Against the backdrop of an increasingly severe energy crisis, developing electrochemical optimization approaches for biomass-fueled engines holds substantial importance. This study constructs a multi-fuel combined supply system based on gasoline port injection (GPI), bio-ethanol direct injection (BEDI), and negative pressure oxy-hydrogen intake, and explores the coupled influence mechanisms of bio-ethanol direct injection ratio (BEDIr), direct injection pressure (DIP), and oxy-hydrogen flow rate on the performance of a spark ignition (SI) engine. Experimental findings indicate that the introduction of oxy-hydrogen significantly improves combustion characteristics. Compared to the baseline condition without oxy-hydrogen and with 0% BEDIr, at an oxyhydrogen flow rate of 16 L/min, the peak cylinder pressure (Pmax) was increased by 7.4%, and the coefficient of variation of indicated mean effective pressure (COVIMEP) decreases by 37.4%. Oxy-hydrogen improves the engine's tolerance to high BEDIr conditions, and moderately increasing DIP can synergistically enhance the brake mean effective pressure (BMEP) under high BEDIr and significantly shorten flame-development and rapidcombustion intervals. In terms of emissions, increased BEDIr reduces CO and HC emissions, an effect strengthened further by oxy-hydrogen addition. For the increase in NOx emissions by oxy-hydrogen, high BEDIr (60-80%) effectively suppresses combustion temperature through its high latent heat of vaporization, forming a favorable complementary relationship in emissions.

Addressing the critical role of spray characteristics in mixture formation and combustion characteristics, this study investigates methanol-blended, ethanol-blended, and five pure alcohol fuels (methanol, ethanol, isopropanol, *n*-butanol, isobutanol). Spray dynamics were captured via high-speed photography and Schlieren imaging in a constant-volume chamber. Computational simulations isolated the effects of fuel properties (density, viscosity, surface tension) and ambient conditions (pressure, temperature) on spray morphology. A Random Forest (RF) model leveraging out-of-bag (OOB) error estimation quantified parameter importance and predicted spray characteristics (penetration, cone angle, area). After hyperparameter tuning, the optimized RF model achieved high-fidelity predictions (e.g., spray area: R2 = 0.98, RMSE = 118). Key findings reveal ambient pressure as the dominant factor influencing spray characteristics, followed by ambient temperature, while fuel properties exhibit operating-condition-dependent effects. This integrated experimental-computational-ML framework enables rapid spray parameter optimization and facilitates inverse design of alcohol-fuel nozzles and combustion chambers, advancing cleaner and more efficient combustion technologies.

Owing to its superior combustion performance, abundant reserves and clean combustion products, hydrogen has been regarded as a promising alternative fuel for internal combustion engines (ICEs). The direct injection hydrogen internal combustion engines can offer larger power without the risk of backfires, compared to port fuel injection. To avoid the NOx emission and improve the power output, lean boosting is adopted to reduce the pressure-increase rate and combustion temperature. As well as the low exhaust temperature of hydrogen engine, which pose significant challenges for the turbocharger system. Against this background, an experimental study was carried out on an 8.2 L DI hydrogen engine. Three types of turbochargers were compared: wastegate turbocharger (WG), variable geometry turbocharger (VGT), and two-stage turbocharger (TS). All tests were performed under operating conditions constrained by pre-ignition and super-knock, aiming to maximize thermal efficiency. The brake thermal efficiency (BTE) varied from 31.1% to 34.4% across the tested engine speeds. The TS turbocharger yielded the maximum BTE of 34.4% at 1300 rpm. This work further investigated the effects of different turbochargers on the energy and exergy distributions of the engine. Energy analysis results reveal that the two-stage turbocharge exhibits the lowest exhaust energy loss, the variable geometry turbocharger demonstrates the lowest coolant energy loss, and the wastegate turbocharger achieves the lowest intercooler loss. Exergy analysis results indicated that the TS system achieved the maximum total available exergy efficiency of 67.9% at 1600 rpm, accompanied by the minimum exergy destruction.

