Vincenzo De Bellis

A numerical procedure for the calibration of a turbocharged spark-ignition variable valve actuation engine at part load

Referring to spark-ignition engines, the downsizing, coupled to turbocharging and variable valve actuation systems are very common solutions to reduce the brake-specific fuel consumption at low-medium brake mean effective pressure. However, the adoption of such solutions increases the complexity of engine control and management because of the additional degrees of freedom, and hence results in a longer calibration time and higher experimental efforts. In this work, a twin-cylinder turbocharged variable valve actuation spark-ignition engine is numerically investigated by a one-dimensional model (GT-Power™). The considered engine is equipped with a fully flexible variable valve actuation system, realizing both a common full-lift strategy and a more advanced early intake valve closure strategy. Refined sub-models are used to describe turbulence and combustion processes. In the first stage, one-dimensional engine model is validated against the experimental data at full and part load. The validated model is then integrated in a multipurpose commercial optimizer (modeFRONTIER™) with the aim to identify the engine calibration that minimizes brake-specific fuel consumption at part load. In particular, the decision parameters of the optimization process are the early intake valve closure angle, the throttle valve opening, the turbocharger setting and the spark timing. Proper constraints are posed for intake pressure in order to limit the gas-dynamic noise radiated at the intake mouth. The adopted optimization approach shows the capability to reproduce with good accuracy the experimentally identified calibration. The latter corresponds to the numerically derived Pareto frontier in brake mean effective pressure–brake specific fuel consumption plane. The optimization also underlines the advantages of an engine calibration based on a combination of early intake valve closure strategy and intake throttling rather than a purely throttle-based calibration. The developed automatic procedure allows for a ‘virtual’ calibration of the considered engine on completely theoretical basis and proves to be very helpful in reducing the experimental costs and the engine time-to-market.

Numerical analysis of the transient operation of a turbocharged diesel engine including the compressor surge

The paper deals with the simulation of a multi-cylinder turbocharged diesel engine for automotive applications, employing a one-dimensional approach with the aim of refining the turbocharger modelling during transient manoeuvres. The proposed methodology is able to handle stable compressor behaviour and also compressor surge. In addition, a waste-gate model is introduced to account for the instantaneous variation in the valve section as a result of the control signal, which is provided by the engine control unit, and the engine state. Preliminarily, the engine model is tuned against experimental data in terms of both the global performance parameters and the in-cylinder pressure cycles. The compressor performance is described through an ‘extended’ map obtained using a one-dimensional turbocharger model; in this way, a refined surge analysis can be performed, accounting for both direct flow compressor operations and reverse flow compressor operations. The one-dimensional model is applied to analyse different transient manoeuvres. First, the vehicles maximum speed is predicted and compared with the manufacturers data, during an acceleration manoeuvre. Then, a sudden part-to-full-load step is described with the aim of analysing in detail the turbo-lag. Finally, a full-to-part-transient manoeuvre is also analysed to verify the capability of the model to represent the compressor surge phenomenon. The numerical results provided in this work qualitatively reproduce the experimental observations available in the literature for transient operation of engines. Thus, the developed computational tool can be successfully used to support the design process and the transient analysis of turbocharged internal-combustion engines.

One-Dimensional Simulations and Experimental Analysis of a Wastegated Turbine for Automotive Engines under Unsteady Flow Conditions

In this paper, the unsteady-state behaviour of a turbocharger wastegated turbine (IHI-RHF3) is investigated using both an experimental approach and a numerical approach. First, an experimental campaign is performed in a specialized test rig operating at the University of Genoa, for different openings of the wastegate valve and under steady flow and unsteady flow operations. An appropriate configuration of the turbine outlet circuit fitted with a separating wall is used to carry out instantaneous measurements downstream of the turbine wheel and the wastegate valve. The above data constitute the basis for the tuning and validation of a one-dimensional turbine model recently developed at the University of Naples. Preliminary model tuning is carried out on the basis of the characteristic map measured for a completely closed wastegate valve under steady flow operations. A refined one-dimensional schematization of the experimental apparatus is implemented within the commercial GT-Power® software, including the turbine, the wastegate circuit and the upstream and downstream measuring stations. In particular, the classical map-based approach is suitably corrected with a sequence of pipes that schematizes each component of the turbine (the inlet and outlet ducts, the volute and the wheel) to account for the wave propagation and storage phenomena inside the machine. A detailed one-dimensional schematization of the wastegate circuit is also implemented and independently tuned. The turbine model capability under unsteady flow conditions is tested for different wastegate openings and pulse frequencies, by applying time-dependent boundary conditions. In particular, the pressures and temperatures measured upstream and downstream are imposed at the model ends, and the instantaneous mass flow rate and the actual power are numerically evaluated. The results are compared with the experimental data, demonstrating good accuracy and showing some improvements with respect to the standard turbine modelling in the case of the mass flow rate prediction. On the contrary, the computed actual power shows some inaccuracies, especially at higher pulse frequencies.