A computational lumped parameter model (MTU-Filter- Lumped) was developed to describe the performance of diesel particulate filters (DPFs) during loading and regeneration processes. The model was formulated combining three major sub-models: a filtration model, a pressure drop model, and a mass and an energy balance equation for the total filter volume. The first two submodels have been widely validated in the literature, while the third sub-model is introduced and combined with the first two sub-models in the present study. The three sub-models combined can give a full description of diesel particulate filter behavior during loading and regeneration processes, which was the objective of the present work. The total combined lumped parameter model was calibrated using experimental data from the literature covering a range of experimental conditions, including different catalytic regeneration means and engine- operating conditions. The model predictions showed very good agreement with the experimental data in terms of pressure drop across the filter, mass retained in the filter, and filter temperature. A diesel particulate filter system was selected to illustrate the control application of the lumped model equations. This system involves a diesel particulate filter for the collection and oxidation of the engine out particulate matter emissions, and the injection of hydrocarbons upstream of an oxidation catalytic converter (OCC) in order to raise the exhaust gas temperature and in turn achieve filter regeneration. Two model-based control strategies were developed aiming to minimize the fuel penalty of the regeneration process described above.
A one-dimensional, two layer computational model was developed to predict the behavior of a clean and particulate-loaded catalyzed wall-flow diesel particulate filter (CPF). The model included the mechanisms of particle deposition inside the CPF porous wall and on the CPF wall surface, the exhaust flow field and temperature field inside the CPF, as well as the particulate catalytic oxidation mechanisms accounting for the catalyst-assisted particulate oxidation by the catalytic coating in addition to the conventional particulate thermal oxidation. The paper also develops the methodology for calibrating and validating the model with experimental data. Steady state loading experiments were performed to calibrate and validate the model. The experimental data were collected on a Corning EX-80 cordierite filter (100 cpi) with a loading of 5-g/ft 3 Pt in the MEX catalyst type formulation using a 1995 Cummins M11-330E heavy-duty diesel engine with manual EGR and conventional low sulfur fuel (375 ppm sulfur). Good agreement was obtained between the model predictions for pressure drop, particulate mass filtration efficiency, particulate mass retained, and filter temperature profile. The model was also used with experimental data to estimate the filter clean wall permeability, the packing density of the particulate deposited inside the filter wall, the particulate layer packing density, porosity, and permeability, as well as the activation energy and frequency factor of the particulate thermal and catalytic oxidation, including the maximum thickness of the particulate layer in contact with the catalyst. The particulate layer packing density, porosity, and permeability were then correlated with the exhaust gas Peclet number to provide a better understanding of particulate properties under various engine loads. This model can be used as a tool to predict the pressure drop across the CPF, the particulate mass filtration efficiency, the downstream particulate concentration, and the particulate mass retained inside the CPF for different filter geometries and physical properties, as well as for different engine operating conditions. In conjunction with designed experiments, the model can also be utilized to characterize (by determing the model constants) catalyzed particulate filters with different catalysts and catalyst loadings.
The goal of the project was to demonstrate that low pressure loop EGR incorporating a diesel oxidation catalyst (DOC) and a diesel particulate filter (DPF) can be applied to an off-highway engine to meet Tier 3 (Task I) and Interim Tier 4 (Task II) off-road emissions standards. Task I data was collected using a John Deere 8.1 liter engine modified with a low pressure loop EGR system. The engine and EGR system was optimized and final data over the ISO 8178 eight mode test indicated the NOx emissions were less than 4 g/kWh and the PM was less than 0.02 g/kWh which means the engine met the Tier 3 off-road standard. Considerable experimental data was collected and used by Michigan Tech University to develop and calibrate the MTU-Filter 1D DPF model. The MTU-Filter 1D DPF code predicts the particulate mass evolution (deposition and oxidation) in the diesel particulate filter (DPF) during simultaneous loading and during thermal and NO{sub 2}-assisted regeneration conditions. It also predicts the pressure drop across the DPF, the flow and temperature fields, the solid filtration efficiency and the particle number distribution downstream of the DPF. A DOC model was also used to predict the NO{sub 2} upstream of the DPF. The