Amid the escalating global energy crunch, the sustainable clean energy is a core development direction for power systems. This study employs a combined experimental and numerical analysis approach to systematically investigate the effects of fuel supply strategies on combustion performance for an engine utilizing a ternary-fuel combined supply system consisting of "oxyhydrogen negative pressure inhalation + bio-isopropanol direct injection + and gasoline port injection" (ONPI + BIDI + GPI). The numerical analysis indicates that increasing the oxyhydrogen blend ratio enhances the peak concentrations of key radicals (OH, HO2, HCO) and intermediate products (H2O2, CO), optimizes the OH distribution in the cylinder, and raises the temperature in the cylinder. This explains the experimental observation that ONPI significantly improves engine combustion performance. Bench test results indicate that when the bio-isopropanol direct injection timing (BIDIT) is set in the intake stroke, HRR, Pmax, and combustion process parameters of the engine are significantly better than when set during the compression stroke. The engine's HRR, Pmax, IMEP, combustion process parameters, and CoVIMEP all improved with an appropriate increase in bio-isopropanol direct injection pressure (BIDIP), but excessively high BIDIP actually deteriorated the in-cylinder combustion environment. Notably, the coupled optimization of ONPI and BIDIT can enhance the power performance and combustion process of engines. Finally, this paper proposes "ONPIv = 16 L/min +250 degrees C A BTDC <= BIDIT <= 300 degrees C A BTDC + BIDIP = 9 MPa" as a feasible fuel supply strategy for improving the combustion performance of the system under the test conditions.

To investigate the influences of direct injection strategies on the combustion and emission performances of hydrogen/gasoline dual-fuel engines, a turbocharged direct-injection gasoline engine was modified into a dualfuel engine for hydrogen and gasoline, which was equipped with HDI) Experiments were conducted under stoichiometric and lean-burn conditions to investigate the influence of hydrogen direct injection timing, hydrogen ESR, and 2 on in-cylinder pressure, HRR, combustion phase, cycle variation, BTE, and exhaust emissions. The results show that hydrogen addition most significantly shortens the early flame development, thereby accelerating flame propagation, advancing combustion phase, enhancing Pmax and BTE, reducing HC and CO emissions due to more complete combustion, while increasing NOX emissions caused by elevated in-cylinder temperature. EOI remarkably affects mixture formation and combustion performance, and - 180 degrees CA ATDC is verified as the optimal injection timing, which provides sufficient mixing time and favorable combustion phasing. Under lean-burn conditions, the coupling control of ESR and 2 achieves a balanced trade-off between efficiency and emissions. Excessively lean mixture or overly retarded injection timing would prolong combustion duration, increase cycle variation and incomplete combustion products.

Combining biobutanol and oxyhydrogen in an SI engine can reduce fossil-fuel use and improve power, but oxyhydrogen increases NOx. Without sacrificing combustion stability, this work investigates lean-burn coupled with exhaust gas recirculation for a gasoline port injection + biobutanol direct injection + oxyhydrogen in-cylinder negative pressure indraft engine, across five oxyhydrogen flow levels, four exhaust gas recirculation ratios, and three excess air ratios. Results show that with lean-burn + exhaust gas recirculation, oxyhydrogen more effectively lowers the coefficient of variation of indicated mean effective pressure and increases indicated mean effective pressure, peak cylinder pressure, and peak heat release rate. With 16 L/min oxyhydrogen, the negative effects of 6-12% exhaust gas recirculation on CA 0-10 and CA 10-90 are mitigated for all excess air ratios, and the crank angle corresponding to peak pressure remains optimal under lean conditions when 6% <= exhaust gas recirculation <= 12%. Oxyhydrogen reduces CO and HC after exhaust gas recirculation, while lean-burn dominates CO reduction. Exhaust gas recirculation suppresses NO more than lean-burn. At 1.1 <= excess air ratios <= 1.2, the optimal exhaust gas recirculation is 12%, ensuring favorable in-cylinder conditions. Overall, lean-burn + exhaust gas recirculation effectively controls NO and maximizes thermal efficiency and renewable-fuel substitution. The optimal strategy is "oxyhydrogen = 16 L/min, exhaust gas recirculation = 12%, 1.1 <= excess air ratios <= 1.2".

