In order to better match the tandem composite turbine and study the utilization of engine exhaust gas energy under various working conditions, the thermodynamic model of a gasoline homogeneous charge compression ignition(HCCI) engine and the mathematical model of its tandem composite turbine system were established. A one-dimensional simulation model of the HCCI engine was also constructed. Based on the tandem composite turbine parameters calculated by the mathematical model, an appropriate tandem composite turbine system was selected, and the one-dimensional simulation verification was carried out. Based on the matching results, the energy distribution in the tandem composite turbine system was studied. The results show that the tandem composite turbine system selected by the proposed matching method can effectively recover the exhaust gas energy and improve the performance of the whole machine. At a speed of 2200r/min and an indicated mean effective pressure(IMEP) of 1.26MPa, the power of the whole machine is increased by 3.17kW, and the efficiency of the whole machine is improved by 2.49%. The exhaust gas energy distribution results show that the power turbine is suitable for the working condition of the engine speed higher than 1600r/min and the IMEP greater than 1.07MPa. There is an optimal intake pressure at this time, so that the exhaust gas energy obtained by the power turbine has the best proportion of exhaust gas energy, leading to an optimized engine thermal efficiency.

Utilizing a repetitive control-based proactive controller known as the‘virtual crankshaft’, diverse piston trajectory operating characteristics under the same compression ratio(represented by parameter Ω, which is the ratio between the minor axis and the major axis of the ellipse generating trajectories)were achieved in a free piston engine. Through an analysis considering characteristics such as indicated thermal efficiency, peak pressure location, ignition delay, nitrogen oxide emissions, etc., an optimal piston trajectory based on parameter Ω was obtained. Using this as the initial condition for optimization parameters, a global optimization design of the piston trajectory was performed based on a genetic algorithm, with indicated thermal efficiency as the optimization objective. The research results demonstrate that changes in the piston trajectory affect ammonia HCCI combustion processes. The optimized trajectory obtained through the design optimization process shows a 3.35% improvement in indicated thermal efficiency compared to traditional internal combustion engines, along with a 42.93% reduction in ignition delay time.

Based on a 3-cylinder 1.5L dedicated hybrid engine, the influence of three different ignition modes, using single spark plug, passive pre-chamber and triple spark plugs, on the engine combustion and emission characteristics was studied under lean-burn conditions. The test results at the speed of 2000r/min and the brake mean effective pressure of 0.8MPa show that under high-tumble intensity, the lean limit with passive pre-chamber ignition corresponded to an excess air ratio ϕ_a of approximately 1.70, which is slightly lower than that of single spark plug. In contrast, triple spark plug ignition can maintain stable combustion with a ϕ_a greater than 1.95, thereby improving the fuel consumption and raw NO_x emissions. As the air dilution increases, the advantage of passive pre-chamber ignition in improving combustion phasing gradually weakens, while the corresponding ignition delay and combustion duration increase more obviously. Triple spark plug ignition can effectively accelerate the combustion rate, and this advantage becomes more significant especially when ϕ_a exceeds 1.70, the combustion duration could be shortened by nearly 4°CA compared to those of single spark plug and pre-chamber. When using triple spark plug ignition and a lean mixture with ϕ_a of 1.80, the engine achieved a highest brake thermal efficiency(BTE) of 45.06% with the fuel of 92 research octane number(RON). With RON100 fuel and further utilization of electronic supercharger, the highest BTE can be increased to 45.63% and 47.04% respectively.

Based on a full-scale transparent nozzle internal flow and spray visualization experiment platform, an experimental study on the effect of taper of hole(taper factor of 1, 2 and 3) and number of hole(single, two and three holes) on the cavitation transient flow and spray in the nozzles was carried out for pure diesel(D100) and gasoline-diesel blended fuel of D20G80(20% diesel and 80% gasoline). The results show that the effect of the nozzle taper on cavitation and spraying in the nozzles is obviously different for the pure diesel and the gasoline-diesel blended fuel. The blended fuel has strong string cavitation with larger spray cone angle and fluctuations in the spray edge in the nozzles with all tested nozzle tapers, while the pure diesel only has string cavitation in the nozzle with a taper factor of 2. For the nozzles with a taper factor of 2, string cavitation occurs for both the pure diesel and the blended fuel in two-hole nozzles. It also occurs in the single-hole nozzle, but with less intense. Geometry-induced wall cavitation mainly occurs in the three-hole nozzle and spray edge fluctuations are stronger than that in the single-hole and two-hole nozzles.

