	Control Group

University of Cambridge > Department of Engineering > Control Group > Publications > Publication

Robust Automotive Idle Speed Control in a Novel Framework

Ford R.

2000

Abstract

In this thesis the idle speed control problem in gasoline, port fuel injected engines is examined. A new nonlinear engine model is suggested which evolves in the discrete, (combustion) event based, domain and while it is intuitively sensible, it nevertheless accurately captures the behaviour of the engine whilst in the idle mode. As a logical consequence of this model, a new linearising framework is suggested which takes account of the nature of the combustion system and allows a single linear controller to control the idle speed throughout the idle speed/load envelope. Of particular significance and novelty is the second output from the new framework which is the torque reserve of the engine. This not only creates a `square' system but also enables the engine controller to affect on line the trade-off between disturbance rejection effectiveness and engine operating efficiency. The nature of the framework means that the high fidelity representation of the engine system is achieved with significantly less calibration effort than traditional representations. The base controller for the idle control system in the new framework is designed using the H-infinity loop shaping synthesis technique. Effective anti-windup and bumpless transfer functionality is motivated and successfully implemented with demonstrable benefits. Logical enhancements to the system involving feedforward knowledge and anticipation of loads are incorporated in a transparent way. The model, controller and all enhancements are validated in the engine test cell. The final controller shows a drastic improvement in performance and fuel economy over conventional idle speed control techniques.

BibTex Entry

@PhdThesis{,: author = {Ford R.},; title = {Robust Automotive Idle Speed Control in a Novel Framework },; year = {2000}
}



© 2004 Control Group, Department of Engineering, University of Cambridge www-control-admin@eng.cam.ac.uk