Knock severely restricts the development of downsized and turbocharged gasoline engines, and fuel formulation optimization is a cost-effective knock suppression strategy. This study combines experimental tests, CFD simulations and reactive force field molecular dynamics (ReaxFF-MD) simulations to investigate the knock suppression mechanism of methanol blending in RON92 gasoline for a 2.0 L turbocharged spark-ignition engine, with methanol mass blending ratios (MR) controlled at 0%, 5%, 10% and 15%. Experimental results show that methanol blending significantly smooths in-cylinder pressure traces and reduces knock intensity (KI) without obvious impact on combustion speed, and the in-cylinder thermal environment and flame propagation characteristics show negligible differences under different MR, indicating fuel intrinsic properties dominate knock suppression. ReaxFF-MD simulations at 1800-2400 K reveal methanol has little effect on gasoline's thermal pyrolysis and initial fragmentation, but effectively inhibits the formation of key intermediates (CH2O and CO) associated with low-temperature oxidation and chain reactions. This work verifies the excellent knock suppression performance of low-proportion methanol blending, and clarifies its atomic-scale mechanism from intermediate evolution and product formation, providing a theoretical basis for high-performance anti-knock fuel design and engine knock control.

To enhance fuel-air mixing in compression ignition engines, the Lateral swirl combustion system (LSCS) was proposed and demonstrated excellent performance in diesel engines. However, no studies have yet applied LSCS to the increasingly popular diesel/ammonia dual-fuel engines. This study investigates the role of LSCS in diesel/ ammonia Reactivity Controlled Compression Ignition (RCCI) combustion and its potential to enhance zerocarbon fuel utilization. By introducing an LSCS chamber with a rim structure and developing a 3D computational model based on nitrogen tracking method, we examined the impact of Start of injection (SOI) and Swirl Ratio (SR) on in-cylinder combustion and emission characteristics. Results show that the LSCS rim structure enhances wall jet and improves air utilization. Across all SOI conditions, LSCS achieved higher Indicated mean thermal efficiency (ITE) than non-convex combustion system (NCCS), with a maximum improvement of 1.9%. The greenhouse gas (GHG) emissions were slightly reduced, while emission of unburned ammonia was reduced by approximately 13%. At SR = 1.25, the interaction between wall jet and swirl motion created optimal local equivalence ratio distribution, resulting in maximum ITE, an increase of 1.34% compared to the original LSCS case, as well as the minimum GHG and ammonia emissions. Therefore, LSCS demonstrates potential to further enhance diesel/ammonia RCCI engine performance while reducing GHG and unburned ammonia emissions.

At present, although bio-ethanol has become a widely used renewable energy source, it is still unable to fully replace fossil fuels in power systems. The key issue lies in the fact that its high latent heat of vaporization significantly deteriorates the combustion environment within the cylinder; however, the blended combustion of hydrogen-based fuels holds promise for resolving this critical issue. This paper presents simulation and experimental studies conducted on a gasoline port injection + bio-ethanol direct injection + HHO negative pressure inhalation (GPI+BEDI+HNPI) engine. The study investigates the optimizing effects of different fuel supply strategy coupling schemes (BEDIr, BEDIT, HNPIv) on the engine’s combustion and emission characteristics. The results indicate that earlier BEDIT and HNPI coupling can effectively mitigate the negative effects of a larger BEDIr on engine power performance, the combustion process, and combustion stability, while also improving the engine’s tolerance to a larger BEDIr. While HNPI can effectively cut HC and CO emissions, it results in a rise in NOx emissions. Increasing the BEDIr helps mitigate this issue. Overall, the fuel supply strategy “HNPIv = 16 L/min + BEDIr = 60% + 300°CA BTDC ≤ BEDIT ≤ 350°CA BTDC” enables the engine to achieve relatively ideal emission levels while maintaining combustion characteristics.

