolith and Partition Fault-Detection Schemes for Broken Rotor Bar in Induction Motors

Broken rotor bars in induction motors can be detected by monitoring any abnormality of the spectrum amplitudes at certain frequencies in the motor-current spectrum. It has been shown that these broken-rotor-bar specific frequencies are located around the fundamental stator current frequency and are termed lower and upper sideband components. Broken-rotor-bar fault-detection schemes should rely on multiple signatures in order to overcome or reduce the effect of any misinterpretation of the signatures that are obscured by factors such as measurement noises and different load conditions. Multiple discriminant analysis (MDA) and artificial neural networks (ANNs) provide appropriate environments to develop such fault-detection schemes because of their multiinput-processing capabilities. This paper describes two fault-detection schemes for a broken-rotor-bar fault detection with a multiple signature processing and demonstrates that the multiple signature processing is more efficient than a single signature processing. The first scheme, which will be named the "monolith scheme," is based on a single large-scale MDA or ANN unit representing the complete operating load-torque region of the motor, while the second scheme, which will be named the "partition scheme," consists of many small-scale MDA or ANN units, each unit representing a particular load-torque operating region. Fault-detection performance comparison between the MDA and the ANN with respect to the two schemes is investigated using the experimental data collected for a healthy and a broken-rotor-bar case. Partition scheme distributes the computational load and complexity of the large-scale single units in a monolith scheme to many smaller units, which results in the increase of the broken-rotor-bar fault-detection performance, as is confirmed with the experimental results