The Problem:
Traditionally, the IPM motor drive is controlled by combining a flux weakening plus maximum torque per ampere (MTPA) block, a lookup table block, and a current controller block. However, the reference commands generated from currently available methods are inaccurate, which causes reduced motor efficiency. Thus, the neural networks (NN) allow for the optimal performance of internal permanent magnetic (IPM) motors.
The Solution:
Researchers at The University of Alabama have developed neural networks and adaptive dynamic programming as a means of improving the efficiency of electric vehicles (EV) and hybrid (HEV) vehicles. Evs and HEVs use IPM motors which are controlled by reference commands. The NNs in this technology allow for the most efficient IPM motor control, the identification of motor parameters in real-time operating conditions, and the replacement of the traditional lookup methods which allow for increased accuracy in mapping currents to motor parameters.
Neural network controller trained with adaptive dynamic programming
Benefits:
• Improves efficiency.
• Improves motor size and reliability.
• Reduces harmonics and torque oscillation.
• Maximizes motor output torque is enhanced.
The University of Alabama Research Office of Innovation and Commercialization (OIC) is a non-profit corporation that is responsible for commercializing University of Alabama technologies and for supporting University research. At OIC, we seek parties that are interested in learning more about our technologies and commercialization opportunities, and we welcome any inquiries you may have.
