| |
Design and DSP Implementation of AI-Based Medium Performance Sensorless Induction Motor Drives
| [View]
[Download]
|
| Author(s) |
P. Vas; M. Rashed; A. F. Stronach; M. Neuroth |
| Abstract |
The DSP implementation of three speed-sensorless medium-performance induction motor drives incorporating Artificial Intelligence (AI) is discussed. The first drive is a scalar induction motor drive and contains a minimal configuration multi-layer feed-forward neural-network-based speed estimator. Although the neural network was trained by using simulation results onle, it was successfully implemented in an induction motor drive employing a 3 kW cage induction motor. However, the same speed estimator neural network (ANN) was also successfully used in the second induction motor drive with a 2.2 kW motor. The performance of speed estimators using feed-forward multi-layer and recursive artificial neural networks are also compared for this drive. In addition to an ANN-based speed estimator, the third induction motor drive contains a simple fuzzy-logic-based system with a minimum rule-base, which improves the low-speed performance. The experimental results show that the implemented artificial-intelligence-based-based drives give satisfactory performance in a wide speed range. All the drive schemes are simple to implement and the memory requirements are modest. The DSP used is the Texas Instruments TMS320C30. |
| Download |
| Filename: | EPE1999 - PP00032 - Vas.pdf |
| Filesize: | 130.6 KB |
|
| Type |
Members Only |
| Date |
Last modified 2004-04-07 by System |
| |
 |