By Topic

Engine emissions modeling for a hybrid electric vehicle

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Gray, D.L. ; Purdue Univ.Calumet, Hammond, IN, USA ; Hentea, T.I.

The objective of this work is the development of engine emissions models for an internal combustion engine of a hybrid electric vehicle. From the vast variety of models initially considered, the following three classes were extensively investigated: static engine maps, dynamic multiple regression linear models, and time delay dynamic neural network models. All the models were built and validated with experimental data from a fuel injected SI engine burning conventional 87 octane unleaded fuel. The total time duration of the experimental data was about 10,000 seconds. The data was divided into two groups: 6700 seconds used for model development and a 3300 second validation set were used to verify the performance of the developed models. The two groups of data contain sections of highway and urban driving, as well as limited cold starting. The models use input variables of engine speed, engine torque, output power, throttle angle, coolant, exhaust, and engine block temperatures. The model outputs are the engine output NOx, CO2, CO, and hydrocarbon (HC) emissions. The models are intended to operate on a per second rate. All the dynamic models predict satisfactorily the grams per second engine emissions of NOx, CO2, and CO for hot engine operation. HC emissions for a hot engine functioning at positive torque output are well predicted only by the neural network model. In the cases of engine operation at negative torque outputs, as well as for cold engine starts, all of the models have large prediction errors mainly due to the small data set utilized. In these conditions the neural network models outperformed both the linear dynamical models and the static map.

Published in:

Energy Conversion Engineering Conference, 2002. IECEC '02. 2002 37th Intersociety

Date of Conference:

29-31 July 2004