Raed Almozel

Artificial Intelligence Specialization · Cohort 2024

Real-Time Intelligent Control System to Address Faults for DC Motors

This project focuses on enhancing the modeling of Permanent Magnet DC (PMDC) motors through the application of advanced machine learning techniques. The primary objective is to develop accurate predictive models for key motor parameters, which are essential for efficient motor control and performance optimization. Initially, we developed models to predict motor velocity based on voltage inputs. This involved training machine learning algorithms on a dataset comprising various voltage inputs and their corresponding motor velocities. Additionally, we created models to estimate rotor angular velocity from pulse width modulation (PWM) signals. These models were trained using data that mapped PWM signals to the resulting angular velocities of the rotor. The predictive models for motor velocity and rotor angular velocity were then integrated to estimate rotor temperature. Rotor temperature is a critical parameter that significantly affectsmotor efficiency and longevity, yet it is challenging to measure directly due to the nonlinear behavior of PMDC motors.

Read the final report (PDF)

See it in the program report (PDF)