To reduce the wear of the full-floating piston pin friction pair in a certain high-power marine diesel engine, a polynomial profile expression method with continuous curvature at the segment was proposed, and the profile of the pin hole was optimized. A sub-model for the contact area between the pin and pin hole was established based on the thermo-mechanical coupling model of the piston pin friction pair for pin hole profile optimization, which can improve computational efficiency by 120 times. A high-order curve was designed based on the foundation of the previous job to replace the original conical profile. The pin hole profile was optimized using the sub-model and a Python-driven automatic surface design method, resulting in a 12.7% reduction in the maximum stress of the pin hole. A thermo-elastic hydrodynamic lubrication model for the piston-pin-connecting rod was constructed to compare the lubrication and wear performance before and after the pin hole profile optimization. The effectiveness of the profile optimization was verified through a single-cylinder engine test.

A three-dimensional numerical simulation model was developed to investigate the impact of key parameters including injection pressure, orifice diameter, injection direction, and incoming flow velocity on the droplet size distribution of methanol spray. The results reveal that increased injection pressure leads to a shift of the droplet size distribution towards smaller droplet sizes, accompanied by reduction in Sauter mean diameter（SMD）as well as characteristic diameters DV10, DV50 and DV90（DVX refers to the droplet size of the droplets which accounts for X% of the spray dropet cluster）. As the orifice diameter decreases, SMD and characteristic diameters of methanol spray droplets decrease, and the span of droplet size（SDS）and wall film quality also decrease. When the injecting angle is 45° against reverse incoming flow, more small-sized droplets are generated. Increasing the incoming flow velocity causes the spray droplet size distribution to shift towards smaller sizes, resulting in decreased SMD and characteristic diameters.

Based on the actual composition and physical properties of diesel fuel, n-dodecane was selected as the test fluid to observe and analyze the dynamic behavior characteristics of a single 100 ℃ droplet impacting cryogenic walls. The research results showed that, unlike the low-temperature wall impingement process of room temperature droplets, the upper surface of the liquid film in the process of high-temperature droplet impingement produced a translucent shadow structure which spreaded with the liquid film. As the wall temperature decreased and the initial velocity increased, the color of the shadow structure deepened, and the structure was gradually twisted in a wave-like shape. The spreading process of high-temperature droplet impacting the wall was also greatly changed compared with that of room temperature droplets. The existing models for predicting the maximum spreading factor max and the time of predicting the maximum spreading diameter tmax under room temperature conditions are not applicable to the test conditions of 100 ℃ droplet impacting the low-temperature wall. An analytical fit was obtained to model the liquid film spreading of high temperature droplet hitting the wall. The wall splashing threshold of 100 ℃ high-temperature droplets also changed greatly compared with that of room temperature droplets, and a droplet splashing threshold model suitable for this experiment by comparing and analyzing the existing models was obtained.

Based on the reaction path analysis method, the generation and consumption of NOx（NO, NO2, N2O）and the NOx chemical kinetic mechanism were studied. The NOx mechanism was combined with the NH3/H2 combustion mechanism, and the fuel-type N*Ox and air-type NOx generation characteristics of NH3/H2 combustion were investigated using the elemental labelling method. The results show that the combined mechanism can accurately predict the ignition delay time, laminar burning velocity, and the concentration of important components in a wide range of equivalence ratios. Adding 50% H2 does not significantly affect the oxidation pathway of NH3. The addition of H2 promotes the generation of large amounts of O, H, and OH radicals, which further promotes the oxidation of NH3 and is the main reason for the significant increase in the generation of fuel-type N*O and N*O2, as well as air-type NO, NO2, and N2O. The H radicals generated by H2 combustion promote the conversion of fuel-type N*2O and fuel/air-type NN*O, thus reducing the final production of both. In addition, fuel-type N*O, N*O2, and N*2O dominate the NO, NO2, and N2O produced by pure NH3 combustion or NH3/H2 combustion with a 50% H2 ratio.