The diesel/ammonia combustion mode can significantly cut down carbon emissions, while the growth in fuel NOx due to the introduction of ammonia poses a potential emission problem. To explore the generation and reduction mechanisms for different sources of NO (oxidant-source NO and fuel N*O), an ammonia/n-heptane chemical kinetic mechanism was constructed. With the N element tracking method, the different sources NO emission characteristics of a diesel/ammonia reactivity-controlled compression-ignition (RCCI) engine were probed through varying diesel start of injection (SOI), injection pressure, nozzle hole number, and spray angle. The results indicated that in a diesel/ammonia RCCI engine, the production of fuel-source N*O occurred earlier than that of oxidant-source NO. During the NO reduction process, thermal DeNOx mainly occurred at the center of the combustion chamber and near the edge oxidant-source NO, and fuel-source NO emissions initially increased and then decreased. The area of unburned ammonia distribution is most enriched near the cylinder wall. As the SOI increased, at higher injection pressures, fuel-source NO was generally higher, while at medium injection pressures, oxidant-source NO emissions were higher. The trends of oxidant-source NO and fuel-source N*O with an increasing nozzle hole number were opposite, while the change in total NO emissions was not significant. When the spray angle was reduced to 122 degrees, the total NO emissions decreased by 49% compared to the original 162 degrees. This suggests that adjusting the spray angle can significantly lower NO emissions without reducing engine efficiency.

The combustion state of methanol/polyoxymethylene dimethyl ether (PODE) spark ignition dual fuel engine at high premix ratios requires further optimization. The combustion behavior of spark ignition dual fuel engines is governed by the coordinated control of spark ignition and PODE autoignition, and the influence of in-cylinder PODE distribution on combustion and emissions remains unexplored. Changing the configuration of direct-injection injector can create different fuel distributions within the cylinder, and the research on the synergistic relationship between in-cylinder fuel distributions and spark plug ignition in dual fuel engines is still a blank field. To change the injection area of PODE in the cylinder and explore the influence of in-cylinder PODE distribution on combustion and emissions, the original 5-hole injectors were modified to create a 2-hole injector with spray near spark plug and a 2-hole injector with spray far away from spark plug. Research findings indicate that compared to original 5-hole injector, the 2-hole injector with spray close to spark plug enhances PODE distribution near the spark plug, resulting in more stable combustion. However, the 2-hole injector with spray far away from spark plug reduces PODE distribution in the vicinity of spark plug, resulting in a cyclic variation of more than 10% at all test points. The overall fuel injection rate of the 2-hole injector is slow, which leads to delayed combustion phase and needs to be improved by advancing PODE injection timing. Compared with the original 5-hole injector, the 2-hole injector requires early injection timing to optimize the combustion state, and the advance degree of injection timing decreases with the increase of premix ratio. Additionally, the uneven distribution of PODE generated by the 2-hole injector leads to an increase in incomplete combustion and a slight decrease in thermal efficiency. The coordinated optimization of the arrangement of direct-injection injector and spark plug needs to be considered.

