Vol. 158, Issue 11, November 2013, pp. 302-311

Abstract: In the speed sensorless induction motor drives system, the Rotor Flux based Model Reference Adaptive System (RF-MRAS) is the most common strategy. It suffers from parameter sensitivity and flux pure integration problems. As a result, it leads to the deterioration of speed estimation. Simultaneously, the traditional PI parameters design may cause speed estimation instability or have gross errors in the regenerative mode. To overcome above-mentioned problems, a suitable Artificial Neural Networks (ANN) based on Ant Colony Optimization (ACO) is presented to replace the reference model of the RF-MRAS. Furthermore, the ANN learning by the modified ACO can enhance the ANN convergence speed and avoids the trap of local minimum value of algorithm. In the meantime, a fuzzy PI adaptation mechanism is also put forward, so the proportional coefficient kp and the integral coefficient ki can be adjusted dynamically to adapt the speed variations. Finally, the simulation results suggest that the speed estimation is more accurate in both the dynamic and static process, and the stability of speed estimation in regenerative mode was improved.

Keywords: Vector control, Speed sensorless, ANN, Modified ACO, Fuzzy PI adaptation mechanism.

Click <here> or title of paper to download the full pages article in pdf format

Download <here> the Library Journal Recommendation Form

1999 - 2018 Copyright ©, International Frequency Sensor Association (IFSA). All Rights Reserved.

Home - News - Links - Archives - Tools - Voltage-to-Frequency Converters - Standardization - Patents - Marketplace - Projects - Wish List - e-Shop - Sensor Jobs - Membership - Videos - Publishing - Site Map - Subscribe - Search

 Members Area -Sensors Portal -Training Courses - S&T Digest - For advertisers - Bookstore - Forums - Polls - Submit Press Release - Submit White Paper - Testimonies - Twitter - Facebook - LinkedIn