Search Results

The paper describes the results of a parallel 1D thermo-fluid dynamic simulation and experimental investigation of a DI turbocharged Diesel engine. The attention has been focused on the overall engine performances (air flow, torque, power, fuel consumption) as well as on the emissions (NO and particulate) along the after-treatment system, which presents a particulate filter. The 1D research code GASDYN for the simulation of the whole engine system has been enhanced by the introduction of a multi-zone quasi-dimensional combustion model for direct injection Diesel engines. The effect of multiple injections is taken into account (pilot and main injection). The prediction of NO and soot has been carried out respectively by means of a super-extended Zeldovich mechanism and by the Hiroyasu kinetic approach.

An experimental and theoretical work on the low temperature oxidation of n-heptane in a jet stirred reactor has been carried out at different inlet temperatures. The presence of the typical low temperature pathologies of hydrocarbons (slow combustion, periodic and dumped cool flames) have been observed experimentally and correctly reproduced by the model. The selectivities of the intermediate and final products are also measured and compared with the theoretical evaluations. The agreement is satisfactory for all the investigated species in the whole temperature range (550–800 K). The introduction of 40% (volume) in the fuel has allowed to investigate the antiknock effect of toluene on the autoignition of n-heptane. At the same inlet temperature the n-heptane conversion shows the same general behaviour, but it is about 10% lower when toluene is fed in the mixture.

A computational, three-dimensional approach to investigate the behavior of diesel soot particles in the micro-channels of wall-flow Diesel Particulate Filters is presented. The KIVA3V CFD code, already extended to solve the 2D conservation equations for porous media materials [1], has been enhanced to solve in 2-D and 3-D the governing equations for reacting and compressible flows through porous media in non axes-symmetric geometries. With respect to previous work [1], a different mathematical approach has been followed in the implementation of the numerical solver for porous media, in order to achieve a faster convergency as source terms were added to the governing equations. The Darcy pressure drop has been included in the Navier-Stokes equations and the energy equation has been extended to account for the thermal exchange between the gas flow and the porous wall.

Nowadays 1D fluid dynamic models are widely used by engine designers, since they can give sufficiently accurate predictions in short times, allowing to support the optimization and development work of any prototype. According to the last requirements in terms of pollutant emission control, some enhancements have been introduced in the 1D code GASDYN, to improve its ability in predicting the composition of the exhaust gas discharged by the cylinders and the transport of the chemical species along the exhaust system. The main aspects of the methods adopted to model the combustion process and the related formation of pollutants are described in the paper. To account for the burnt gas stratification, two different approaches have been proposed, depending on the expected turbulence levels inside the combustion chamber. The reliability of the simulation of the pollutant formation process has been enhanced by the integration of the thermodynamic module with the Chemkin code.

The paper describes recent advances in the research work concerning the 1-d fluid dynamic modeling of unsteady flows in i.c. engine pipe systems. A comprehensive simulation model has been developed, which is based on different numerical techniques for the solution of the fundamental conservation equations. Classical (MacCormack method plus TVD algorithm) and innovative (the CE-SE method, the discontinuous Galerkin FEM) shock-capturing schemes have been compared, considering the shock-tube problem and the shock-turbulence interaction problem. Moreover, the tracking of the chemical species along the intake and exhaust duct systems has been investigated, introducing the species continuity equations in the numerical model. The engine test case reported in the paper points out the predicted transport of chemical species in the ducts.

Aim of this work is to present and discuss the possibility and the limits of two zone models for spark-ignition engines using a detailed kinetic scheme for the characterization of the evolution of the air-fuel mixture, while an equilibrium approach is used for the burnt zone. Simple experimental measurements of knocking tendency of different fuels in ideal reactors, such as rapid compression machines and shock tube reactors, cannot be directly used for the analysis of octane numbers and sensitivity of hydrocarbon mixtures. Thus a careful investigation is very useful, not only of the combustion chamber behavior, including the modelling of the turbulent flame front propagation, but also of the fluid dynamic behavior of the intake and exhaust system, accounting for the volumetric efficiency of the engine.

This work presents a general model for the prediction of octane numbers and knock propensity of different fuels in SI engines. A detailed kinetic scheme of hydrocarbon oxidation is coupled with a two zone, 1-D thermo-fluid dynamic simulation code (GASDYN) [1]. The validation of the kinetic scheme is discussed on the basis of recent experimental measurements. CFR engine simulations for RON and MON evaluation are presented first to demonstrate the capabilities of the coupled model. The model is then used to compare the knock propensity of a gasoline “surrogate” (a pure hydrocarbon mixture) and PRFs in a current commercial engine, resulting in a simulation of “real world” octane number determination, such as Bench Octane Number (BON). The simulation results agree qualitatively with typical experimental trends.

