Digital Twin of an Induction Motor: Fault Analysis and Predictive Maintenance
Details:
In this model, we basically present an efficient
method to predict the fault in motors using digital twin and
predictive machine learning models to estimate the fault in
induction motor. The faults in an induction motor can be broken
down into three different categories which are bearing faults,
stator faults and unbalanced voltages and centricity. These faults
have direct impact on two important parameters which are
vibrational signals and stator currents, and these two parameters
will be used in our predictive machine learning model for further
analysis. All the other parameters are physical in nature and
cannot be modelled using simulated motors. This includes using
detection models on sensory thermography, oil analysis, and
ultrasound data.
The block Diagram is:
