Fault Detection of the Electrical Motors Based 3D Accelerometer Sensor and 1D CNN

In the modern industrial production environment, maintaining the stable operation of machines is extremely important to ensure performance and safety. Therefore, early detection and diagnosis of technical problems has become necessary. In this context, this paper presents a fault detection method to diagnose faults in electrical motors using 3D an accelerometer sensor. Vibration data of the motor measured from the sensor characterizes its state, and thus it can indicate different faults. A ID convolution neural network (CNN) model is proposed classify four states of the motors consisting normal, bearing fault, misalignment, and unbalance. The experimental result showed that the accuracy of the proposed algorithm is up to 99.8%.