Broken Rotor Bar Detection in Line-Fed Induction Machines Using Complex Wavelet Analysis of Startup Transients

Fault detection of line-connected induction machines using complex vector wavelets to analyze the transient stator currents during startup is proposed in this paper. When a machine is connected to the line, the startup transient is characterized by large stator (and rotor) currents as well as by large slips (i.e., the rotor speed is significantly smaller than the excitation frequency). The stator current of machines with damaged rotors includes large rotor speed dependent components during the startup transient. Such components, however, fade away or coincide with components not containing fault-related information (e.g., saturation-induced components) once the machine reaches steady state. Because of this, the startup transient provides an opportunity for performing diagnostics on the machine. This paper shows that the information contained in the startup transient signal can be effectively separated and detected using a complex vector wavelet transform.