Unbalance and misalignment detection in induction motors under oscillating loads using current and Fast Fourier Transform (FFT)

This paper presents a method for detecting unbalance and misalignment faults in induction motors under oscillating loads. The main objective is to acquire the electrical current signal from the stator and process it to discriminate between three motor conditions: healthy, unbalance, and misalignment. To conduct this study, an experimental matrix comprising 15 tests was designed, covering four different frequencies (15, 30, 50, and 60 Hz) and three motor conditions mentioned earlier. The current signal was accurately acquired and processed using GNU Octave software. By analyzing the Fast Fourier Transform of the current signal, it was possible to identify and characterize the sidebands associated with unbalance and misalignment faults. These sidebands provided valuable information for the detection and diagnosis of abnormal motor conditions. Finally, an intuitive user interface was developed, providing two reports. The first report, of a technical nature, allows the operator to visually and easily detect the analyzed faults in the motors. A traffic light indicator is used to indicate the motor's status (healthy or presenting a fault). The second report provides the amplitude values of the frequency sidebands in the Fourier spectrum associated with faults.