This study experimentally and numerically evaluates the impact of ammonia-water direct injection (DI) nozzle configurations on knock suppression in gasoline spark-ignition (SI) engines under high-load conditions. The original six-hole injector was modified into two single-hole variants: DI-1 (nozzle holes oriented toward the piston) and DI-2 (holes directed toward the cylinder head). Experiments analyzed 12 end-of-injection (EOI) timings (-330 degrees CA to 0 degrees CA ATDC) and varying direct injection ratios (DIr) to assess knock suppression, combustion stability, and temperature distribution. Numerical simulations using CONVERGE software validated experimental results and provided insights into in-cylinder spray dynamics and flame propagation. Findings revealed that DI-1 significantly outperformed DI-2, achieving a broader effective knock suppression range (KI reduction up to 42 %) at minimal pulse widths (DIr = 3.5 %) due to enhanced cooling of the end-gas mixture via piston-directed spray alignment with tumble flow. Early-phase injection near bottom dead center (BDC, -180 degrees CA ATDC) optimized cooling efficiency by maximizing spray dispersion during compression, whereas delayed injection post-BDC diminished suppression due to altered spray-wall interactions. The original six-hole injector, despite requiring higher DIr (9.5 %), demonstrated robust cooling performance but faced challenges in precise low-flow control. Combustion analysis highlighted that ammonia-water injection reduced peak cylinder pressure (by 8-12 %) and delayed combustion phasing (CA50 by 4-6 degrees CA), necessitating a balance between DIr and ignition timing to avoid excessive retardation. Temperature field simulations confirmed DI-1's superior peripheral cooling, reducing high-temperature regions (>730 K) by 18 % compared to DI-2. This work underscores the viability of ammonia-water as a knock inhibitor and proposes nozzle orientation optimization as a critical pathway for minimizing inhibitor consumption while maintaining thermal efficiency. The results provide actionable insights for designing high-performance DI systems in downsized, high-efficiency SI engines.

The use of alcohol fuels in automotive engines holds significant potential for reducing reliance on fossil fuels and mitigating greenhouse gas emissions. This study investigated the macroscopic spray characteristics of methanol, ethanol, isopropanol, n-butanol, and isobutanol using a constant volume chamber and the schlieren method. The effects of varying ambient temperatures, ambient pressures, injection pressures, and fuel temperatures on the spray characteristics of these alcohol fuels were systematically analyzed. An artificial neural network (ANN) was employed to predict spray characteristic parameters, with Bayesian optimization applied for hyperparameter tuning, leading to the development of an optimal ANN model utilizing fuel properties as input. The results revealed that although the overall trends in spray characteristics were consistent among different alcohol fuels under varying conditions, the magnitude of changes differed significantly due to variations in their physicochemical properties. The developed ANN model demonstrated high predictive accuracy (R-2 > 0.98), closely matching experimental data. This study highlights the potential of ANN-based models to reduce experimental workload in the development of carbon-neutral fuels and advanced fuel injection systems, contributing to cleaner and more efficient combustion technologies.

The injection behavior of fuel is intimately linked with the engine's combustion process and emissions. Accurate modeling and predicting the spray characteristics of these fuels are essential to the effective large-scale deployment of alternative fuels and the development of corresponding fuel supply systems. Spray test data for methanol, ethanol, isopropanol, n-butanol, isobutanol, and gasoline under various operating conditions were collected through schlieren imaging in a constant volume chamber. This study addresses two primary objectives: first, to predict the spray tip penetration (STP) for each tested fuel using both response surface methodology (RSM) and artificial neural networks (ANN), and to compare the predictive performance of these two methods. Second, to fill the gap in the absence of a universal model for the spray characteristics of alcohol fuels, multiple topological indices, which are tools for converting molecular structure information into numerical descriptors, were utilized to describe the alcohol fuels and to develop an ANN model. The findings revealed that both the RSM and ANN models exhibited high accuracy and reliability in predicting the STP for individual fuel datasets, with the RSM model marginally outperforming the ANN model in terms of fit quality. The ANN model, developed using topological indices to describe alcohol fuels, achieved R2 of 0.9939, 0.9936, and 0.9949 on the training set for predicting three spray characteristic parameters, demonstrating high precision, robustness, and reliability. The ANN model developed in this study, based on topological indices, shows significant potential to serve as a universal model for predicting the spray characteristics of alcohol fuels across a range of operating conditions. Expanding the database to incorporate additional fuels and spray conditions would allow the universal model to be used for promoting the efficient and clean use of alcohol fuels and advancing the development of highefficiency, low-carbon internal combustion engines.