In order to study the effects of diesel injection pressures and methanol substitution rates on the flame shapes, reaction zone properties and distribution of reaction intermediates in the development of methanol–diesel dual direct injection flame, the visualization of methanol–diesel dual direct injection flame was studied by using OH* chemiluminescence and flame self-luminescence high-speed synchronous imaging techniques on constant volume combustion chamber（CVCC）. The results show that when the injection pressure is low, the atomization is poor, and some oil droplets are fully burned after colliding with methanol. When the injection pressure is high, the injection pulse width is shorter and the combusting effect is weaker. The selection of methanol substitution rate needs to balance combustion performance and exhaust emissions, which also has an important impact on the shape of the dual direct jet flame and the distribution of diesel fuel in the flame. When the methanol substitution rate increases from 60% to 90%, the peak heat release rate decreases by 33.2%.

The methanol port injection with diesel ignition technology was tested on a marine methanol and diesel dual fuel engine. The study was focused on the combustion and thermal load for the combustion chamber between the methanol fuel mode and the diesel fuel mode. The test results show that the ignition delay is increased and combustion duration is decreased below 75% engine load in the methanol fuel mode. Meanwhile the exhaust temperature before turbine is also reduced in methanol mode. The temperature of the cylinder cover and liner is generally reduced in the methanol mode based on the measured temperature distribution of the cover and liner. But the temperature at parts of measured points varies greatly due to the combustion nonuniformity.

To improve the combustion efficiency and indicated thermal efficiency of ammonia diesel dual-fuel engines while reducing pollutant emissions, based on a large-bore marine engine, the 3D simulation software CONVERGE was used to analyze the combustion and emissions under different ammonia energy substitution ratios, different diesel injection timings and initial cylinder temperatures. The results show that with the gradual increase of ammonia energy substitution ratio and the increase of premixed combustion, the indicated thermal efficiency of the engine is improved, and the carbon emission is significantly reduced. However, the excessively high ammonia substitution ratio reduces the completion of combustion and the combustion efficiency, thereby decreasing the indicated thermal efficiency. When the diesel injection timing is appropriately advanced, the combustion efficiency is improved while the NH3 and CO emissions are reduced. Increasing the initial temperature in the cylinder improves the combustion environment, promotes the combustion rate of ammonia fuel, improves the indicated thermal efficiency of the engine, and significantly reduces N2O and unburned ammonia emissions. At an ammonia substitution ratio of 40%, a diesel injection timing of -21°, and an initial cylinder temperature of 338 K, the engine’s indicated thermal efficiency reached the highest level of 51.37%, while the N2O and unburned ammonia emission levels were very low.

To solve the issues of low thermal efficiency and high emissions caused by long injection duration and poor atomization quality of single injector mode under high-load condition of diesel engine, the diesel dual direct injection strategy was proposed. The combustion and emission characteristics of single and double injector modes under high-load conditions were studied by experiments, and further analyzed in detail by computation fluid dynamics（CFD）simulation method. The results show that the dual direct injector mode presents three stages of heat release. The first stage is the low temperature heat release of the auxiliary injector, the second stage starts from the high temperature heat release of the auxiliary injector, and the third stage is the high temperature heat release caused by the main injector. In addition, the ignition timing under the dual direct injection strategy is advanced, and the center of combustion is closer to the top dead center, thereby leading to the higher thermal efficiency.

According to requirements of engine performance improvement and exhaust emission optimization, engine bench test data including 6 factors, i.e., rail pressure, fuel injection timing, air inlet volume, pre-injection fuel quantity, pre-injection timing and turbine inlet pressure, were collected in the world harmonized steady-state cycle（WHSC）. The central composite design was used in design of experiment（DoE）. A response surface model was established by regression analysis and the effects of 6 factors on fuel consumption and exhaust emissions were studied. Then the optimal solution was obtained by the model to obtain the fuel consumption and the emission value. The results proved that the response surface method（RSM）model was accurate. The model showed that when the turbine inlet pressure increased from 700 to 830 Pa（gauge pressure）, and the fuel consumption rate changed by 2.0%. The fuel consumption rate can be reduced by calibrating turbine inlet pressure.

