: The main objective of this dissertation is to develop a new technique for detecting stator winding short circuit faults in Induction Motor (IM). So modeling turn-to-turn short circuit fault in more than one phase and also phase-to-ground fault is the first novelty of this research. Then a novel online and nondestructive algorithm is proposed using motor current signature analysis besides infrared imaging. This technique is able to accurately detect all kinds of stator winding short circuit in three-phase IM in early stage. In addition, locating and defining severity of the fault are the important outcomes of this technique. Experimental results are included to validate the proposed models and demonstrate the effectiveness of the proposed method using a 2-hp IM. Key words Condition monitoring, Induction motor, Motor current signature analysis, Thermography, Feature extraction
