| The human being's society has been made progress with the development of world's automobile industry powered by internal combustion engines, but automobiles also brought seriously environment problems. The automobile emission control regulations have become stricter than before to limit the emissions of pollutants. Diesel engines have been widely used because of its high power density, low fuel consumption, and high work reliability, the mostly attention have been paid on its particulate emissions. The performances of diesel engines under transient conditions are much different from that under steady operating conditions. The combustion process, the emission behavior and the performance improvement of diesel engines under transient operating conditions have been a very important research field.An important way to control emissions is to study the creation behavior of pollutants through combustion process of diesel engines. These techniques include in-cylinder sampling, optic measurements, CFD (Computational Fluid Dynamics), and so on. The total in-cylinder sampling technique has several advantages than other techniques, such as lower cost, more reliability of the measuring results, more prior to be used to measure in-cylinder mass-averaged particulate histories. The sampling mechanism is the key of total in-cylinder sampling technique. The conventional sampling mechanism uses cutter and diaphragm to obtain samples which brought some flaws to this technique. The conventional total in-cylinder sampling system is more complicated and the experimental period has been prolonged by the replacing of diaphragm due to its broken after each sampling action. So improve the sampling mechanism is very meaningful to meet with the high requirements of convenience and efficiency of modern control and measure system.This research discussed on emission formation histories, such as particulate, nitrogen oxides and total hydrocarbon in a single direct injection (D.I.) diesel engine by using a developed total in-cylinder sampling system. The effects on emission formation histories by using GTL fuel and EGR technique were analyzed. The differences of emission formation histories under transient operating conditions and steady operating conditions were also discussed.The main study objects and achievements are as follows.1. Establishment of control and measure system for transient operating conditionsThe test system was composed of three components: parameter measure system, operating conditions control system and sampling control system.1) A high speed, real time data recording and monitoring system was developed. The data of in-cylinder pressure, speed of engine, et al, could be recording and displaying at the same time by software programming of an A/D converter board.2) A repeatable transient condition control system was realized by using SCP (Single Chip Processors) of MCS-51 series as the core control elements and a step motor as the actuator. Two kinds of typical transient operating conditions at constant speed or constant torque could be realized by a conventional eddy current dynamometer combined with the developed transient condition control system.3) The signals of engine cycles and crank angles obtained by sensors were identified by the ECU (Electronic Control Unit) to trigger the action of sampling mechanism. Before each sampling action, released determinate volume of nitrogen gas under NTP (Standard Conditions for Temperature and Pressure,NTP) into the flexible sampling bag to quench the chemical reaction of cylinder contents gathered latterly. 2. Development of sampling mechanismAn impact theory based total in-cylinder sampling mechanism was developed. The kinetic performances of the developed sampling mechanism were simulated by ADAMS software.1) The sampling mechanism was composed of electronic magnet, spring, impact anvil, sampling valve, and so on. This system was simplified by removing the intake and exhaust valve deactuators and injection system controller which indispensably in conventional sampling system. The sampling tube could be sealed and shut off timely by sampling valve, so the sampling process could be completed without shutting down the engine.2) The two key factors which affected the intensity of sampling mechanism were the structure of sampling valve and the impact energy.3) The smaller valve head's diameter, the shorter valve length, the bigger valve staff's diameter, and the heavier impact anvil were chosed, the higher reliability of the sampling mechanism could be obtained satisfied with three conditions: sampling more than 80% mass of cylinder contents, the maximal impact force less than the value that sampling valve could endure, and the maximal height during valve rising period more than 3mm.4) If the valve head's diameter is 26mm, the valve staff's diameter is 10mm, the valve length