By using the CONVERGE software and the SAGE combustion model, the combustion process of dilute hydrogen ignited by active prechamber jet in a direct injection hydrogen engine was numerically simulated. Based on the experimental data of the hydrogen engine in the literature, the reaction rate of the key elementary reaction in the hydrogen sub-mechanism of the kinetics mechanism version 3.0 for methane oxidation proposed by the American Natural Gas Research Institute（GRI）was modified. Based on the modified hydrogen sub-mechanism of GRI3.0, the indicated mean effective pressure（IMEP）, knocking tendency and emission level of hydrogen engine with active jet ignition and spark ignition were compared under the condition of an equivalence ratio of 0.45. Under the condition that the crank angles at 10% heat release（CA10）are close, compared with spark ignition, four-hole pre-chamber jet ignition achieves a higher peak heat release rate, while the knocking tendency and NOx emission level increases slightly. The highest IMEPs of pre-chamber jet ignition and spark ignition are close. The flame propagation modes of jet ignition present difference because of the different jet penetration distances caused by different ignition advance angles.

In order to improve the response characteristics of the methanol injector, a one-dimensional simulation model of the methanol injector was built based on the AMEsim hydraulic simulation platform. Through the single factor analysis of the structural parameters such as the diameter of the control plunger, the preload force of the needle valve spring, the diameter of the spray hole, and the inlet and return oil throttling holes, the influences of the structural parameters on the response characteristics of the injector was researched. The parameters and their interaction were optimized by variance analysis using design of experiment method. The results show that changing the diameter of the return oil orifice has the most significant effect on the injector opening response, while changing the inlet oil orifice diameter has the most significant effect on the injector closing response. After comprehensive consideration of all parameters, the optimal parameter configuration is determined. The diameter of the control plunger is 8.8 mm. The pre-tightening force of the needle valve spring is 599.972 N. The diameter of the oil inlet orifice is 0.31 mm. The oil return orifice is 0.55 mm, and the diameter of the jet orifice is 0.86 mm. With this parameter configuration, the response characteristics of the methanol injector are significantly improved. And the opening response time and closing response time of the optimized design reduced by 28.7% and 21.1%, respectively, compared with the original state.

The development history of hydrogen internal combustion engines（H2ICEs）around the world was sorted out, and the detailed analysis of physicochemical characteristics and flame propagation of hydrogen was conducted with an emphasis on the current research status of key technologies including hydrogen injection, abnormal combustion, NOx emission, lubrication, safety, and engine performance. Driven by the global carbon neutrality, energy saving and emission reduction regulations, a significant progress has recently been made in hydrogen direct injection and its mixture preparation, hydrogen pre-ignition and detonation suppression, NOx emission aftertreatment and dedicated lubricants for H2ICEs . Under lean-burn conditions, the brake thermal efficiency of H2ICEs has now exceeded 45%, and near-zero NOx emission can be achieved. Finally, critical factors are outlined for the large-scale application of H2ICEs, and the needs for further improvement in safety and reliability are highlighted. Notably, hydrogen direct injection and ultra-high supercharging technologies are identified as pivotal solutions for enhancing both power output and fuel economy of H2ICEs.

In order to study the effects of blending biofuel and heavy oil on the particulate matter emission characteristics of marine engines, experimental studies of load characteristics（E2）and recommended characteristics（E3）were carried out on a two-stroke marine engine. Five fuels including B15 biofuel（i.e., a fuel blend of biodiesel and 180 marine low-sulfur heavy fuel oil（HFO）mixed in a mass ratio of 15%∶85%, and so on）, B24 biofuel, B30 biofuel, B50 biofuel, and 180 marine low-sulfur HFO were selected. The quantitative concentration of particulate matter in the exhaust gas at different particle sizes was measured by a particle sizer, and the results showed that the total number of particles in the accumulation mode（greater than 50 nm）of biofuel was reduced or converted into particles of a smaller nuclear mode（less than 50 nm）compared to 180 marine low-sulfur HFO. The emission factor of particulate matter was measured and calculated by membrane weighing, and it was found that the emission factor of particulate matter decreased as the proportion of biofuel oil increased under the load characteristic cycle condition. The particulate matter samples were collected by a thermophoresis probe, and the microscopic appearance of the particles was observed with transmission electron microscopy（TEM）. And it was found that particulate matter aggregates exhibited a chain-like structure.

