Search Results

A convenient way of modelling turbochargers is based on data maps. These models are easy to put into place, require low CPU charge and are control-oriented. Data relative to compressor and turbine are read from tables: pressure ratio and efficiency are determined as functions of mass flow rate and rotary speed on two distinct data maps. Nevertheless, this type of model has drawbacks: Usually, only higher turbocharger speed data are mapped (> 90000 rpm) although the low rpm zone is the most useful zone for normalized driving cycles simulations. Moreover, maps are poorly discretized, leading to the use of specific extra-interpolation methods (many are identified in [5]). These methods are purely mathematical, which gives inaccurate results in extrapolation zones. Relation between pressure ratio and efficiency is then broken (i.e., if one implements a pumping model for the compressor, the pressure ratio will be affected, but not the efficiency).

The objective of this paper is to present and to validate a numerical model of a single-cylinder pneumatic-combustion hybrid engine. The model presented in this paper contains 0-D sub-models for non-spatially distributed components: Engine cylinder, Air tank, wall heat losses. 1-D sub-models for spatially distributed components are applied on the compressive gas flows in pipes (intake, exhaust and charging). Each pipe is discretized, using the Two-Steps Lax-Wendroff scheme (LW2) including Davis T.V.D. The boundaries conditions used at pipe ends are Method Of Characteristics (MOC) based. In the specific case of a valve, an original intermediate volume MOC based boundary condition is used. The numerical results provided by the engine model are compared with the experimental data obtained from a single cylinder prototype hybrid engine on a test bench operating in 4-stroke pneumatic pump and 4 stroke pneumatic motor modes.

This paper presents energy management strategies for a new hybrid pneumatic engine concept, which is specific by its configuration: It is not a vehicle but only an engine itself which is hybridized. This arrangement could provide as much as 30% of fuel saving depending on the driving cycle. Therefore different energy management strategies are proposed and compared in this paper. The first of them is called Causal Strategy and implements a rule-based control technique. A second strategy called Constant Penalty Coefficient is based on minimization of equivalent consumption, where the use of each energy source is formulated in a comparative unit. The balance between consumption of different energy source (chemical or pneumatic) is reached by introduction of an equivalence factor. The third strategy is called Variable Penalty Coefficient, where the equivalence factor is consider as variable within the amount of pneumatic energy stored in the air-tank.

Model-based tuning is a way followed by car manufacturers to reduce development costs. In this context, a new methodology has been developed in order to adapt a tur-bocharged diesel engine in the case of non-standard external conditions. Indeed, variable geometry turbine and fuel injection command laws are developed for standard conditions (20°C, altitude=0m). Turbocharger and fuel injection actuators pre-positioning maps should be adjusted regarding the inducted air mass density (influenced by the external temperature and pressure), in order to meet thermal, mechanical and pollutant emissions constraints. In order to reduce the use of climatic tests bench and extreme conditions tests in foreign countries, a model of a turbocharged diesel engine coupled to an optimization loop has been used to take into account the effect of non-standard external conditions on pre-positioning maps.

Nowadays, numerical modeling/simulation is an essential tool. It is used in all the design stages in order to improve quality and to reduce costs and time-to-market. As models address increasingly multiphysics and complex phenomena throughout the V-cycle process, modelers are confronted, sooner or later, with the need for model reduction. This paper describes the set-up of a virtual prototyping platform and highlights the interest of integrating energetic aspects, which are used to compute new model reduction criteria. A similar energy approach, “MORA” using “Activity”, is already used by Bond Graphists in order to obtain the “Proper Model” ([7] and [8]). The methodology presented here (PEMRA: “Power & Energy -based Model Reduction Algorithm”), with new power and energy criteria, makes it possible to obtain a simpler and more accurate reduced model than with MORA methodology, while improving the system's energy information.

The amount of fresh air induced into the cylinder is the main parameter to be taken into account when developing the engine control laws. However, the instantaneous amount of induced air cannot be directly measured. Additionally, as the engine air ducting becomes more and more complex (high and low pressure exhaust gas recirculation, variable valve timing, pneumatic hybridization…), models used to develop engine control laws must be as predictive as possible. It has therefore been decided to use 1d aerodynamics simulation to provide accuracy to the control laws development and validation process. Commercial engine codes have been tested but did not give satisfactory results in terms of calculation time and flexibility. Additionally, in the case where no experimental data are available to determine valve discharge coefficient, simulation results were in total disagreement with the engine bench measurements.

In the last decade, the drastic strengthening of engine emission regulations has conducted the automotive industry towards more and more sophisticated engine control strategies requiring more and more sensor inputs and actuator outputs. The testing and setting up of the ECUs implementing such strategies becomes more and more difficult, requiring numerous engine tests on test benches. ECUTEST is a hardware and software package from KADRA CONSULTANTS that offers the following features: Simulation of sensors including variable reluctance sensor, lambda sensor, knock sensor… Measurement of output signals (injection, ignition, EGR…) timing and amplitude. A predefined test pattern can be replayed on the ECU to perform end of line testing. A real-time model can be used for testing and setting up embedded closed loop strategies. The present paper will cover the implementation of a real-time SI engine model on ECUTEST.

Recrystallized Silicon carbide (R-SiC) honeycombs have been widely used over the last couple of years as filtration media for diesel particulates filtration in passenger cars applications. Although such filters are very reliable thanks to SiC good properties and smart designs, existing devices can still be improved. This paper describes several new features developed for R-SiC honeycomb filters in order to increase their durability and reduce their cost. Durability improvements can be obtained through the optimization of different filter properties such as thermo-mechanical resistance and thermal diffusivity. Specific tests have been performed in order to optimize new R-SiC filters.

Nowadays, (engine) downsizing using turbocharging appears as a major way for reducing fuel consumption. With this aim in view, the air actuators (throttle, Turbo WasteGate) control is needed for an efficient engine torque control especially to reduce pumping losses and to increase efficiency. This work proposes Nonlinear Model Predictive Control (NMPC) of the air actuators for turbocharged SI engines where the predictions are achieved by a neural model. The results obtained from a test bench of a Smart MCC engine show the real time applicability of the proposed method based on on-line linearization and the good control performances (good tracking, no overshoot) for various engine speeds.

Compressor models play a major role as they define the boost pressure in the intake manifold. These models have to be suitable for real-time applications such as control and diagnosis and for that, they need to be both accurate and computationally inexpensive. However, the models available in the literature usually fulfill only one of these two competing requirements. On the one hand, physics-based models are often too complex to be evaluated on line. On the other hand, data-based models generally suffer insufficient extrapolation features. To combine the merits of these two types of models, this work presents an extended approach to compressor modeling with respect to thermo- and aerodynamic losses. In particular, the model developed by Martin et al. [1] is augmented to explicitly incorporate friction, incidence and heat transfer losses. The resulting model surpasses the extrapolation properties of data-based models and facilitates the generation of extended lookup tables.

The paper presents a comparative study of various models used to estimate gas dynamics in internal combustion engine (ICE) ducts. 1D models provide a sufficient accuracy, but they are still not implementable on current ECUs. On the other hand, low order models can be real-time but their lack of accuracy and high calibration cost are still a challenging problem. This work aims at presenting a comparison of currently used gas dynamics models to predict transient phenomena in engine ducts. It emphasizes on 1D and low order models. To test under engine-like conditions, the intake path of a virtual engine implemented in GT-Power and a production two cylinder engine are used. Results show a contrast in the performance of the different models, which gives the possibility to evaluate the various approaches. Based on this assessment and depending on the application in hand, the models can be chosen properly to estimate the gas dynamics in internal combustion engine ducts.