Dissipation Elements at the Flame Surface in Methane Diffusion Flame (B. Hentschel and D. Denker)

Flame in Slotburner (S. Kruse)

Particle Charged Flow (E. Varea)

DNS of a scaled-up Diesel injector

Dissipation Element Analysis of Methane Diffusion Flame (D. Denker)

DNS of a scaled-up Diesel injector (M. Bode)

Quartz nozzle sampling in a methane counterflow flame (M. Baroncelli)

Oxyfuel coal combustion in a hot gas stream (D. Felsmann)

Turbulent/non-turbulent interface in high Reynolds number Jet (D. Denker and B. Hentschel)

Engine Test Bench

Control of partially homogenized diesel engine combustion by variable injection strategies

The aim of the project is the development of an efficient and clean low temperature diesel combustion process. At part load, this process has to work almost homogenously while at full load, the experience from conventional diesel combustion is utilized. 

The low temperature combustion strategy PCCI (premixed charge compression ignition) is of high importance for diesel engines due to its huge potential in reducing pollutant emissions. This method is characterized by moderate local temperatures and the avoidance of locally rich or lean zones and thereby inhibiting soot- and NOx-formation. Figure 1 shows the PCCI combustion process in comparison to other methods.

Figure 1: The PCCI combustion process

In order to ensure efficient operation with low emissions at part and full load operation, constructive and thermodynamic optimizing process is required. The potential of various piston bowl geometries and spray cone angles has been thoroughly investigated. On a continuous basis, detailed thermodynamic analysis is performed in order to reduce exhaust emissions without decreasing engine efficiency, which would increase the specific fuel consumption. 

The Institute for Combustion Technology (ITV) is steadily working on further development of this combustion method. Investigations on this promising future technique are performed on a modern 4 cylinder diesel engine equipped with a common rail direct injection system. The engine specifications are listed in Table 1 and Figure 2 shows the engine test bench of ITV. 

Table 1: Engine specifications (1.9 l diesel engine)

Figure 2
: The engine test bench of ITV

The ITV has high-performance measurement equipment for the engine test bench. A rapid, dynamic measurement of the EGR-rate is possible by the interpretation of lambda sensor signals in the air intake and the exhaust. Fast sensors allow the measurements of nitric oxides (NOx) as well as unburnt hydrocarbons (HC) within cycle-to-cycle resolution. The probe for these measurements is directly located at the exhaust manifold. This ensures undelayed and unaltered emission-measurements. The acquistion of results within milliseconds is essential for real-time emission control. Further emissions are be determined by stationary, slower measurement technique (soot, CO, CO2 etc.). 

In corporation with the Institute for Automatic Control (IRT) model-based control strategies are developed and applied at the ITV engine test bench. As a conclusion, the best results are reached with a multivariable cascade control, containing a fast secondary auxiliary control loop for the air path (10ms) and a slower main controller for the combustion (60ms). It was shown, that the combustion control achieves a high performance using a one-step optimal control approach. In addition, the potential of a predictive control strategy based on global optimization and PWA-models (piecewise affine) was pointed out by simulations. Currently the focus is on the implementation of a model-based predictive control (MPC) using a quadratic programming optimization. Figure 3 shows the simulated control results of an implemented MPC with 3 manipulated variables (SOI, FMI, EGR) and 3 controlled variables (CA50, IMEP, NOX). Here, the parameters for the control horizon are Hu = 5 and for the prediction horizon are Hp = 30. In summary, the set point values are reached adequately. The corresponding sequences of the control values are shown in Figure 3 (bottom part). The plotted MIN and MAX lines delimit the reachable ranges for the controlled variables and the manipulated variables at the current operating point. 

Figure 3: Control results of the MIMO system with 3 manipulated variables  (SOI, FMI, EGR) and 3                      controlled variables (CA50, IMEP, NOx) 

The experimental work at the engine test bench and the development of model-based control algorithms is complemented by computational work at the Institute for Combustion Technology (ITV). Simulations are performed with models of varying complexity. On one hand, CFD calculations are carried out using an in-house RANS flow solver that is coupled with the well-known combustion model RIF to account for the chemical source term. Detailed chemical mechanisms can be applied, which enable a realistic description of the chemical in-cylinder processes. On the other hand, reduced order combustion models are developed, based on the multi-zone concept. These efficient but accurate models allow a coupling with control algorithms to establish model predictive control of the Diesel engine process. 

Figure 4 depicts the comparison of experimental and simulated data for a variation of the external EGR-rate. On the left-hand side, the transient in-cylinder pressure curves are shown, which enable the validation of the CFD model. On the right-hand side, indicated mean effective pressure and combustion phasing are presented for both models as well as the experiment to indicate the predictive capability of the reduced order model. Both models are characterized by a very accurate description of the engine behavior. 

Figure 4: Comparison of experimental and computed in-cylinder pressure (above) as well as                              indicated mean effective pressure (IMEP) of experiment, CFD and multi-zone model (MZM)                  (below) for varying external exhaust gas recirculation (EGR)

Currently, the following tasks are being worked on:

  • Application of fuels with a lower cetane number in order to optimize the center of combustion
  • Development of a multiple injection strategy for PCCI combustion
  • Enhancement of the control strategy for the minimization of exhaust emissions and fuel consumption

Contact Person