To meet the high real-time computation requirements for the Diesel cycle process within the cylinder of a marine engine digital twin model, a single-zone real-time rapid prediction model with variable step-size explicit calculation was proposed. To improve computational efficiency, calculation points were discretized, step sizes for each phase were allocated according to precision and stability requirements, and physical equations were explicitly formulated to minimize computational errors. The model simplification involved adopting a polytropic process and improving the calculation method of specific heat ratio for the compression and expansion model, reducing combustion zones, simplifying combustion into a single-step chemical reaction, and optimizing combustion calculation points. The validation of the model was based on a 320 mm medium-speed marine engine. The results show that the calibrated peak pressure error is less than 2%, while the predicted peak pressure error is less than 3%. Both the errors for CA10（the crank angle corresponding to 10% of total heat release）and CA50（the crank angle corresponding to 50% of total heat release）are less than 1.5°. The computation time for the model is within 0.25 ms, enabling the model to achieve accurate predictions while simultaneously meeting the requirements for real-time performance.

To explore the influence of injection strategy on in-cylinder mixing and combustion, the iterative coupling method was used to study the effects of single hydrogen injection and secondary hydrogen injection in both conventional and linear engines, as well as the influence of different injection ratios on the linear engine. The results showed that the pressure and temperature in the two engines with secondary hydrogen injection were lower than those with single hydrogen injection. For the conventional engine, the pressure and temperature with secondary hydrogen injection were 0.1 MPa and 6.5 K lower than those with single hydrogen injection, while for the linear engine, the pressure and temperature with secondary hydrogen injection were 0.5 MPa and 29.0 K lower than those with single hydrogen injection. The pressure and temperature of the linear engine were lower than those of the conventional engine, with single hydrogen injection being 1.6 MPa and 93.0 K lower, and secondary hydrogen injection being 2.0 MPa and 116.0 K lower. The indicated thermal efficiency of the linear engine was higher than that of the conventional engine, with a difference of 1.5 percentage points for single hydrogen injection and 1.4 percentage points for secondary hydrogen injection. For both the linear engine and the conventional engine, the indicated thermal efficiency of secondary hydrogen injection was higher than that of single hydrogen injection, by 0.5 percentage points for the linear engine and 0.3 percentage points for the conventional engine. As the first hydrogen injection quantity decreased, both the in-cylinder velocity and turbulent kinetic energy of the linear engine gradually decreased until they reached a relatively uniform distribution within the cylinder. There were partial differences in the combustion reaction rate and mixture fraction changes in the cylinder after the top dead center, and the combustion in the cylinder was influenced by factors other than the hydrogen injection strategy.

A three-dimensional multiphase non-isothermal proton exchange membrane fuel cell（PEMFC）electrochemical model with coupled cooling channels was developed to investigate the effect of Al2O3 nanofluids on the heat transfer and output performance of the PEMFC. The heat transfer performance was evaluated by using the average membrane temperature, average membrane water content, and the index of uniform temperature（IUT）, while the feasibility of using Al2O3 nanofluids in the PEMFC cooling system was assessed based on net power density and power consumption ratio. The results indicate that Al2O3 nanofluids exhibit superior cooling performance compared to ethylene glycol coolant, with particularly enhanced cooling effects at lower coolant flow velocities, significantly reducing the average membrane temperature, thus increasing the average membrane water content. However, while the nanofluid coolant enhances the cooling effect, it may also lead to a decrease in the uniformity of the membrane temperature distribution. Specifically, when the coolant flow velocity was 0.1 m/s and the output voltage was 0.6 V, the average membrane temperature decreased from 359.66 K to 353.10 K, the average membrane water content increased from 9.91 to 11.53, and the IUT rose from 1.71 to 1.96. Al2O3 nanofluids can significantly improve the output performance of the PEMFC at low coolant flow velocities, leading to an increase of approximately 3.3% in net power density. Furthermore, under identical cooling conditions, Al2O3 nanofluids can enhance the cooling efficiency of the PEMFC while simultaneously reducing the parasitic power caused by the coolant.

