Artificial neural network based controller for permanent magnet DC motor drives
- Memorial Univ. of Newfoundland, St. John`s, Newfoundland (Canada). Faculty of Engineering and Applied Science
This paper introduces a novel approach of designing a controller using multi-layer feed-forward neural network (FFNN) for the speed control of a permanent magnet (PM) dc motor. Artificial neural network (ANN) controller with its massive parallel properties and learning capabilities offers a promising way to solving the problem of system non-linearity, parameter variations and unexpected load excursions associated with a PM dc motor drive system. Self-tuning technique of the controller in real time is achieved through an improved on-line back-propagation training algorithm based on an output error propagation. The proposed ANN controller is implemented with a PM dc motor drive system in the laboratory. The laboratory test results validate the efficacy of the based controller for a high performance PM dc motor drive.
- OSTI ID:
- 415565
- Report Number(s):
- CONF-9510203-; TRN: IM9704%%196
- Resource Relation:
- Conference: IEEE/Industrial Application Society conference, Orlando, FL (United States), 8-12 Oct 1995; Other Information: PBD: 1995; Related Information: Is Part Of Conference record of the 1995 IEEE Industry Applications Society thirtieth IAS annual meeting. Volume 2; PB: 954 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Laboratory implementation of a neural network trajectory controller for a DC motor
A continually online-trained neural network controller for brushless DC motor drives