When injecting hydrogen gas into the intake air path and in-cylinder environment of an internal combustion engine, there is a transition from supercritical to subcritical between the injection pressure and ambient pressure. High-speed schlieren imaging technology was used to study the injection characteristics under different conditions in a constant volume combustion bomb. The results show that the penetration distance, jet surface area, and jet volume of the hydrogen jet vary with time under different conditions. However, the change in the jet cone angle after 1ms is not significant, and the difference remains within 2°. When the supercritical injection enters the supercritical environment, the interaction between the jet and surrounding gases enhances turbulence intensity, which facilitates the entrainment of more surrounding gases. An empirical formula for the jet penetration distance of the hydrogen jet was proposed, which has proved to be high accuracy. These results can be utilized to optimize the mixture formation and combustion of hydrogen internal combustion engines, providing references for hydrogen injection strategies.

The concentration distribution characteristics of ultra-high pressure natural gas jet were studied by laser induced fluorescence imaging method based on a high-pressure dual-fuel direct injection valve. The data processing method of natural gas jet concentration distribution was optimized by dividing the natural gas jet into shock core region, concentration rapid decline region, concentration mixing region and boundary diffusion region. The results show that the core region of shock wave is affected by the Mach disk, and the concentration peak appears at and is highly correlated with the location of the Mach disk. It is found in the study of the axial concentration of natural gas that the injection pressure has the greatest influence on the shock core region and the concentration rapid decline region near the nozzle. When the injection pressure is increased from 22MPa to 31MPa with a step of 3MPa, the length of the shock core region and the concentration rapid decline region increases by 60%, while the length of the concentration mixing region decreases by 2.2%. The boundary diffusion region is least affected, and the region length increases first and then decreases. The horizontal concentration study of natural gas also shows that the concentration of natural gas has a strong influence on the area near the nozzle. The axial concentration in the shock core region and the concentration rapid decline area increases with the increase of the injection pressure, while the concentration fluctuation area appears in the concentration mixing area and the boundary diffusion area far away from the nozzle.

An engine-after-treatment virtual simulation model was built for emission prediction of in-use light-duty vehicles under real drive emission(RDE) conditions. First, a one-dimensional engine model was built and calibrated with steady-state and transient experimental engine emissions. The results show that the relative errors of cumulative mass of CO and NO emissions are 9.2% and 5.8%, respectively, under RDE conditions. Secondly, the after-treatment performance tests were conducted, and the after-treatment temperature, light-off time and aging models were verified. The results show that during the first 35s in world light vehicle test cycle(WLTC), the maximum error of temperature in the axial direction of the front catalytic converter is 52℃, and the absolute(relative) errors of post-catalytic cumulative masses of CO and NO are 0.0385g(5.90%) and 0.0011g (0.75%), respectively. Finally, the emission patterns of WLTC, RDE and standardized random test-aggressive driving style (RTS-95) working conditions were investigated under different catalytic converter aging conditions using the virtual simulation model. It is found that NO emission is more sensitive to the working conditions and the risk of exceeding the limit is higher. The NO instantaneous mass emission threshold is set to 8mg/s, and the warning window width was investigated for the RTS-95 operating condition with a catalytic converter dispersion factor of 0.6. The result shows that increasing the warning window width will reduce the misjudgment rate and the recognition probability of cold start overrun.

