Compact combustion concept for lowest PN and High Efficiency

Integration, testing and validation on Dynometer set up of a Renault demonstrator vehicle and Integration into a demonstrator vehicle. The objectives for Renault are:

  • To develop and validate the selected engine system technologies (to TRL 7) to achieve Fuel savings: 5% saving comp. to EU6 state of the art; Emissions: Half of Euro 6; RDE with conformity factor 1.5; special focus on reduction of emitted particles (number) including sub 23 nm particles (>80% reduction based on Eiro6 limit); Demonstration: Preparation of a demonstrator vehicle which contains the selected system technologies and which will be provided for an independent testing (carried out in WP7)
  • To develop new techniques to reach the targets given above: Combustion concept with reduced heat losses from the combustion chamber by partly isolated pistons and reduction of heat transfer from hot gas to the walls by optimizing the in-cylinder gas flow; Implementation of improved control functions for fuel injection to utilize demanding fuel injection characteristics such as multiple injections with low fuel quantities and dwell times (“digital rate-shaping”); Integration of high capacity air and EGR systems to support a high charged gas flow and EGR rates with minimum pumping losses

Starting point of the efficiency improvement is the reduction of heat losses from the combustion chamber. This requires design measures such as e.g. a partial thermal insulation of the piston surface and also a new combustion concept with optimized flow characteristics of the charged gas and improved fuel injection strategies. The aim is to accomplish a stable and fast heat release without any delayed reactions and optimal soot oxidation. In order to achieve a considerable improvement of the efficiency various side effects of heat insulation will be exploited and optimally combined such as the utilization of the elevated exhaust energy, friction reduction due to higher component temperatures and reduced drive power of auxiliaries (coolant pump, fan).

Update August 2018: Combustion system assessment by CFD simulation

In order to prepare the development work of the engine on the test bench and to determine the best combustion system dedicated to the engine performance targets, we have realized a CFD simulation study to compare three possible combustion bowl geometries. These geometries are the following:


From the analysis on the different speed-load points of the study, the behaviour of the “chamfer” geometry shows a better performance:

  • At rated power, efficiency is higher, soot emissions are lower
  • At low engine speed, the maximum torque is higher
  • At part load, efficiency and NOx / soot trade-off are better than for other variants.

Based on these results, the demonstrator engine is equipped with the “chamfer” design of the combustion chamber. The benefits in efficiency and NOx soot trade-off will contribute to the project target achievement.

Update July 2018: Modelling of engine emissions in transient operation

The objective of this task is to model the engine emissions in transient operation. With this purpose, a platform for engine simulation available at CMT Motores Térmicos (Virtual Engine Model -VEMOD-) has been completely refined and adapted to perform an assessment of different strategies at cold and low pressure ambient conditions as the main objective of this task.


The integrated Virtual Engine Model (VEMOD) has been rebuilt as a standalone tool to simulate new standard testing cycles. The VEMOD is based on a wave-action model that carries out the thermo-and fluid dynamics calculation of the gas in each part of the engine. In the model, the engine is represented by means of 1D ducts, while the volumes, such as cylinders and reservoirs, are considered as 0D elements. VEMOD includes different sub-models that have been refined and/or adapted to retain the main physics of the most relevant phenomena whilst providing fast response for accurate simulation of engine fast transients (boosting and EGR behaviour) and low transient (engine warm-up) cycles. The improvements carried out deal with the following:

