LEARNING TECHNIQUES TO TRAIN NEURAL NETWORKS AS A STATE SELECTOR FOR INVERTER-FED INDUCTION MACHINES USING DIRECT TORQUE CONTROL

Citation
La. Cabrera et al., LEARNING TECHNIQUES TO TRAIN NEURAL NETWORKS AS A STATE SELECTOR FOR INVERTER-FED INDUCTION MACHINES USING DIRECT TORQUE CONTROL, IEEE transactions on power electronics, 12(5), 1997, pp. 788-799
Citations number
13
Categorie Soggetti
Engineering, Eletrical & Electronic
ISSN journal
08858993
Volume
12
Issue
5
Year of publication
1997
Pages
788 - 799
Database
ISI
SICI code
0885-8993(1997)12:5<788:LTTTNN>2.0.ZU;2-D
Abstract
Neural networks are receiving attention as controllers for many indust rial applications. Although these networks eliminate the need for math ematical models, they require a lot of training to understand the mode l of a plant or a process. Issues such as learning speed, stability, a nd weight convergence remain as areas of research and comparison of ma ny training algorithms. This paper discusses the application of neural networks to control induction machines using direct torque control (D TC), A neural network is used to emulate the state selector of the DTC , The training algorithms used in this paper are the backpropagation, adaptive neuron model, extended Kalman filter, and the parallel recurs ive prediction error, Computer simulations of the motor and neural-net work system using the four approaches are presented and compared, Disc ussions about the parallel recursive prediction error and the extended Kalman filter algorithms as the most promising training techniques is presented, giving their advantages and disadvantages.