The paper describes the 1-D fluid dynamic modeling of unsteady flows with chemical specie tracking in the ducts of a four-cylinder s.i. automotive engine, to predict the composition of the exhaust gas reaching the catalyst inlet. A comprehensive simulation model, based on classical and innovative numerical techniques for the solution of the governing equations, has been developed. The non-traditional shock-capturing CE-SE (Conservation Element-Solution Element) method has been extended to deal with the propagation of chemical species. A comparison of the MacCormack method plus FCT or TVD algorithms with the CE-SE method has pointed out the superiority of the latter scheme in the propagation of contact discontinuities. A realistic composition of the exhaust products in the cylinder, evaluated by a two-zone combustion model including emission sub-models, has been imposed at the opening of the exhaust valve, considering the effect of short-circuit of air during valve overlap.

Aim of this work is to develop a general purpose model for combustion and knocking prediction in SI engines, by coupling a thermo-fluid dynamic model for engine simulation with a general detailed kinetic scheme, including the low-temperature oxidation mechanism, for the prediction of the auto-ignition behavior of hydrocarbons. A quasi-D approach is used to describe the in-cylinder thermodynamic processes, applying the conservation of mass and energy over the cylinder volume, modeled as a single open system. The complex chemistry model has been embedded into the code, by using the same integration algorithm for the conservation equations and the reacting species, and taking into account their mutual interaction in the energy balance. A flame area evolution predictive approach is used to evaluate the turbulent flame front propagation as function of the engine operating parameters.

The paper describes the development and validation of a quasi-dimensional multi-zone combustion model for Gasoline Direct Injection engines. The model has been embedded in the 1D thermo-fluid-dynamic code for the simulation of the whole engine system named GASDYN and developed by the authors [1, 2 and 3]. The GDI engine combustion model solves mass, energy and species equations using a 4th order Runge-Kutta integration method; the fuel spray is initially divided into a number of zones fixed regardless of the injected amount and the time step, considering the following break-up, droplet evaporation and air entrainment in each single zone. Experimental correlations have been used for the spray penetration and spatial information. Once the ignition begins it is assumed that the flame propagates spherically, evaluating its velocity by means of a fractal combustion approach and considering the local air-fuel ratio, which is the result of the spray evolution within the combustion chamber.

A 1-D thermo-fluid dynamic simulation code, including a quasi-D combustion model coupled with a detailed kinetic scheme, is used to analyze the combustion process in HCCI engines. The chemical mechanism has previously been validated in comparison with experimental data over a wide range of operating conditions. To explore the impact on model predictions, the cylinder was divided into multiple zones to characterize the conditions of the in-cylinder charge. Particular attention is devoted to the numerical algorithm in order to ensure the robustness and efficiency of the large system solution. This numerical model allows study of the autoignition of the air fuel mixture and determines the chemical evolution of the system. The proposed model was compared with in-cylinder temperature and chemical species profiles. The experimental activity was carried out in the combustion chamber of a single cylinder air cooled engine operating in HCCI mode.

Nowadays a great attention is paid to the level and quality of noise radiated from the tailpipe end of intake and exhaust systems, to control the gas dynamic noise emitted by the engine as well as the characteristics of the cabin interior sound. The muffler geometry can be optimized consequently, to attenuate or remark certain spectral components of the engine noise, according to the result expected. Evidently the design of complex silencing systems is a time-consuming operation, which must be carried out by means of concurrent experimental measurements and numerical simulations. In particular, 1D and multiD linear/non-linear simulation codes can be applied to predict the silencer behavior in the time and frequency domain. This paper describes the development of a 1D-multiD integrated approach for the simulation of complex muffler configurations such as reverse chambers with inlet and outlet pipe extensions and perforated silencers with the addition of sound absorbing material.

A new approach for the fluid-dynamic simulation of the Diesel Particulate Filters (DPF) has been developed. A mathematical model has been formulated as a system of nonlinear partial differential equations describing the conservation of mass, momentum and energy for unsteady, compressible and reacting flows, in order to predict the hydrodynamic characteristics of the DPF and to study the soot deposition mechanism. In particular, the mass conservation equations have been solved for each chemical component considered, and the advection of information concerning the chemical composition of the gas has been figured out for each computational mesh. A sub-model for the prediction of the soot cake formation has been developed and predictions of soot deposition profiles have been calculated for different loading conditions. The results of the simulations, namely the calculated pressure drop, have been compared with the experimental data.

The present work focuses on the simulation of the hydrodynamics, transient filtration/loading and catalytic/NO2-assisted regeneration of Diesel after-treatment systems. A 1D unsteady model for compressible and reacting flows for the numerical simulation of the behavior of Diesel Oxidation Catalysts (DOCs) and Diesel Particulate Filters (DPFs) has been developed. The numerical model is able to keep track of the amount of soot in the flow; the increasing of back-pressure through the exhaust system (mainly due to the Diesel Particulate Filter) can be predicted by the calculation of the permeability variation of the porous wall, as the soot particles goes inside the DPF. A sub-model for the regeneration of the collected soot has been developed: the collected particulate is oxidized by the Oxygen (O2) and by the Nitrogen Dioxide (NO2).