Abstract
The Homogeneous Charge Compression Ignition (HCCI) engine offers an interesting compromise between fuel consumption, pollutant emissions and efficiency. The challenge is to control cycle-to-cycle variations in order to maintain the stability of combustion. Recently, the effect of ozone addition has been experimentally investigated and it has been shown that the use of a small mass flow of ozone, a strong oxidizer species, can improve the combustion and forward the phasing of combustion. A solution to compensate cycle-to-cycle variations is then to combine dilution using EGR valve (slow response) and adding a proportion of ozone in the mixture (fast response).
The objective of this work is to develop a real time control of cycle-to-cycle HCCI combustion by ozone addition. For this, a control-oriented model is developed in order to give a good estimation and prediction of the combustion phasing parameters (in-cylinder maximum pressure, corresponding crank angle and the crank angle degree when 50% of the fuel mass is consumed) using position of the EGR valve, equivalence ratio and ozone mass flow that acts as a chemical actuator . A developed 0D physical model is used to create 3D-maps of EGR and ozone rate values that can lead to the optimal combustion parameters. Finally, the real time control laws in addition to the maps are employed to control HCCI combustion by ozone addition.
The 0D physical model is validated against in-cylinder experimental data with different equivalence ratios (0.3 to 1), with various EGR rates and ozone concentrations, at constant engine speed 1500 rpm. The results show three different cases during HCCI combustion. An excess EGR rate tends to misfire but the ozone addition improves the combustion. For an EGR rate corresponding to complete combustion, the ozone seeded moves forward the combustion phasing. An EGR rate lower than the required amount for an operating point leads to knock, in this case, the ozone seeded is useless; and only by decreasing equivalence ratio combined with ozone addition, the optimal operating point can be reached. The simulated results show a good agreement with experimental data for all combustion parameters. A control-oriented model, based on this physical model, is developed and shows its capacity to predict the stability of the combustion during transient operation.
The control law and the control oriented model must be integrated into a standard Engine Control Unit (ECU). The limitation of this approach is linked to sample time. The HCCI engine is a non-linear dynamics system, creating performance problems such as cylinder to cylinder stability and robustness. Optimal and robust control laws (emissions/efficiency) will be designed to meet cylinder-to-cylinder dynamics and to stabilize the HCCI combustion. Finally, the control laws will be validated on a multi-cylinder engine with a high dynamic bench in real-time.
A new 0D combustion model dedicated to HCCI combustion is developed in this study; based on a probability density function (PDF) approach to describe temperature fluctuations inside the combustion chamber. A chemical actuator (ozone concentration) is integrated in a control-oriented model in order to control the stability combustion.
A comparison of in-cylinder pressure curves between simulated and experimental results is done through the developed 0D HCCI combustion model. Simulated results show a good agreement with experimental data for in-cylinder pressure and combustion parameters. The results of the control-oriented model make it suitable for real-time HCCI engine control.
KEYWORDS - HCCI engine, 0D combustion model, control-oriented model, ozone addition