DPF model was calibrated to the experimental data at temperatures from 230 C to 550 C, and volumetric flow rates from 9 to 39 actual m{sup 3}/min. Model predictions of the solid particulate mass deposited in the DPF after each loading and regeneration case were in agreement within +/-10g (or +/-10%) of experimental measurements at the majority of the engine operating conditions. The activation temperatures obtained from the model calibration are in good agreement with values reported in the literature and gave good results in the model calibration by using constant pre-exponential factors throughout the entire range of conditions evaluated. The average clean filter permeability was 2.372 x 10{sup -13} m{sup 2}. Estimates of the solid particulate mass packing density inside the porous wall were 1 to 5 kg/m{sup 3}; and percolation factors were 0.81 to 0.97. Average particulate layer permeability was 1.95 x 10{sup -14} m{sup 2}. Solid particulate layer packing density values were between 11 and 128 kg/m{sup 3}. These values were in good agreement with the Peclet number correlation theory reported in the literature. NO{sub 2}-assisted oxidation of PM in the DPF showed experimentally that a significant reduction of the pressure drop can be achieved (<8 kPa) when sufficient NO{sub 2} (>120 ppm) is available and high exhaust gas temperatures ({approx}360-460 C) can be maintained, even at high PM loadings (low NO{sub 2}/solid PM ratios). The CRT{trademark} (DOC-DPF system) showed limited advantages when used with high PM rates (low NOx/PM ratios) in combination with a low pressure loop EGR strategy for a continuous operation of an engine-exhaust aftertreatment system. The 8.1-liter engine was not designed for low-pressure loop EGR and when the EGR was added the NOx emissions were reduced but the PM emissions increased. This corresponds to the well known NOx to PM relationship in which if the NOx is reduced the PM emissions increase. In order for this technology to be successful on this engine family, the engine out PM emissions must be reduced. These results led to Task II. Task II objective was to meet the interim Tier 4 standards using the CCRT{trademark} technology applied to an advanced 6.8 liter John Deere engine. The advanced engine incorporated a 4 valve head, required additional EGR, an advanced high pressure common rail fuel system and a better matched turbocharger. The EGR system was optimized and the goal of less than 2 g/kWh NOx and less than 0.02 g/kWh PM were achieved over the 8 mode test. Again, experimental data was provided to Michigan Tech to study the passive regeneration of the CCRT{trademark} technology. Two computer models, i.e., the MTU 1-D DOC model and the MTU 1-D 2-layer CPF model were developed as part of this research and calibrated using the data obtained from experiments. The 1-D DOC model employs a three-way catalytic reaction scheme for CO, HC and NO oxidation, and is used to predict CO, HC, NO and NO{sub 2} concentrations downstream of the DOC. The 1-D 2-layer CPF model used '2-filters in series' approach for filtration, PM deposition and oxidation in the PM cake and substrate wall via thermal (O{sub 2}) and NO{sub 2}/temperature-assisted mechanisms, and production of NO{sub 2} as the exhaust gas mixture passes through the CPF catalyst washcoat. The bottom line is the MTU models were improved and the models better predict the pressure drop across the DOC and CPF and the models do a good job estimating the amount of PM entering the CPF and the amount oxidized in the CPF and the amount exiting. The idea is to use this information to predict how much soot is in the DPF and predict when active regeneration is needed.
A 1-D 2-layer model developed previously at MTU was used in this research to predict the pressure drop, filtration characteristics and various properties of the particulate filter and the particulate deposit layer. The model was calibrated and validated for this CPF with data obtained from steady state experiments conducted using a 1995 Cummins M11-330E heavy-duty diesel engine with manual EGR and using ULSF. The CPF used is a NGK filter having a cordierite substrate with NEX catalyst type formulation (54% porosity, 15.0 μm mean pore diameter and 50 gms/ft 3 Pt). The filter was catalyzed using a wash coat process. The model was used to predict the pressure drop, particulate mass retained inside the CPF, particulate mass filtration efficiency and concentration downstream of the CPF with agreement between the experimental and simulated data. The model was also used to predict the clean substrate permeability, packing density of the particulate deposited inside the substrate, permeability and packing density of the particulate layer deposited on the substrate, activation energies and frequency factors for the thermal and catalytic oxidation path and the thickness of the particulate layer I, which is in contact with the catalyst. In addition, the model also predicted the velocities in the inlet channel, outlet channel and through the wall, variation of particulate layer thickness with axial distance and time along with the particle number distribution downstream of the CPF.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)