In this paper a new method based on pattern recognition by using artificial neural networks is presented for detection and classification of induction motor ball bearing faults. And also one set of data is used for testing network. In this method, time and frequency of domain characteristics of the vibration signal of the engine body are extracted and used as the neural network input. These characteristics are obtained through direct processing of the different parts of measured signal. Three modes of safe, the outer ring fault, and inner ring fault are classified using the proposed algorithm and type of fault is determined. The results show that pattern recognition algorithms can determine fault location using time domain features with high accuracy and without need to complex calculations.
کلید واژگان :Fault detection, Ball Bearing, Induction Motor, Pattern Recognition.
ارزش ریالی : 300000 ریال
با پرداخت الکترونیک