Effect of Optimization Criteria on Direct-Injection Homegeneous Charge Compression Ignition Gasoline Engine Performance and Emissions Using Fully Automated Experiments and Microgenetic Algorithms

Homogeneous charge compression ignition (HCCI) is a new low-emission engine concept. Combustion under homogeneous, low equivalence ratio conditions results in modest temperature combustion products, containing very low concentrations of NO_x and PM as well as providing high thermal efficiency. However, this combustion mode can produce higher HC and CO emissions than those of conventional engines. Control of the start of combustion timing is difficult with pre-mixed charge HCCI. Accordingly, in the present study charge preparation and combustion phasing control is achieved with direct injection. An electronically controlled Caterpillar single-cylinder oil test engine (SCOTE), originally designed for heavy-duty diesel applications, was converted to a direct-injection gasoline engine. The engine features an electronically controlled low-pressure direct injection-gasoline (DI-G) injector with a 60 deg spray angle that is capable of multiple injections. The use of double injection was explored for emission control, and the engine was optimized using fully automated experiments and a microgenetic algorithm optimization code. The variables changed during the optimization include the intake air temperature, start of injection timing, and the split injection parameters (percent mass of fuel in each injection, dwell between the pulses) using three different objective (merit) functions. The engine performance and emissions were determined at 700 rev/min with a constant fuel flow rate at 10 MPa fuel injection pressure. The results show the choice of merit or objective function (optimization goal) determines the engine performance, and that significant emission reductions can be achieved with optimal injection strategies. Merit function formulations are presented that minimized PM, HC, and $NOx$ emissions, respectively.