Evaluation of Virtual NOx Sensor Models for Off Road Heavy Duty Diesel Engines

Stadlbauer, Stephan; Alberer, Daniel; Hirsch, Markus; Formentin, Simone; Benatzky, Christian; Re, Luigi del

doi:10.4271/2012-01-0358

Cited by 19 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the previous paper, 25 it has been already proposed that the overall in-cylinder pressure trace can be reconstructed by only the most important principal components. In our case, the first four of them are shown in the upper subplot of Figure 4.…”

Section: Combustion Information Extractionmentioning

confidence: 99%

Real time soot models for production CI engines based on pressure trace information

Zhang

Waschl

Fuerhapter

et al. 2016

International Journal of Engine Research

Self Cite

View full text Add to dashboard Cite

Better knowledge of the instantaneous soot and nitrogen oxides (NOx) emissions in Diesel engines would allow better choices of injection parameters and thus lead to better raw emissions than, for instance, with classical averaged measures, like smoke maps. Unfortunately, while for most regulated emissions fast measurements are possible and reasonably fast sensors are available, the same is not true for soot. Against this background, this paper proposes a real time, delay-free soot model based on the pressure trace information. The model essentially maps changes of the pressure trace into changes of the raw soot emissions. The key elements of the approach are the design of experiments, segmentation of the pressure trace and the use of the principal component analysis technique to extract the essential information. The model is based only on data; that is, it needs an initial calibration with working points measured with a standard device, such as an AVL Microsoot or Opacimeter. This paper describes the method and then shows validation results with an additional amount of pilot injection in one cylinder which confirms the delay-free soot estimation.

show abstract

Section: Combustion Information Extractionmentioning

confidence: 99%

Real time soot models for production CI engines based on pressure trace information

Zhang

Waschl

Fuerhapter

et al. 2016

International Journal of Engine Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…This approach, discussed in more detail in Formentin et al 17 and Stadlbauer et al, 18 is similar to the “ECU approach” since the general idea is to also look at the A/F ratio inside the cylinder during combustion. However, unlike directly taking into account variables from air and fuel path, the focus is placed on the in-cylinder pressure trace.…”

Section: Modeling Approachesmentioning

confidence: 99%

“…16 Recent results employ feature extraction from the in-cylinder pressure trace via singular value decomposition (SVD) utilizing them as inputs to polynomial NARX models. 17,18…”

Section: Introductionmentioning

confidence: 99%

Indicated pressure-based data-driven diesel engine NOx modeling

Benatzky

Stadlbauer

Formentin

et al. 2014

International Journal of Engine Research

Self Cite

View full text Add to dashboard Cite

This article addresses the design and evaluation of virtual NO x sensors for heavy-duty off-road diesel engines based on static polynomial black box modeling. Three approaches, differing in the chosen sets of regressors, are analyzed regarding their NO x prediction capability. As regressor sets only quantities available on standard production-type electronic control units, features extracted from the in-cylinder pressure trace via singular value decomposition extended by the engine speed as well as geometric values from the pressure trace and heat release curves are utilized, respectively. It is shown that while each of the approaches alone has its drawbacks, a systematic combination, especially of the first two methods, results in high-accuracy NO x models while at the same time keeping the number of regressors low.

show abstract

“…This method has been used and validated in a variety of applications (see e.g. Stadlbauer et al (2012) and Passenbrunner et al (2014)) in automotive control. The simple nature of these models is also well suited as a basis for nonlinear optimal control design.…”

Section: Introductionmentioning

confidence: 99%