  • 1D/0D thermo-and-fluid dynamic: it is the core or the simulation and most of the calculation time is due to this subsystem, thus the numerical method for solving the physical equations has been modified from finite differences to finite volumes. This numerical technique is extremely conservative so big mesh size can be used to reduce computation time from the original 50 times slower than real time to the current 10 times slower than real time. Also the model has been improved through the identification and quantification of water condensation at different engine elements, key issue during cold starting. Finally, the blow-by model, used to evaluate the gas leakage and to set the boundary condition for the friction calculation in the piston, has been reviewed to consider the effect of ambient conditions.
  • Virtual injector: the main input for the combustion model is the injection rate, thus a new simple injection model to reproduce the actual injection rate has been included in VEMOD. The model simulates the instantaneous fuel injection based on straight part that takes into account the initial and final transient periods and the steady state part in long injections.
  • Combustion and emissions formation models: combustion process is calculated based on the Apparent Combustion Time (ACT) 1D model [1], responsible for the prediction of the rate of heat release and NOx formation. ACT is a mixing controlled combustion model that has been improved with a CFD-based correction of the transient processes at SOI and EOI. Thus, during quasi-steady conditions, a physical approach based on turbulent gas jet theory is used to calculate the air entrainment in each fuel parcel injected. During the transient process at the start/end of injection a parameterized correction based on CFD calculations is applied. The ignition delay model, which is based in the Shell model, tracks the Livengood & Wu [2] integral accounting for the instantaneous thermodynamic conditions and the local composition. It has been upgraded to make it more suitable for low T conditions. Finally, the premixed combustion is reproduced by means of an empirical model based on the propagation velocity of a premixed flame. Regarding the emissions, in the case of NOx a physical (chemical) approach is applied, considering all the relevant mechanisms for NOx formation (thermal or Zeldovich, prompt and via N2O) and reduction (reburning mechanism, relevant at high EGR rates and F/A ratios). In order to reduce the computation time, the model is implemented in a computational efficient way by means of tabulated chemistry. For the rest of pollutants (soot, CO and UHC emissions), which fundamentals are unclear and/or affected by many local phenomena, an Artificial Neural Network (ANN) approach has been used to determine them.
  • In order to predict tailpipe pollutant emissions to the ambient, different aftertreatment systems models have been upgraded (DOC, DPF) or completely developed (SCR y LNT) to reproduce the behavior of the devices in the exhaust system. Thus, the DPF model have been improved to consider PSD variation due to filtration, DOC model considers UHC accumulation and SCR and LNT have been included in VEMOD. Main physical/chemical processes taking place in each EATS element are approached to be solved based on a lumped model approach.
  • Heat transfer (HT) model in VEMOD is based on lumped conductance models (engine block, ducts and turbo) to take into account the heat transferred between the different fluids of the engine (gas and liquids –coolant and oil-). Thus, the lumped conductance model allows linking in-cylinder, port and turbocompressor processes with hydraulic circuits through the heat rejection calculation: 0D in-cylinder model, 1D ports model and 0D turbo and compressor models provides boundary conditions of gas temperature and heat transfer coefficient to calculate heat flux, while the nodal model provides detailed wall temperature and heat transfer repartition to coolant and oil circuit. The model has been improved to include the oil-coolant heat exchange through the block, the thermal inertia of liquids (coolant and oil) and the metal heating, all of them key issues during transient operation starting from cold conditions. Finally, the model has been adapted to evaluate different thermal management strategies with split cooling system (independent coolant circuit in the cylinder-head and block).


Being the heat transfer in the chamber the main source of heat rejection, it is a critical issue during the operation in cold conditions such as that in the WLTC after the engine start at -7ºC. With the aim of assessing the validity of the VEMOD HT model, a critical review of convective HT models has been carried out. The main conclusion found is that the influence of room temperature on the heat transfer fitting coefficient is smaller than engine speed influence, and hence there is no evidence that the Woschni-based HT model of VEMOD is not valid for both hot and cold environment conditions.


  • The VEMOD includes coolant and lubricant circuits linked, on the one hand, with the engine block and the turbocharger through heat transfer lumped models; and on the other hand with the engine heat exchangers. The hydraulic models have been upgraded to include all the required heat exchangers (gas-liquid, liquid-liquid…) and the effect of the vehicle velocity on the radiator. Also hydraulic circuit model has been modified in order to consider split cooling in combination with the lumped conductance model.
  • An existent dedicated friction and auxiliaries model [7] has been included in VEMOD to obtain the brake power starting from indicated power. Friction losses (piston pack, bearings and valve train) are calculated taking into account the instantaneously lubrication regime, kinematics and dynamics of the mechanisms. Specific upgrade to consider rolling cam follower in the valve train and oil viscosity at low temperatures (up to -35ºC) have been carried out. Auxiliaries system has been considered in a simple way, taking into account the power to drive fuel, coolant and oil pumps.
  • A control system emulating the ECU along with a vehicle and driver models allow completing the engine simulation tool. On the one hand the control system has been upgraded to include all the necessary sensors (torque, mass flow, p and T in pipes…) and actuators (EGR, swirl valves, turbine position, engine speed, injection settings…) in order to simulate both stationary and transient tests (WLTP…) by considering different engine settings and operations modes. Thus, the VEMOD is now able to reproduce experimental conditions by imposing measured settings and engine speed (no vehicle is considered) or a fully predictive evolution where the vehicle is considered and the setting are taken from the calibration tables in the virtual ECU in order to follow a track.


After the VEMOD upgrading, sub-model geometry was adapted to the characteristics of the specific engine and the properties of different elements (turbocharger maps, pump curves, thermostat settings, heat exchangers efficiencies, virtual ECU calibration…) were set according to the information provided by Renault or specific tests performed at CMT. The calibration of the model was performed in steady state operating condition using experimental results at different operating points and ambient temperature conditions (from -7ºC to 25ºC). When possible, the sub-models were calibrated independently of the rest:


  • The injector model was calibrated with a complete experimental matrix in an injector test rig.
  • Aftertreatment models were calibrated (and validated) in stand-alone executions in against experimental data.
  • The friction model calibration, aimed at obtaining an accurate prediction of the brake efficiency, was performed with stationary operating points in the complete engine map both at cold and hot ambient conditions.
  • Heat tranfer model calibration was performed by means of a methodology developed at CMT in which a combination of experimental and simulated in-cylinder cycles in motoring and combustion tests allows obtaining the heat transfer model constants.

