Gasoline particulate filters (GPFs) are practically adoptable devices to mitigate particulate matter emissions
from vehicles using gasoline direct ignition engines. This paper presents a newly developed control-oriented
model to characterize the thermal and soot oxidation dynamics in a ceria-coated GPF. The model utilizes the
GPF inlet exhaust gas temperature, exhaust gas mass flow rate, the initial GPF soot loading density, and air–
fuel ratio to predict the internal GPF temperature and the amount of soot oxidized during regeneration events.
The reaction kinetics incorporated in the model involve the rates of both oxygen- and ceria-initiated soot
oxidation reactions. Volumetric model parameters are calculated from the geometric information of the coated
GPF, while the air–fuel ratio is used to determine the volume fractions of the exhaust gas constituents. The
exhaust gas properties are evaluated using the volume fractions and thermodynamic tables, while the cordierite
specific heat capacity is identified using a clean experimental data set. The enthalpies of the regeneration
reactions are calculated using thermochemical tables. Physical insights from the proposed model are thus
enhanced by limiting the number of parameters obtained from fitting to only those which cannot be directly
measured from experiments. The parameters of the model are identified using the particle swarm optimization
algorithm and a cost function designed to simultaneously predict both thermal and soot oxidation dynamics.
Parameter identification and model validation are performed using independent data sets from laboratory
experiments conducted on a ceria-coated GPF. This work demonstrates that the proposed model can be
successfully implemented to predict ceria-coated GPF dynamics under different soot loading and temperature
conditions.
more »
« less
Initial Development of a Physics-Aware Machine Learning Framework for Soot Mass Prediction in Gasoline Direct Injection Engines
Calibration of automotive engines to ensure compliance with emission regulations is a critical phase in product development. Control of engine-out particulate emissions, which directly impact the environment and public health, is particularly important. Detailed physics-based models are typically used to gain a rich understanding of the complex physical phenomena that drive the soot particle formation in an engine cylinder. However, such models often fail to correctly represent the highly dynamic nature of the underlying mechanisms under transient combustion conditions. Moreover, most physics-based models were initially developed for diesel engine applications and their applicability to gasoline engines remains questionable due to differences in flame structure and fuel-wall interactions. Black-box models have been previously proposed to predict engine-out soot emissions, but their lack of physical interpretability is an unsolved drawback. To address these limitations, we present a physics-aware twin-model machine learning framework to predict and analyze engine-out soot mass from a gasoline direct injection (GDI) engine. The framework combines a physics-based model with a bagging-type ensemble learning model that both maintains high accuracy and allows physical interpretation of results without using computationally intensive high-fidelity models. This work shows why a one-model-fits-all approach fails in the case of predicting soot emissions due to clustered co-occurrences of operating conditions that cause non-compliant behavior. We compare the performance of the proposed framework with that of the standalone baseline model and a feed-forward deep neural network. Using WLTP data from a 2.0L naturally aspirated GDI engine, the proposed framework predicts engine-out soot mass with an improvement of 29% in the R2 value and 21% in the root mean squared error from the baseline physics-based model, without compromising physical interpretability. These improvements are significant enough to warrant further framework development with additional engine datasets.
- Award ID(s):
- 1901099
- NSF-PAR ID:
- 10524840
- Publisher / Repository:
- 16th International Conference on Engines & Vehicles
- Date Published:
- Format(s):
- Medium: X
- Location:
- Capri, Italy
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
A fractal engine simulation (FES) sub-model was integrated into three-dimensional simulations for modeling turbulent combustion for a gasoline direct injection (GDI) engine. The FES model assumes that the effects of turbulence on flame propagation are to wrinkle and stretch the flame, and fractal geometry is used to predict the surface area increase and thus the turbulent burning velocity. Different formulas for the four sequential stages of combustion in SI engines are proposed to account for the changing effects of turbulence throughout the combustion process. However, most prior studies related to the FES model were quasi-dimensional simulations, with few found in multi-dimensional studies, and none under cold start conditions or stratified charges. This paper describes how the model was implemented into multidimensional simulations in CONVERGE CFD, and what the formulas are in the four sequential stages of combustion in SI engines. The capabilities of the FES model for simulating the cold start cases, under the conditions of the dramatically changing engine speed and mixture stratification in a complex engine geometry, are presented in this study. The FES model was able to not only simulate the steady-state cases with constant engine speed, but also predict the in-cylinder pressure traces in all four cylinders for the very first firing cycle with transient engine speed, and gave good agreement with the experimental measurements under these extremely transient conditions. The uncertain maximum fractal dimension was chosen as 2.37 in this research, and a simple linear correlation with engine speed was used to obtain the coefficient used in calculating the kernel formation time which controls the so-called combustion or ignition delay.
