Search Results

The flow in engine turbocharger compressors and turbines is highly unsteady in nature, as it responds to the intake and exhaust manifolds of the internal combustion engine. The optimization of the turbocharger system is therefore a very difficult task, since the devices operate at off-design conditions for most of the engine cycle. Experimental studies allow for improving the understanding on the behavior of the engine components, in particular when tests are performed under real engine operating conditions; however, the experimental tests can be more efficient if they are combined with theoretical simulation tools, which help to select significant engine operating conditions. In this paper, experimental investigations were performed on a flexible component test rig (expressly suited to perform tests on automotive turbochargers) at ICE Laboratory of the University of Genoa (ICEG-DIMSET).

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.

The paper describes the research work concerning the simulation of 1D unsteady reacting flows in s.i. engine pipe-systems, including pre-catalysts and main catalysts. The numerical model GASDYN has been developed to enable the concurrent prediction of the wave motion in the intake and exhaust ducts, the chemical composition of the gas discharged by the cylinder of a s.i. engine, the chemical and thermal behavior of catalytic converters. The effect of considering the transport of chemical species with reactions in gas phase (post-oxidation of unburned HC in the exhaust manifold) and in solid phase (conversion of pollutants in the catalyst) on the predicted wave motion is reported.

The paper describes the experimental and simulation work recently carried out to investigate the effects of secondary air injection on the emission conversion in the exhaust after-treatment system of a S.I. automotive engine. The modeling of the 1D unsteady reacting flows in the complete exhaust system of a spark ignition engine, designed to satisfy the Euro IV limits, has been performed including the secondary air injection system, to predict the possible shortening of catalyst light-off time and the speed-up of the after-treatment system warm-up. The transport of chemical species with reactions in gas phase (post-oxidation of unburned HC in the exhaust manifold) and in solid phase (conversion of pollutants in the catalyst) with and without secondary air has been simulated by the 1D thermo-fluid dynamic model GASDYN, developed by the authors.

The paper describes the research work carried out on the thermo-fluid dynamic modeling of an S.I. engine coupled to the vehicle in order to predict the engine and tailpipe emissions during the ECE European driving cycle. The numerical code GASDYN has been extended to simulate the engine + vehicle operation during the first 90 seconds of the NEDC driving cycle, taking account of the engine and exhaust system warm-up after the cold start. The chemical composition of the engine exhaust gas is calculated by means of a thermodynamic multi-zone combustion model, augmented by kinetic emission sub-models for the prediction of pollutant emissions. A simple procedure has been implemented to model the vehicle dynamic behavior (one degree of freedom model). A closed-loop control strategy (proportional-derivative) has been introduced to determine the throttle opening angle, corresponding to the engine operating point when the vehicle is following the ECE cycle.

The unsteady flows in the exhaust system of a single cylinder Diesel engine have been simulated by a 1-d fluid dynamic code, and the pulse noise spectrum radiated by the open termination has been calculated. The simulation model developed is based on different symmetric shock-capturing schemes for the solution of the 1-d conservation equations: the classical MacCormack and two-step Lax-Wendroff methods, and the new Conservation Element-Solution Element method (CE-SE). Several complex exhaust muffling systems have been modeled, such as Helmholtz resonators, expansion chambers, perforates, achieving a satisfactory agreement between predicted and measured results.

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.

Increasing demands on the capabilities of engine simulation and the ability to accurately predict both performance and acoustics has lead to the development of multiple approaches, ranging from fully 3D to simplified 1D models. In this work it will be described the development and application of hybrid 1D-3D approaches and an innovative one based on the 3D cell element. This is designed to model the acoustics of intake and exhaust system components used in internal combustion engines. Models of components are built using a network or grid of 3D cells based primarily on the geometry of the system. This means that these models can be built without fundamental knowledge of acoustically equivalent systems making their range of application larger as well as making them simpler to construct. Due to the 3D nature of these models it is also possible to predict higher order modes and improve the accuracy of models at high frequencies compared to conventional plane wave approaches.