In the case of the combustion model (and hence emissions sub-models that are directly dependent on it) a dedicated calibration was performed after the heat transfer model calibration, which affects the experimental heat release (main input for the combustion model).

After the calibration of the different models, the complete VEMOD has been validated with experimental tests using steady and WLTC tests. The result showed that key variables were well predicted during WLTC such as engine torque (e=2 %), turbine outlet temperature (e=6 %), CO2 emissions (total cumulated e< 1), and NOx with a good performance during the WLTC cycle except at the last part (high load and speed) having a mean error of 14 %.


Dissemination of the project results:

  1. Martin, J., Arnau, F., Piqueras, P., and Auñon, A., “Development of an Integrated Virtual Engine Model to Simulate New Standard Testing Cycles,” SAE Technical Paper 2018-01-1413, 2018, doi:10.4271/2018-01-1413.
  2. Broatch, A., Olmeda, P., Martin, J., and Salvador-Iborra, J., “Development and Validation of a Submodel for Thermal Exchanges in the Hydraulic Circuits of a Global Engine Model,” SAE Technical Paper 2018-01-0160, 2018, doi:10.4271/2018-01-0160.
  3. Payri, F., Arnau, F.J., Piqueras, P., and Ruiz, M.J., “Lumped Approach for Flow-Through and Wall-Flow Monolithic Reactors Modelling for Real-Time Automotive Applications,” SAE Technical Paper 2018-01-0954, 2018, doi:10.4271/2018-01-0954.


Development of an Integrated Virtual Engine Model to Simulate New Standard Testing Cycles by Jaime Martín, Francisco Arnau, Pedro Piqueras and Angel Auñón. Universitat Politecnica de Valencia

The combination of more strict regulation for pollutant and CO2 emissions and the new testing cycles, covering a wider range of transient conditions, makes very interesting the development of predictive tools for engine design and pre-calibration. This paper describes a new integrated Virtual Engine Model (VEMOD) that has been developed as a standalone tool to simulate new standard testing cycles. The VEMOD is based on a wave-action model that carries out the thermo-and fluid dynamics calculation of the gas in each part of the engine. In the model, the engine is represented by means of 1D ducts, while the volumes, such as cylinders and reservoirs, are considered as 0D elements. Different sub-models are included in the VEMOD to take into account all the relevant phenomena. Thus, the combustion process is calculated by the Apparent Combustion Time (ACT) 1D model, responsible for the prediction of the rate of heat release and NOx formation. Experimental correlations are used to determine the rest of pollutants. In order to predict tailpipe pollutant emissions to the ambient, different sub-models have been developed to reproduce the behavior of the aftertreatment devices (DOC and DPF) placed in the exhaust system. Dedicated friction and auxiliaries sub-models allow obtaining the brake power. The turbocharger consists of 0D compressor and turbine sub-models capable of extrapolating the available maps of both devices. The VEMOD includes coolant and lubricant circuits linked, on the one hand, with the engine block and the turbocharger through heat transfer lumped models; and on the other hand with the engine heat exchangers. A control system emulating the ECU along with vehicle and driver submodels allow completing the engine simulation. The Virtual Engine Model has been validated with experimental tests in a 1.6 L Diesel engine using steady and transient tests in both hot and cold conditions. Engine torque was predicted with a mean error of 3 Nm and an error below 14 Nm for 90 % of the cycle duration. CO2 presented a mean error of 0.04 g/s, while during 80 % of the cycle, error was below 0.44 g/s. Read more

Development and Validation of a Submodel for Thermal Exchanges in the Hydraulic Circuits of a Global Engine Model by Broatch, A.; Olmeda, P.; Martín, J.; Salvador-Iborra, J.

