Comparison between adaptive neuro-fuzzy inference system and general regression neural networks for gearbox fault detection using motor operating parameters

Condition monitoring of a gearbox is a crucial activity due to its importance in power transmission for many industrial applications. Thus, there has always been a constant pressure to improve measuring techniques and analytical tools for early detection of faults in gearboxes. This study focuses on developing gearbox monitoring methods using the operating parameters obtained from machine control processes rather than the traditional measures such as vibration and acoustics. To monitor the gearbox conditions, an adaptive neuro-fuzzy inference system (ANFIS) is used to capture the nonlinear connections between the electrical motor current and control parameters such as load settings and temperatures. The predicted values generated by the ANFIS model are then compared with the measured values to indicate the abnormal condition in gearbox. Furthermore, a comparative
study of the results this technique and the general regression neural networks (GRNN) is also carried out. The comparison results show that the ANFIS model performs more accurately than the other model in gearbox condition monitoring and fault detection.