Support Vector Machine-Based Determination of Gasoline Direct Injected Engine Admissible Operating Envelope Ford Motor Company, Ford Research Laboratory

Format:

Book

Conference/Event

Author/Creator:

Kolmanovsky, Ilya V., 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:

Support Vector Machines (SVMs) have been gaining popularity as classifiers with good generalization ability. In an attempt to study their applicability to typical automotive problems, this paper investigates the modeling of the operating envelope for a direct injection gasoline (GDI) engine. This envelope defines the admissible ranges for key engine operating variables so that specified conditions on engine roughness and misfire are satisfied. The SVM model of the operating envelope is subsequently used by the engine control strategy to set engine operating variables such as spark and injection timing to avoid excessive engine roughness and misfire. Findings and conclusions from this study related to generalization ability and complexity of the SVM classifier models are summarized

Notes:

Vendor supplied data

Publisher Number:

2002-01-1301

Access Restriction:

Restricted for use by site license