To investigate the feasibility of eliminating the oil cooler in a hybrid electric vehicle（HEV）gasoline engine with a power density of 65 kW/L or above, a high-efficiency turbocharged engine specifically designed for HEVs was selected as the research object. Through experimental and simulation analysis methods, the engine operating conditions were analyzed. The changes in oil pressure, thermal balance, reliability, and fuel economy of both the engine and the vehicle before and after eliminating the oil cooler under the same operating conditions were compared. The research results show that, the cooling capacity of the system improves without the oil cooler. When the coolant temperature is 90 ℃, the wall temperatures of the cylinder head and cylinder block decrease by about 5 ℃ and 8 ℃, respectively. Meanwhile, the oil temperature rises, oil viscosity decreases, and the specific fuel consumption of the engine can be reduced by more than 0.8%. Under world light vehicle test cycle（WLTC）conditions, after eliminating the oil cooler, the oil temperature rise rate during the warm-up process decreases by 13.8%, while the coolant temperature rise rate increases by 19.0%, and the fuel consumption of the vehicle increases by 0.29%. Eliminating the oil cooler presents a low risk of oil temperature exceeding limits or causing engine failure under HEV operating conditions, having minimal impact on overall vehicle fuel consumption while effectively reducing vehicle cost and enhancing market competitiveness. The engine dyno thermal balance test can evaluate safe operating conditions and boundaries for the engine without an oil cooler under HEV conditions and predict engine oil temperatures under certain HEV operating conditions, thereby effectively saving vehicle test resources.

To investigate the impact of pump structural parameters on performance and internal flow characteristics, experiments were conducted on hydrogen circulation pumps with different structural configurations. Additionally, computational fluid dynamics（CFD）simulations were performed to compare results when using air and hydrogen as working fluids, with an in-depth analysis of the internal flow field and temperature distribution. Results show that HP-2 can achieve the same pressure increase and flow rate as HP-1 can at lower rotational speeds but with higher power consumption. The simulation trends for both air and hydrogen were consistent, suggesting that air can be used as a substitute for experimental studies. HP-2, with a smaller impeller diameter-to-width ratio, exhibited a higher proportion of the flow channel occupied by the blade passages, which weakened momentum exchange between the impeller and the flow channel. This resulted in a lower proportion of medium-to-high velocity regions in HP-2 compared to HP-1. The increased axial clearance at the volute tongue region also had a more significant impact on the inlet flow. Despite similar temperature distribution patterns under different operating conditions, the temperature difference between the inlet and outlet of HP-2 was 29% higher than that of HP-1. Moreover, at the same flow rate, HP-1 was 23% more efficient than HP-2. Additionally, HP-2 exhibited more chaotic velocity vector distributions, leading to increased secondary flow and mechanical losses. The maximum angular deviation of the velocity vectors from the radial direction within the impeller passages of HP-2 was approximately 10°higher than that of HP-1, resulting in greater impact losses due to stronger impingement on the impeller suction surfaces and outlet duct walls, ultimately reducing overall efficiency.

To address the issue of insufficient exhaust energy at low engine speeds in opposed-piston two-stroke（OP2S）engines, research on an electrically-assisted boosting system was conducted to enhance the boosting capability. A 1D simulation model for an electrically assisted turbocharged OP2S engine was established, and the in-cylinder energy and exhaust energy utilization efficiencies of pre-installed, mid-installed, and rear-installed electrically assisted turbocharging modes were analyzed. The results indicate that the exhaust energy is insufficient for the turbocharger to operate stably at low engine speed under 2 300 r/min. The electrically assisted turbocharging can extend the stable operating speed range of the OP2S engine. In addition, the increase of the exhaust energy at low engine speed makes the exhaust gas turbocharger which could not operate stably also have a supercharging effect, further increasing the pressure ratio. It suggests a synergistic effect between the two systems. The mid-installed mode exhibits the highest in-cylinder energy utilization efficiency, surpassing the lowest（the pre-installed mode）by 1.1 percentage points without compromising exhaust energy utilization.