-
SAE (Ed.)more » « lessAn investigation of the performance and emissions of a Fischer-Tropsch Coal-to-Liquid (CTL) Iso-Paraffinic Kerosene (IPK) was conducted using a CRDI compression ignition research engine with ULSD as a reference. Due to the low Derived Cetane Number (DCN), of IPK, an extended Ignition Delay (ID), and Combustion Delay (CD) were found for it, through experimentation in a Constant Volume Combustion Chamber (CVCC). Neat IPK was analyzed in a research engine at 4 bar Indicated Mean Effective Pressure (IMEP) at three injection timings: 15°, 20°, and 25° BTDC. Combustion phasing (CA50) was matched with ULSD at 10.8° and 16° BTDC. The IPK DCN was found to be 26, while the ULSD DCN was significantly higher at 47 in a PAC CID 510. In the engine, IPK’s DCN combined with its short physical ignition delay and long chemical ignition delay compared to ULSD, caused extended duration in Low Temperature Heat Release (LTHR) and cool flame formation. It was found in an analysis of the Apparent Heat Release Rate (AHRR) curve for IPK that there were multiple Negative Temperature Coefficient (NTCR) regions before the main combustion event. The High Temperature Heat Release (HTHR) of IPK achieved a greater peak heat release rate compared to ULSD. Pressure rise rate for IPK was observed to increase significantly with increase in injection timing. The peak in-cylinder pressure was also greater for IPK when matching CA50 by varying injection timing. Emissions analysis revealed that IPK produced less NOx, soot, and CO2 compared to ULSD. CO and UHC emissions for IPK increased.
-
A parametric study was carried out for the first firing cycle of a 4-cylinder, 2.0-liter, turbocharged gasoline direct injection (GDI) engine. The primary goal was to see how changes in the fuel injection parameters would affect the GDI engine combustion and emissions for the first four combustion events that constitute the first firing cycle. Experimental studies were carried out with a custom-designed powertrain control system to measure the HC emissions and pressure development for the first firing cycle. The quantitative experimental results were accompanied by simulations of the detailed temporal and spatial fuel concentration profiles using Converge CFD engine simulation software. An alternative calculation method was used to calculate the average combustion equivalence ratio for each of the four cylinders. This method showed that the majority of the cold start HC emissions during the first firing cycle was unburned gasoline and its possible decomposition products, which did not contribute significantly to the combustion and heat release. For the same amount of fuel injected into a cylinder, increased fuel rail pressure resulted in better evaporation and combustion, while slightly increasing the HC emissions during the cold start process. A multiple injection strategy was studied that split the fuel delivery between the intake stroke and the compression stroke with either one or two injections in each of those strokes (two or four injections total). The quadruple injection strategy led to better first cycle combustion, with higher engine IMEP and lower HC emissions. This resulted from a richer fuel mixture in the region near the spark plug due to better fuel evaporation and a better spatial fuel distribution. While increasing fuel rail pressure with either injection strategy failed to significantly lower the HC emissions given the same amount of injected fuel mass, higher rail pressure with the quadruple injection strategy resulted in higher IMEP for the same amount of injected fuel; this may provide the possibility to reduce the total fuel injection mass which may have benefits for both fuel consumption and emissions.more » « less
-
Reactivity controlled compression ignition (RCCI) engines center on a combustion strategy with higher thermal efficiency, lower particulate matter (PM), and lower oxides of nitrogen (NOx) emissions compared to conventional diesel combustion (CDC) engines. However, real time optimal control of RCCI engines is challenging during transient operation due to the need for high fidelity combustion models. Development of a simple, yet accurate control-oriented RCCI model from physical laws is time consuming and often requires substantial calibrations. To overcome these challenges, data-driven models can be developed. In this paper, a data-driven linear parametervarying (LPV) model for an RCCI engine is developed. An LPV state space model is identified to predict RCCI combustion phasing as a function of multiple RCCI control variables. The results show that the proposed method provides a fast and reliable route to identify an RCCI engine model. The developed model is then used for the design of a model predictive controller (MPC) to control crank angle for 50% fuel burnt (CA50) for varying engine conditions. The experimental results show that the designed MPC with the data-driven LPV model can track desired CA50 with less than 1 crank angle degree (CAD) error against changes in engine load.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.4271/2023-24-0174
