Degree Type

Dissertation

Date of Award

2009

Degree Name

Doctor of Philosophy

Department

Mechanical Engineering

First Advisor

Song-charng Kong

Abstract

The objective of this research was to develop advanced diesel combustion strategies for emissions reduction in a multi-cylinder diesel engine. The engine was equipped with an electronically-controlled, common-rail fuel injection system, and an exhaust gas recirculation (EGR) system. This experimental setup allowed a wide range of operating conditions to be explored.

Effects of various injector parameters with various EGR levels on emissions were studied. Injector parameters included the injector flow number, nozzle hole geometry (straight, convergent), and nozzle arrangement (6-hole, 10-hole, 16-hole). The included spray angle was kept constant at 133 deg. Other engine parameters included the EGR rate (0-41%), injection pressure (150-225 MPa), start of injection (SOI) (-20 to 5 ATDC), start of pilot injection (-40 to -15 ATDC), and pilot fuel percentage (0-25%).

For single injection operations, a simultaneous reduction of NOx and particulate matter (PM) was achieved by using high EGR (30%) with late injection timing (0 to 5 ATDC) at high injection pressures (150 MPa). For double injection operations, NOx and PM emissions were reduced using 30% EGR, 15% pilot injection at an early pilot timing (-30 ATDC) and late main injection (5 ATDC).

Injectors with low flow numbers were able to produce low emissions at high EGR levels (>35%) and high injection pressures (>150 MPa). The combustion was stable at these high EGR levels as the SOI was held at 0 ATDC. On the other hand, injectors with high flow numbers were not able to produce stable combustion at such high EGR levels with late SOI.

Small nozzle holes in the 10-hole injector helped reduce NOx and PM emissions significantly. However, a 16-hole injector with a similar nozzle hole diameter produced very high PM emissions due to poor air utilization.

To improve the speed of optimization for lower emissions, particle swarm optimization (PSO), a stochastic, population-based evolutionary optimization algorithm, was applied to both engine experiments and numerical simulation. The algorithm was tested using test functions that were used in the field of optimization to ensure reaching a global optimum. A merit function was defined to help reduce multiple emissions simultaneously. The PSO was found to be very effective in finding the optimal operating conditions for low emissions. The optimization usually took 40-70 experimental runs to find the optimum. High EGR levels, late main injection, and small pilot amount were suggested by the PSO. Multiple emissions were reduced simultaneously without a compromise in the brake specific fuel consumption. In some cases, the NOx and PM emissions were reduced to as low as 0.41 and 0.0092 g/kW-h, respectively. The operating conditions at this point were 34% EGR, 5 ATDC main SOI, -24 ATDC pilot SOI, and 5% pilot fuel.

The PSO was also integrated with an engine simulation code and applied to engine optimization numerically. The results showed that optimization of engine combustion using PSO with numerical simulation was an effective means in the development of future emission reduction strategies.

DOI

https://doi.org/10.31274/etd-180810-2128

Copyright Owner

Prashanth Kumar Karra

Language

en

Date Available

2012-04-30

File Format

application/pdf

File Size

156 pages

Share

COinS