Internal Combustion Engines

Real-Time Transient Soot and NOx Virtual Sensors for Diesel Engine Using Neuro-Fuzzy Model Tree and Orthogonal Least Squares

[+] Author and Article Information
Rajit Johri

Mechanical Engineering,  University of Michigan, Ann Arbor, MI 48109rajit@umich.edu

Ashwin Salvi

Mechanical Engineering,  University of Michigan, Ann Arbor, MI 48109asalvi@umich.edu

Zoran Filipi

Mechanical Engineering,  University of Michigan, Ann Arbor, MI 48109filipi@umich.edu

J. Eng. Gas Turbines Power 134(9), 092806 (Jul 23, 2012) (9 pages) doi:10.1115/1.4006942 History: Received December 02, 2011; Revised May 25, 2012; Published July 23, 2012; Online July 23, 2012

Diesel engine combustion and emission formation is highly nonlinear and thus creates a challenge related to engine diagnostics and engine control with emission feedback. This paper presents a novel methodology to address the challenge and develop virtual sensing models for engine exhaust emission. These models are capable of predicting transient emissions accurately and are computationally efficient for control and optimization studies. The emission models developed in this paper belong to the family of hierarchical models, namely the “neuro-fuzzy model tree.” The approach is based on divide-and-conquer strategy, i.e., to divide a complex problem into multiple simpler subproblems, which can then be identified using a simpler class of models. Advanced experimental setup incorporating a medium duty diesel engine is used to generate training data. Fast emission analyzers for soot and NOx provide instantaneous engine-out emissions. Finally, the engine-in-the-loop is used to validate the models for predicting transient particulate mass and NOx .

Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Quasi-steady state model prediction compared with measured soot emissions

Grahic Jump Location
Figure 2

Engine-in-the-loop test cell configuration

Grahic Jump Location
Figure 3

Algorithm for creating virtual emission sensors

Grahic Jump Location
Figure 4

Schematic for generating m-level pseudo random signal

Grahic Jump Location
Figure 5

Staircase test at 2000 rpm

Grahic Jump Location
Figure 6

Throttle signal to engine for SYSID

Grahic Jump Location
Figure 7

Speed signal to engine for SYSID

Grahic Jump Location
Figure 8

Engine visitation points during SYSID test

Grahic Jump Location
Figure 9

Cross correlation of fast NOx and boost

Grahic Jump Location
Figure 10

Network structure of a dynamic neuro-fuzzy model with M local models and n inputs with k tapped-delay output feedback for transient systems

Grahic Jump Location
Figure 11

Construction of a neuro-fuzzy model

Grahic Jump Location
Figure 12

Hierarchical model structure of neuro-fuzzy model tree based emission sensors

Grahic Jump Location
Figure 13

Soft model partition based on engine speed with Gaussian validity function

Grahic Jump Location
Figure 14

Predicted versus measured transient soot emission for a series hydraulic hybrid over FTP75

Grahic Jump Location
Figure 15

Predicted versus measured transient NOx emission for a series hydraulic hybrid over FTP75



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In