To face the current challenges of the automotive industry, there is a need for computational models capable to simulate the engine behavior under low-temperature and low-pressure conditions. Internal combustion engines are complex and have interconnected systems where many processes take place and influence each other. Thus, a global approach to engine simulation is suitable to study the entire engine performance. The circuits that distribute the hydraulic fluids –liquid fuels, coolants and lubricants- are critical subsystems of the engine. This work presents a 0D model which was developed and set up to make possible the simulation of hydraulic circuits in a global engine model. The model is capable of simulating flow and pressure distributions as well as heat transfer processes in a circuit. After its development, the thermo-hydraulic model was implemented in a physical based engine model called Virtual Engine Model (VEMOD), which takes into account all the relevant relations among subsystems. In the present paper, the thermo-hydraulic model is described and then it is used to simulate oil and coolant circuits of a diesel engine. The objective of the work is to validate the model under steady-state and transient operation, with focus on the thermal evolution of oil and coolant. For validation under steady-state conditions, 22 operating points were measured and simulated, some of them in cold environment. In general, good agreement was obtained between simulation and experiments. Next, the WLTP driving cycle was simulated starting from warmed-up conditions and from ambient temperature. Results were compared with the experiment, showing that modeled trends were close to those experimentally measured. Thermal evolutions of oil and coolant were predicted with mean errors between 0.7ºC and 2.1ºC. In particular, the warm-up phase was satisfactorily modeled. Read more

Lumped Approach for Flow-Through and Wall-Flow Monolithic Reactors Modelling for Real-Time Automotive Applications by Payri, F.; Arnau Martínez, FJ.; Piqueras, P.; Ruiz Lucas, MJ.

The increasingly restrictive legislation on pollutant emissions is involving new homologation procedures driven to be representative of real driving emissions. This context demands an update of the modelling tools leading to an accurate assessment of the engine and aftertreatment systems performance at the same time as these complex systems are understood as a single element. In addition, virtual engine models must retain the accuracy while reducing the computational effort to get closer to real-time computation. It makes them useful for pre-design and calibration but also potentially applicable to on-board diagnostics purposes. This paper responds to these requirements presenting a lumped modelling approach for the simulation of aftertreament systems. The basic principles of operation of flow-through and wall-flow monoliths are covered leading the focus to the modelling of gaseous emissions conversion efficiency and particulate matter abatement, i.e. filtration and regeneration processes. The model concept is completed with the solution of pressure drop and heat transfer processes. The lumped approach hypotheses and the solution of the governing equations for every submodel are detailed. While inertial pressure drop contributions are computed from the characteristic pressure drop coefficient, the porous medium effects in wall-flow monoliths are considered separately. Heat transfer sub-model applies a nodal approach to account for heat exchange and thermal inertia of the monolith substrate and the external canning. In wall-flow monoliths, the filtration and porous media properties are computed as a function of soot load applying a spherical packed bed approach. The soot oxidation mechanism including adsorption reactant phase is presented. Concerning gaseous emissions, the general scheme to solve the chemical species transport in the bulk gas and washcoat regions is also described. In particular, it is finally applied to the modelling of CO and HC abatement in a DOC and DPF brick. The model calibration steps against a set of steady-state in-engine experiments allowing separate certain phenomena are discussed. As a final step, the model performance is assessed against a transient test during which all modelled processes are taking place simultaneously under highly dynamic driving conditions. This test is simulated imposing different integration time-steps to demonstrate the model´s potential for real-time applications. Read more

Update March 2018

Starting from the Euro6 version of the 2,0 litre Diesel engine currently in development at Renault, we have integrated in the Dieper demonstrator engine several new techno-bricks in order to improve both the engine out emissions level and the fuel efficiency to reach the targets of the project. In parallel a precise optimization of the fuel and air systems was carried in accordance with the full load performances requirement of the project: 120 kW and 360Nm to take maximum benefit in the part load range of the engine map.

The fuel injection and the combustion systems have been specified in collaboration between AVL, Continental and Renault starting from state of the art low swirl intake ports, nine holes low flow injector, and dedicated bowl geometry. The turbocharger has been precisely matched by MHI according to the maximum power and the high EGR rates requirements obtained using both cooled high-pressure and low-pressure circuits.

Moving to the new techno-bricks integrated on the base engine, at first, we have implemented with MHI a cooled compressor housing to allow high EGR rates without exceeding the maximum temperature limit of this component.

In addition, thermal coatings of pistons, cylinder head, intake valves and exhaust manifold will improve both engine fuel efficiency with reduced heat losses and higher exhaust temperature for the after-treatment system performance.

Finally, a twin stage intake air cooling with continuous flow repartition management designed in cooperation with Mahle will give the possibility to adjust the intake temperature of the engine depending on the operating point. For cold start and low ambient temperature, an additional electrical heater is also integrated in the intake system.

In conclusion, we have produced a very refined demonstrator engine, with advanced thermal management features, which will contribute to improve the thermal efficiency, to reduce NOx production and to optimize the after-treatment system function.

Two of these engines will be assembled, one for the test bench activities, and the second one for the demo-car. Read more


Want to know more? Please contact Philippe Mallet (Renault)

Share on FacebookTweet about this on TwitterShare on Google+Share on LinkedInEmail this to someonePrint this page