In order to explore the suspected rust problem of local position on the surface of 38MnVS6Ti hypoeutectoid low alloy steel piston, the metallographic structure, microhardness, surface morphology and energy spectrum of the suspected rusted piston were comprehensively tested and analyzed. The results show that there is ledeburite structure on the surface of the local rust position of the piston. Through a comprehensive analysis of the manufacturing process and process of steel pistons, it is found that the high-carbon release agent sprays on the surface of the mold during the original forging process entered the molten iron after the partial high-temperature surface oxide of the blank melted, forming a high-carbon molten iron. The high-carbon molten iron undergoes eutectic reaction during the forging process of the blank to form a high-temperature ledeburite structure. The high-temperature ledeburite forms a room-temperature ledeburite after the forging blank is cooled at a controlled temperature. By reducing the forging temperature and diluting the release agent, the surface temperature and surface carbon content of the piston blank can be reduced, thereby blocking the formation conditions of the ledeburite on the local surface of the steel piston and eliminating the ledeburite structure. The optimized piston forging process has been verified by actual production. The verification results show that the improved forging process can solve the local rust problem of steel piston caused by ledeburite.

Considering the speed control characteristics of a low-speed two-stroke marine diesel engine, a modified active disturbance rejection control（ADRC）strategy was proposed, which is more suitable for low-speed diesel engine speed control. In order to reduce the impact of inherent speed fluctuations in diesel engines under small error condition, and to reduce the recovery time under big error situation, a strategy combined with linear extended state observer（LESO）for estimating and compensating the total disturbance was designed for constructing an improved ADRC method, in which linear feedback control law（EFCL）was employed when the absolute value of the error was smaller than 1, while nonlinear EFCL with “large error, large gain”was used in the else condition. To verify the proposed method, a cylinder-by-cylinder mean value engine model（MVEM）that can embody the cyclic speed fluctuations was built for fully considering the inherent speed fluctuations of diesel engines. Based on the proposed engine model, the performance of the proposed controller was verified through a hardware-in-loop（HIL）testing platform. The results demonstrate that the improved ADRC can balance multiple static and dynamic performance indexes. The proposed controller not only has good robustness to changes in the rotational inertia of the diesel engine, but also is insensitive to variations in its own parameters. Moreover, it is easy to tune the control parameters, which is of great significance for the practical application of ADRC technique in low-speed two-stroke marine engines.

In response to the inadequacy of traditional crankshaft vibration dampers in adapting to system structure and operating conditions, as well as their limited ability to achieve multi-harmonic, wide-frequency vibration control, a novel crankshaft vibration damper featuring controllable magnetorheological（MR）damping characteristics was designed. The damping torque characteristics and adaptive damping control effect of the crankshaft were analyzed and validated. A new disc-type MR grease torsional damper was designed based on theoretical analysis of variable damping torsional control and the structural configuration of traditional silicone oil torsional dampers. Mechanical properties of the MR torsional damper were experimentally tested under various vibration excitation conditions and control currents on a dedicated test platform. A dynamic model for the device based on the tangent hyperbolic model was established. Combined control simulations using AMESim and MATLAB were conducted to compare and analyze the torsional vibration control effects under different scenarios including uninstalled damper, passive control, skyhook control, and fuzzy control during steady-speed and steady-acceleration engine operation. The results indicate that the designed MR torsional damper exhibits variable damping mechanical output characteristics with an adjustable torque range of 4.36, meeting design requirements for variable damping torsional vibration control. The implementation of semi-active control strategy for MR torsional vibration damper can significantly enhance the vibration damping effectiveness of the crankshaft system. Among these, the fuzzy vibration damping control algorithm demonstrates the most comprehensive vibration damping effect, achieving a reduction in torsional angular velocity and angular acceleration by 27.2% to 69.5%. Following closely is the skyhook vibration damping control effect, while the traditional passive vibration damping control exhibits relatively inferior performance.