0
Research Papers

# NO and $NO2$ Concentration Modeling and Observer-Based Estimation Across a Diesel Engine Aftertreatment System

[+] Author and Article Information
Ming-Feng Hsieh

Department of Mechanical and Aerospace Engineering, Ohio State University, Columbus, OH 43210hsieh.122@osu.edu

Junmin Wang1

Department of Mechanical and Aerospace Engineering, Ohio State University, Columbus, OH 43210wang.1381@osu.edu

1

Corresponding author.

J. Dyn. Sys., Meas., Control 133(4), 041005 (Apr 07, 2011) (13 pages) doi:10.1115/1.4003380 History: Received February 14, 2010; Revised November 24, 2010; Published April 07, 2011; Online April 07, 2011

## Abstract

This paper presents an experimentally validated control-oriented model and an observer for diesel oxidation catalyst (DOC)-diesel particulate filter (DPF) system in the context of exhaust gas NO and $NO2$ concentration estimations. NO and $NO2$ have different reaction characteristics within DPF and selective catalytic reduction (SCR) systems, two most promising diesel engine aftertreatment systems. Although the majority of diesel engine-out $NOx$ emissions is NO, the commonly used DOC located upstream of a DPF and a SCR can convert a considerable amount of NO to $NO2$. Knowledge of the $NO/NO2$ ratio in exhaust gas is thus meaningful for the control and diagnosis of DPF and SCR systems. Existing onboard $NOx$ sensors cannot differentiate NO and $NO2$, and such a sensory deficiency makes separate considerations of NO and $NO2$ in SCR control design challenging. To tackle this problem, a control-oriented dynamic model, which can capture the main NO and $NO2$ dynamics from engine-out, through DOC, and to DPF, was developed. Due to the computational limitation concerns, DOC and DPF are assumed to be standard continuously stirred tank reactors in order to obtain a 0D ordinary differential equation model. Based on the model, an observer, with the measurement from a commercially available $NOx$ sensor, was designed to estimate the NO and $NO2$ concentrations in the exhaust gas along the aftertreatment systems. The stability of the observer was shown through a Lyapunov analysis assisted by insight into the system characteristics. The control-oriented model and the observer were validated with engine experimental data and the measured $NO/NO2$ concentrations by a Horiba gas analyzer. Experimental results show that the model can accurately predict the main engine-out/DOC/DPF $NO/NO2$ dynamics very well in semisteady-state tests. For the proposed observer, the predictions converge to the model values and estimate the NO and $NO2$ concentrations in the aftertreatment system well.

<>

## Figures

Figure 1

A commonly used diesel engine aftertreatment setup

Figure 2

Diesel engine and aftertreatment system test bench setup

Figure 3

Schematic presentation of the setup

Figure 4

Experiment results of NO2 and NOx relation

Figure 5

CSTR model of DOC

Figure 6

Simplified model of PM contact area inside DPF

Figure 7

CSTR model of DPF

Figure 8

Measurements of Horiba gas analyzer and NOx sensors from test 1

Figure 9

Engine speed and torque profile of Fig. 8

Figure 10

Comparison of tests 3 and 4 engine-out NOx emissions

Figure 11

Comparison of exhaust variables of tests 3 and 4 at high speed

Figure 12

NO and NOx measurements of test 2 and comparisons to engine model predicted NO

Figure 13

Comparisons of engine exhaust NO concentrations measured by Horiba gas analyzer and estimated by engine model

Figure 14

Comparison of engine-out exhaust NO and NO2 measurements and model predicted values during transient operations at a higher engine speed

Figure 15

Steady-state NO and NOx measurements from test 4 and model predicted engine exhaust NO

Figure 16

Comparisons of NO and NO2 measurements after DOC and model simulation

Figure 17

Validation of DOC model by experimental data

Figure 18

Comparisons of engine exhaust NOx, downstream DOC NO, and downstream DPF NO

Figure 19

CO and CO2 concentrations after DOC (before DPF)

Figure 20

Comparisons of experimental data from Test 3 and DPF model simulation

Figure 21

Validation of DPF model estimation with experimental data from test 3

Figure 22

Comparisons of model values observer estimations

Figure 23

Zoom-in of the initial part of Fig. 2

Figure 24

Comparisons of measurements and observer estimations (after DOC)

Figure 25

Comparisons of measurements and observer estimations at higher speeds (after DOC)

Figure 26

Comparisons of measurements and observer estimations (after DPF)

Figure 27

Comparisons of measurements and observer estimations at higher speeds (after DPF)

## Errata

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 Proceedings Articles
Related eBook Content
Topic Collections