 |
|
| Direct dynamic instantaneous emission modelling (2001 - 2005) |
 |
Road traffic accounts for an important part of air pollution. Therefore, it is necessary to quantify the emission levels as accurately as possible. Due to the quantity of information necessary to determine the different parameters related to the traffic emissions, direct measurement becomes impractical and expensive. Thus, models for predicting emissions, although expensive to develop, represent an alternative to direct measurement.
For more than a decade attempts have been made to store or map emission measurements of test on chassis dynamometers or engine test benches in a neutral way, such that emissions of other driving conditions can be calculated out of them.
The terms “instantaneous”, “modal”, “continuous”, “online” emission models are used as synonyms. The instantaneous emissions and fuel consumption models operate at a high level of complexity. They are designed to provide an estimation technique at a microscale or mesoscale level, such as for streets, cities or regions. Extensive vehicle measurements are necessary to provide vehicle operation, emissions and fuel consumption data at a high time resolution (one to ten Hz). The data are then analysed in terms of vehicle's events (modes) such that instantaneous emissions and fuel consumption rates can be estimated on a 10 Hz basis (Ajtay et al., 2004a). |
|
 |
| Figure 1: Exhaust gas transport systems |
| | Usually, instantaneous emission modelling maps emissions at a given time to their generating “engine state”, like speed, acceleration, engine power, etc. This makes it possible to integrate new, unmeasured driving patterns over the model and calculate their emission factors without further measurements. |
|
But, the original emission signals measured in a test are delayed to their time of formation, since the exhaust gas is transported from the engine to the analysers (see Figure 1), and the emission peaks are flattened by convolution.
If these dynamic aspects of the exhaust transport are neglected, the emission events are correlated to “the wrong second”, resulting in incorrect engine status in emission modelling. For instantaneous emission modelling, emission values can be correlated to the correct engine state of the car only if they are at their correct location on the time scale. Therefore, these delays and mixing dynamics must be compensated, i.e. the behaviour of the gas transport systems must be modelled.
For the modelling purposes, the exhaust gas transport model may be split up into three parts, with increasing level of complexity:
- Modelling of the raw gas measurement system, as a linear, time-invariant differential system of order three (Weilenmann et al., 2003).
- Modelling of the exhaust system of the car, as a linear, time-varying system of order one (Ajtay et al., 2004b).
- Modelling of the dilution measurement system (an alternative to the raw measurement) as a nonlinear, time-varying system of order six (Ajtay et al., 2004b).
|
|
 |
| Figure 2: Reconstruction of catalyst-out signal from the signal recorded at the analyser |
| | With this approach, we are now able to reconstruct the emission signals at their location of formation (engine-out, catalyst-out) from the signal at the analyzer with a time quality of about one second (Figure 2). |
|
 |
| Figure 3: Typical NOx map |
| | Using the reconstructed signals, the mapping of emissions is performed by relating them to engine speed and engine torque (Figure 3). |
|
 |
| Figure 4: Quality prediction for a Euro-2 diesel vehicle |
| | This static instantaneous model can accurately predict the emission factors for various driving situations (rural, highway, urban, urban with stops-and-go) for vehicles without after treatment system. Figure 4 shows the quality of emission factors prediction for a diesel Euro-2 vehicle. |
|
 |
| Figure 5: Measured and simulated engine-out emissions of a Euro-3 gasoline vehicle, during a highway cycle |
| | In order to take into account the transient generation of emissions for the vehicles with after treatment system, the model was extended by adding a dynamic variable. At this moment, we are able to predict engine-out emissions, and not just the integrated values (i.e. emission factors), but also the instantaneous ones (Figure 5). This dynamic model is going to be linked with a catalyst model, such that more accurate predictions for emissions of vehicles with after treatment systems shall be obtained. |
|
Literature
[1] Ajtay D., Weilenmann M.: Static and dynamic instantaneous emission modelling, International Journal of Environment and Pollution, Vol. 22, No. 3, p. 226-239, 2004a.
[2] Ajtay D., Weilenmann M.: Compensation of the exhaust gas transport dynamics for accurate instantaneous emission measurements, Environmental Science and Technology, Vol. 38, No. 19, p. 5141-5148, 2004b.
[3] Weilenmann M., Soltic P., Ajtay D.: Describing and Compensating Gas Transport Dynamics for Accurate Instantaneous Emission Measurements, Atmospheric Environment, Vol. 37, p. 5137-5145, 2003. |
|
| |
|
|
|
|
|
|
back
Links
Send article
print view