is 158.4mm, the impact anvil's mass is 3kg, the gas's pressure is 10MPa, the gas's temperature is 2000K, the impact velocity is 2.1m·s-1, then the safety factor of sampling mechanism is 2.1 according to the simulation results.3. Performance analysis of developed total in-cylinder sampling systemThe performances of the developed sampling system were obtained by two kinds of tests: simulative cylinder tests and engine bench tests. The simulative cylinder tests were operated with a fixed volume cylinder under atmosphere temperature. The engine bench tests referred to the sampling actions on diesel engine under the conditions of 1500r·min-1/30N·m and 1500r·min-1/50N·m. 1) The maximal difference of sampling system's delay time is 9ms and the mean square error (MSE) is 2.18ms. They are 1.125 times and 1.5 times of the conventional sampling systems respectively.2) The results of simulative cylinder tests showed that the maximal height of valve rise, and the duration of valve open were enlarged with the decreasing of initial gas pressure under the conditions of same impact energy, but the duration of pressure drop to 50% and 80% of its initial pressure were almost steady, they were 1.4ms and 3.4ms respectively.3) The results of engine bench tests showed that the sampling process was different from simulative cylinder tests because of the changing cylinder volume and temperature during sampling process. The mass ratio of cylinder contents decreased from 80% to 50% when the blowdown angle increased from 0°CA ATDC (Crank Angle After Top Dead Center) to 35°CA ATDC. The half decay period of pressure drop was 10 to 20 degrees of crank angle when the blowdown angle just between 0°CA ATDC to 60°CA ATDC, it was 2～3 times of the conventional sampling systems.4) The sampling process had a little affect on engine's speed. The speed dropped nearly 3% but could return to the same level a short time later.5) The tests proved the high reliability of the sampling system.4. Experiment study on emission formation historiesThe experimental study was carried out to produce cylinder histories of particulate, nitrogen oxides and total hydrocarbon.1) The mass-averaged histories of PM (Particulate Matter) and DS (Dry Soot) had the shapes of one peak value. DS appeared when the combustion starts. At about 20°CA ATDC, DS reached its peak value, then largely decreased after that. Compared to the fuel of 0# diesel, GTL (Gas to Liquid) diesel could produce less PM and DS emissions, but EGR (Exhaust Gas Recirculation) made the PM and DS emissions worse. Compared to the fuel of 0# diesel, the peak values of DS forming histories were both reduced when the GTL or EGR techniques were used.2) The histories of particulate quantity, surface area, and volume concentration had the shapes of one peak value. The position of the peak at about 30°CA ATDC. When engine's load increased from 30N·m to 50N·m, the total particulate quantity, surface area, and volume increased 2.3, 1.7, and 1.7 times respectively.3) In this research, the distribution of particulate quantity had two peak values within 0.01～1.0μm. The first one was not evident, but the second one occurred when the particulate diameter was 0.133μm. The second peak value was much smaller than the first one. The distributions of particulate surface area and volume had one peak value within 0.01～1.0μm. The peak value of surface area occurred when the particulate diameter within 0.075～0.237μm, and the peak value of volume occurred when the particulate diameter was 0.237μm. The concentration of particulates which emerged in the metaphase of combustion was higher than that emerged in the prophase and anaphase of combustion. The concentration of middle-scale particulates which emerged in the anaphase of combustion was higher than that emerged in the prophase of combustion, while the concentration of large-scale particulates which emerged in the anaphase of combustion was on the contrary.4) The volume concentration of HC (Hydrocarbon) and CO (Carbon Monoxide) had the shapes of one peak value while NOx (Nitrogen Oxides) had the"S"shape. When the load was 30N.m, the emission value was about 9% relative to peak value of CO, while the load was 50N.m, the ratio was 11%. Most NOx were produced within 20°CA after the combustion began. The emission of NOx increased with the increasing of load.5) The mass-averaged histories of PM and DS were different between transient operating conditions and steady operating conditions. The peak values of PM and DS were lower while the emission of DS was higher at transient operating conditions than that of the corresponding steady operating conditions. The HC, CO and NOx emissions were reduced appreciably at transient operating conditions. The differences of boundary conditions had greater effects on prophase than anaphase of transient operating conditions.
|
PDF Full Text Request