The pre-chamber engine wall's structure has a significant effect on the ignition and combustion performance pre-chamber turbulent jets. This study identified an abnormal combustion phenomenon in the case of secondary jets, which was characterized by the pre-chamber pressure rise rate exceeding 30 times the normal rate owing pre-chamber jet impingement on the wall. To investigate this phenomenon, we examined the causes of ignition events resulting from the interaction between pre-chamber jets and the wall by using a visualized constant volume combustion chamber experimental platform and by conducting CFD simulations. The results indicated that for a certain range of the wall-nozzle distance, pre-chamber jet impingement on the wall is advantageous for reducing the ignition delay and increasing the combustion speed. Under lean burn conditions, pre-chamber jets impinging on the wall led to both direct and delayed wall-impinging ignition owing to the wall structure. Pre chamber jet impingement on concave or flat wall surfaces generated more intense impinging turbulence than that on convex wall surfaces, resulting in the improvement of both ignition and flame propagation speeds. Thus, the wall surface at the impingement point of pre-chamber jets should not be convex as this would reduce the jet ignition performance and increase the probability of secondary jet occurrence, leading to irreversible damage the pre-chamber structure.

The real driving emissions of gasoline and diesel vehicles are significantly influenced by altitude, temperature, and starting conditions. In this study, the real driving emissions (RDEs) of gasoline and diesel vehicles compliant with China V standards were investigated under various conditions. The adaptability of RDE testing in China was evaluated by analyzing vehicle emissions at different altitudes, ambient temperatures, and starting conditions. The results show that, with increasing altitude, CO, NOx, and PN emissions generally exhibit a downward trend, particularly for gasoline vehicles, whose conformity factors remain well below the China VI limit. However, for China V diesel vehicles relying solely on EGR technology, NOx emissions significantly exceed China VI standards, indicating that EGR alone is insufficient to meet regulatory requirements. Temperature variations have little effect on the emissions of China V PFI gasoline vehicles, while diesel vehicles continue to exhibit excessive NOx emissions under varying temperatures. Although the cold-start phase generates substantial pollutant emissions, the EMROAD evaluation method excludes this phase, resulting in limited differences between cold- and hot-start emission results. Nevertheless, the inclusion of cold-start emissions should be considered in future RDE assessments.

With the ongoing downsizing of engines and the increasing adoption of direct injection systems, flash boiling and spray-wall impingement have become inevitable phenomena, particularly for methanol and ethanol, which are promising alternatives to gasoline. This study experimentally investigated the wall-film characteristics after spray impingement under varying flash-boiling intensities in a constant-volume chamber using high-speed imaging combined with the Refractive Index Matching (RIM) technique. Key parameters-the ambient-tosaturation pressure ratio (Pa/Ps), fuel temperature, and ambient pressure-were systematically varied. The results show that as Pa/Ps decreased from 1.0 to 0.15, sprays transitioned from non-flash to flare flash-boiling regimes, accompanied by spray collapse and droplet coalescence, leading to a morphological change in the wall film from uniform to converging curved structures. Increasing flash-boiling intensity markedly reduced the overall film area but caused localized thickening, which was more evident for methanol than for ethanol. Fuel temperature exhibited nonlinear effects: maintaining moderate superheating (approximately 313-343 K for methanol and ethanol) effectively reduced wall wetting, whereas excessive heating intensified spray collapse and film thickening. At low ambient pressures (0.03-0.01 MPa), flash boiling further accelerated film evaporation, leaving only minimal residues for alcohol fuels. These findings provide guidance for optimizing fuel temperature and injection strategies to minimize wall wetting and improve mixture formation in gasoline direct injection (GDI) engines.

