MODIFIED LMS ADAPTIVE ALGORITHM FOR CMAC NEURAL-NETWORK-BASED CONTROLOF SWITCHED RELUCTANCE MOTORS

Authors
SHANG C REAY DS WILLIAMS BW
Citation
C. Shang et al., MODIFIED LMS ADAPTIVE ALGORITHM FOR CMAC NEURAL-NETWORK-BASED CONTROLOF SWITCHED RELUCTANCE MOTORS, Electronics Letters, 32(12), 1996, pp. 1113-1115
Citations number
4
Categorie Soggetti
Engineering, Eletrical & Electronic
Journal title
Electronics Letters → ACNP
ISSN journal
00135194
Volume
32
Issue
12
Year of publication
1996
Pages
1113 - 1115
Database
ISI
SICI code
0013-5194(1996)32:12<1113:MLAAFC>2.0.ZU;2-6
Abstract
A novel approach to adapting the weights of a CMAC neural network for torque ripple reduction in switched reluctance motors is proposed, usi ng a variable learning rate function within the standard LMS algorithm . Simulation results demonstrate that training CMAC networks following this approach affords low torque ripple with high power efficiency.