Under the engine testing conditions of world harmonized steady-state cycle(WHSC) and world harmonized transient cycle(WHTC), different ammonia nitrogen ratio(ANR) and exhaust temperature of the preceding stage and posterior stage were designed based on the selective catalytic reduction(SCR) reaction mechanism. The performance of the closed coupled SCR(cc-SCR) with different substrates at different ANR and exhaust temperatures was studied. The experiment result indicates that the addition of cc-SCR at the back of the supercharger can significantly improve the NO_x conversion efficiency of the after-treatment system. The ANR of the preceding stage injection should not exceed 0.8 when selecting cc-SCR. The advantage of the vanadium-based cc-SCR over copper-based cc-SCR is lower N₂O production. At low temperature, the NO_x conversion efficiency of the copper-based cc-SCR is superior to that of the vanadium-based cc-SCR. And at 300—450℃, the NO_x conversion efficiency of the two substrates catalysts is similar. At the steady-state operating point and exhaust temperature of 180℃, the overall NO_x conversion efficiency of the vanadium-based & copper-based catalyst program decreases as the ANR increases, while the overall NO_x conversion efficiency of the copper-based & copper-based catalyst program increases as the ANR increases. At high temperatures, the two scenarios convert NO_x equally. Under the hot cycle of WHTC, the overall NO_x conversion efficiency of the copper-based & copper-based catalyst program increases as the ANR increases.

To investigate the feasibility of utilizing diesel oxidation catalytic converter(DOC) technology in marine medium speed diesel engines, an experimental study was conducted to examine the effects of DOC on selective catalytic reduction(SCR) performance and physicochemical properties of particulate matter(PM) under typical low temperature and high flow conditions in marine diesel engines. The experimental results demonstrate that the utilization of DOC can enhance the catalytic efficiency of the SCR system in typical low temperature and high flow conditions, resulting in a 65% increase in catalytic efficiency under high load(18000h^-1) exhaust condition of the diesel engine. The addition of DOC can decrease the degree of graphitization in PM, and this effect is more pronounced at higher exhaust temperatures. The application of DOC can also enhance the concentration of oxygen-containing functional groups on the surface of PM, particularly leading to a 100% increase in the concentration of C—OH functional group. Consequently, this significantly enhances the oxidation activity of PM. In hydrogen-containing functional groups, the concentration of aliphatic C—H functional group, which serves as the growth nucleus of PM, is significantly reduced by 50%, thereby impeding the continued growth and aggregation of PM.

Based on density functional theory(DFT) and transition state theory(TST), the formation process of benzopyrene in the cylinder of internal combustion engines was investigated by phenyl-addition cyclization(PAC) mechanism. The results show that the entrance reactions of both anthracene free edge and sawtooth edge with phenyl exhibit barrierless, high exotherm and fast rate. By analyzing the rate-limiting step rates of each pathway, it is found that the anthracene free edge C1 site is more likely to generate benzopyrene than the sawtooth edge C2 site, and the optimal reaction pathway rate-limiting step energy barrier is only 176.2kJ/mol with a reaction rate of 3.7×10⁹s^-1. The rate-limiting step of the two main paths at the C2 position of the sawtooth edge is the isomerization of the five-membered ring into the six-membered ring. The energy barrier is up to 417.2kJ/mol and 421.7kJ/mol, and the reaction rates are 8.7×10⁴s^-1 and 5.8×10⁴s^-1, respectively. In addition, the H migration reaction on the aromatic ring is more difficult than on the aliphatic chain. This study provides the mechanism of PAC generation of benzopyrene, which gives a theoretical basis for the construction of chemical kinetics model of benzopyrene and is important for predicting the generation of benzopyrene in the cylinder of internal combustion engines.

Based on Simulink platform, a gasoline canister purge control strategy model was built. The model included modules like target purge rate calculation, target purge flow calculation and solenoid valve flow characteristic calibration. Model accuracy and air-fuel ratio control performance were verified on a revolving drum test table with flow gauge. Further, the characteristic parameters of control strategy were optimized based on the built model in order to increase purge flow. It shows that the accumulative purge flow during whole cycle is 402.69L and 373.62L for experiment and simulation respectively with 7.78% deviation. Regular purge rate factor, solenoid valve opening speed, limitation of solenoid valve maximum opening and initial purge rate factor all have significant impacts to canister purge flow independently. After applying comprehensive optimization of characteristic parameters, accumulative purge flow rate increases from 373.62L to 499.61L with superior control performance of air-fuel ratio. This research enlightens control strategy design of the gasoline canister to increase purge flow under the premise of satisfied air-fuel ratio control.