This paper describes the development of a fully 1D and of a 1D-multiD integrated approach for the simulation of complex muffler configurations. The fully 1D approach aims to model the muffler recurring to an equivalent net of 1D pipes. An expansion chamber with offset inlet and outlet pipes was modeled with this preocedure and the resuts compared to CFD simulations, pointing out some critical aspects in the TL prediction. The HLLC Riemann solver and its extension to the second order were implemented both in the 1D and multiD models and exploited to handle the interface between the calculation domains. The integrated 1D-multiD approach was used afterwards to predict the transmission loss of more complex geometries such as series chambers with extended inlet and outlet pipes and with flow reversals. A new procedure has been adopted to calculate the transmission loss, imposing a pressure impulse at the inlet and evaluating the response of the muffler.

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 describes the development, application and coupling of two different numerical codes, respectively based on a 1D (Gasdyn) and 3D (OpenFOAM) schematization of the geometrical domain. They have been adopted for the prediction of the wave motion inside the intake and the exhaust systems of internal combustion engines. The HLLC Riemann solver has been implemented both in the CFD and the 1D codes to solve the Euler system of equations, in order to operate with the same solver on the different calculation domains. Moreover, the HLLC solver has been applied to treat the boundary conditions at the interface between the two domains, in such a way to allow the propagation of flow disuniformities through the domain interface, without affecting the solution accuracy. The hybrid approach was used for the simulation of two different test cases: a complex 5 into 1 pipe junction of a high performance V10 engine and a Venturi tube plus a Helmholtz resonator of a single cylinder S.I. engine.

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.

This paper describes some recent advances in the field of I.C. engine modeling and simulation, concerning the development and application of a 1D thermo-fluid dynamic research code. An extensive experimental analysis has been concurrently carried out, to support the development and validation of the simulation code. A four-stroke, 10V-cylinder, 5.0 liters automotive S.I. engine has been modeled, in order to predict not only the wave motion in the system and its influence on the cylinder gas exchange process, but also the in-cylinder pressure to get a good prediction of pollutant emission concentration along the exhaust system. The gas composition in the exhaust pipe system is dictated by the cylinder discharge process, after the calculation of the combustion process via a thermodynamic multi-zone model, based on a fractal approach to predict the turbulent combustion.

Multi-dimensional simulation of hydrodynamics in full-scale wall-flow Diesel Particulate Filters by GpenFQAM®, an open-source C++ object-oriented CFD code, is presented. A new fast and efficient parallel numerical solver has been developed by authors to simulate flows through porous media and it has been tested for the simulation of diesel particulate filters; errors caused by discretization of filter monoliths have been corrected by the formulation of a correction factor, that has been included in the solver. A set of experimental data, available from literature, has been used for code validation.

A new fast and efficient parallel numerical solver for reacting and compressible flows through porous media has been developed in the OpenFOAM® (Open Field Operation and Manipulation) CFD Toolbox. With respect to the macroscopic model for porous media originally available in OpenFOAM®, a different mathematical approach has been followed: the new implemented solver makes use of the physical normal components resulting from the velocity expansion in the unit orthogonal vector basis to compute the Darcy pressure drop across the porous medium. Also, an additional sink term to account for the increased flow friction over the porous wall has been included into the momentum equation. In the new solver, the pressure correction equation is still able to achieve a faster convergency at very low permeability of the medium, also when it is associated with grid non-orthogonality.

The calculation of the main pollutant emissions discharged into the atmosphere by means of numerical codes requires the development of integrated models, including either an accurate thermodynamic in-cylinder analysis and the simulation of reacting unsteady flows in the duct system. This paper describes the main features of the numerical model GASDYN developed by the authors, which in the last years has been enhanced in order to achieve this kind of objectives. A multi-zone approach has been adopted to predict the combustion process in s.i. engines, whereas the so called super-extended Zeldovich mechanism has been introduced to perform a more detailed description of all the chemical reactions involved in the NOx production process. The simulation of the reacting flows in the exhaust manifold has been completed by the introduction of further enhancements to predict the chemical behavior of gases inside the catalytic converters.

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.

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.