Neural Network Based Feedforward Control for Electronic Throttles Motorola, Incorporated

Format:

Book

Conference/Event

Author/Creator:

Yang, Hanlong, author.

Conference Name:

SAE 2002 World Congress & Exhibition (2002-03-04 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2002

Summary:

This paper addresses feedforward tracking control for electronic throttles. A robust and accurate tracking control scheme based on the training of a Neural Network model and feedback term (PID) is developed. The Neural Network based term can be trained off-line. This feedfoward term serves as a mathematical model capable of describing Electronic Throttle dynamics over a wide range. We have shown that by adding the Neural Network based feedforward control to a common feedback control method, such as the gain-scheduled PID used in many ETC production controllers, that the tracking control performance criteria such as transient errors, steady state errors, response time and overshoot, are greatly improved. Experiments conducted on a production Electronic Throttle Body with a Motorola H-brigde driver IC have shown good results utilizing this approach

Notes:

Vendor supplied data

Publisher Number:

2002-01-1149

Access Restriction:

Restricted for use by site license