Mallikarjuna J M

This article aims to identify the best combination of intake port angle (IPA) and cylinder head pentroof angle (PA) of a gasoline direct injection (GDI) engine to achieve a simultaneous reduction in the fuel consumption and the exhaust emissions using computational fluid dynamics (CFD) and optimization techniques. The present study is carried out on a single-cylinder, four-stroke GDI engine. The design space is bound by the range of the IPA (35°, 80°) and the PA (5°, 20°). The initial data set consists of 80 design points, which are generated using the uniform Latin hypercube (ULH) algorithm. CFD simulations were carried out at all the points in the initial data set using CONVERGE at engine speed of 2,000 rev/min and the overall equivalence ratio of 0.7 ± 0.05. A prediction model based on the support vector machine algorithm is generated between the design inputs and the output parameters viz., indicated specific fuel consumption (ISFC), hydrocarbon (HC), nitric oxides (NOx), and soot. After sufficient validation of the prediction model, it is used for the optimization study. The optimization is carried out using the MOGA-II algorithm. The optimization study predicted that the IPA of 58° and the PA of 13.4° results best, in simultaneous reduction of the fuel consumption and the emissions. The results of the optimization study are further validated using the CFD analysis, which is carried out at the optimum design point. From the results, it is concluded that the optimization-driven design techniques could be effectively used to improve the engine performance and reduce the emissions simultaneously.

The mixture formation in gasoline direct injection (GDI) engines operating at stratified condition plays an important role in deciding the combustion, performance and emission characteristics of the engine. In a wall-guided GDI engine, piston profile is such that the injected fuel is directed towards the spark plug to form a combustible mixture at the time of ignition. In these engines, fuel injection pressure and timing play an important role in creating a combustible mixture near the spark plug. Therefore, in this study, an attempt has been made to understand the effect of fuel injection pressure with single and split injection strategy on the mixture formation in a four-stroke, wall-guided GDI engine operating under stratified conditions by using computational fluid dynamics (CFD) analysis. Four fuel injection pressures viz., 90, 120, 150 and 180 bar are considered for the analysis. All the CFD simulations are carried out at the engine speed of 2000 rev/min., compression ratio of 11.5, with the overall equivalence ratio of about 0.65. The fuel injection and spark timings are maintained at 605 and 705 CADs respectively. In this study, the effect of fuel injection pressure on mixture stratification is carried out by a new parameter called "Stratification Index". It is found that, at the time of the spark, with single fuel injection, with the fuel injection pressure of 180 bar, proper mixture stratification is produced. But, with split injection mode, at all the fuel injection pressures considered, a nearly homogeneous mixture is produced. Also in the single fuel injection cases, with the fuel injection pressures of 120, 150 and 180 bar, the peak in-cylinder pressures are higher by about 4.6, 14.9 and 19.6%; and 1.5, 3.7 and 4.3% respectively, compared to that of 90 bar fuel injection pressure.

Mixture distribution in the combustion chamber of gasoline direct injection (GDI) engines significantly affects combustion, performance and emission characteristics. The mixture distribution in the engine cylinder, in turn, depends on many parameters viz., fuel injector hole diameter and orientation, fuel injection pressure, the start of fuel injection, in-cylinder fluid dynamics etc. In these engines, the mixture distribution is broadly classified as homogeneous and stratified. However, with currently available engine parameters, it is difficult to objectively classify the type of mixture distribution. In this study, an attempt is made to objectively classify the mixture distribution in GDI engines using a parameter called the "stratification index". The analysis is carried out on a four-stroke wall-guided GDI engine using computational fluid dynamics (CFD). All CFD sub-models used, in this study, are validated with the available experimental and CFD results from the literature before carrying out the analysis. Three types of mixture distributions viz., ideally homogeneous, ideally stratified and mal-distributed mixtures are defined and their effect on combustion, performance and emission characteristics of the engine are analyzed. Further, the effect of fuel injector orientation on the mixture distribution in the combustion chamber is analyzed for three different orientations of the fuel injector viz., -15, 0 and 15° with the vertical. From the results, it is found that the early fuel injection doesn't produce an ideally homogeneous mixture. Also, among the cases of the stratified mixtures, it is found that, the fuel injector orientation of 15° results in a mixture that is closer to the ideally stratified one. This is characterized by the value of stratification index that is close to 1.

Preparation of air-fuel mixture and its stratification, plays the key role to determine the combustion and emission characteristics in a gasoline direct injection (GDI) engine working in stratified conditions. The mixture stratification is mainly influenced by the in-cylinder flow structure, which mainly relies upon engine geometry i.e. cylinder head, intake port configuration, piston profile etc. Hence in the present analysis, authors have attempted to comprehend the effect of cylinder head geometry on the mixture stratification, combustion and emission characteristics of a GDI engine. The computational fluid dynamics (CFD) analysis is carried out on a single-cylinder, naturally-aspirated four-stroke GDI engine having a pentroof shaped cylinder head. The analysis is carried out at four pentroof angles (PA) viz., 80 (base case), 140, 200 and 250 with the axis of the cylinder. The entire CFD simulations are performed at the engine speed of 2000 rev/min., and the overall equivalence ratio (ER) of 0.75. Finally, it is observed that the PA of 140 produced a rise of about 10.5% in indicated thermal efficiency (ITE) and 3% rise in peak heat release rate (HRR) with a compromise of 10.7% higher NOx emissions than that of the base case.