Internal combustion engines are evolving to attain advanced combustion modes with higher percentage of compression ignition combustion. Homogeneous charge compression ignition is limited by load range and combustion stability, so hybrid combustion modes have emerged as a key research hotspot. In order to optimize hybrid combustion modes, combustion modes should be accurately identified. Therefore, this study introduces new criterion and method to identify combustion modes. The results show that high compression ratio spark ignition engines are prone to autoignition, however, a trade-off relationship exists between high percentage of compression ignition combustion and the optimal combustion phase. As low percentage of compression ignition is favorable for engine efficiency, reduction of the combustion intensity should be considered. Under spark ignition assistance, both single-fuel and dual-fuel combustion exhibit a combustion mode where spark ignition and compression ignition occur simultaneously, which is defined as spark-assisted compression ignition combustion mode. The peak value (HP) for the second derivative of heat release rate at the onset of a combustion mode is strongly correlated with burning intensity and can be used to identify different combustion modes. The HP of spark ignition combustion is less than 1.5 J/deg3. Spark-assisted compression ignition combustion has an HP between 1.5 J/deg3 and 5.0 J/deg3, while the HP of premixed compression ignition combustion is at least 5.0 J/deg3.

Abnormal combustion phenomena pose significant challenges for the application of hydrogen direct injection engines. This study investigates the suppression of these issues in pure hydrogen direct injection engines through the use of variable valve timing and split injection strategies. The research addresses a critical gap in the current literature, as these strategies have not been widely explored in hydrogen direct injections, despite their successful application in gasoline engines. Results demonstrated that significant reductions in pre-ignition at early-mid compression-stroke events by up to 38 % were observed with specific valve timing adjustments, alongside decreased in knock intensity by 34.5 % and knock probability by 38.6 % with optimal intake valve closing timing set at 227 degrees crank angle before top dead center. Furthermore, configurations utilizing split end of injection timing at 30 degrees crank angle before top dead center and split injection ratio of 0.6 achieved a substantial 64 % reduction in specific pre-ignition events and split injection ratio of 0.7 resulted in a 63.0 % and 63.7 % decrease in knock intensity and knock probability, respectively, compared to initial valve timing and single injection. However, it was found that higher split injection ratio values, while effective in reducing abnormal combustion, led to diminishing returns in brake thermal efficiency. Achieving optimal performance required a careful balance between suppressing pre-ignition at early-mid compression-stroke events and knock phenomena while maintaining or improving brake thermal efficiency. The findings offer critical insights for advancing hydrogen direct injection engine technology, promoting decarbonization in transportation, and supporting global carbon neutrality goals. Future research should focus on optimizing variable valve timing and split injection strategies for various operational conditions and explore their integration with other advanced technologies, such as turbocharging and exhaust gas recirculation, to further enhance hydrogen combustion stability.

Hydrogen fuel holds great promise for accelerating the energy sector's decarbonization efforts. However, hydrogen's unique properties pose challenges for hydrogen direct injection (HDI) engines, such as differences in airflow exchange capabilities for very low density and reduced combustion rates due to lean burn conditions. This study investigates the impact of combined variable valve timing (VVT) and split injection strategies on HDI engine combustion, efficiency, and emissions. Using Artificial Neural Network (ANN) models and experimental data, predictive models for Brake Thermal Efficiency (BTE) and Nitrogen Oxide (NOx) emissions were developed. Adjusting intake and exhaust valve timings improved BTE by up to 2.5% and 1.8%, respectively, while reducing NOx emissions by 48.9% and 31.1% under specific conditions. Optimizing split injection alongside VVT further increased BTE by 1.2% and 0.9%, but increased NOx emissions by 29.9% and 26.5%. The flexible application of VVT and split injection strategies aims to balance efficiency gains with emissions control. The ANN-based models demonstrated high accuracy (R2 > 0.974), proving reliable substitutes for real-engine testing in certain conditions. In conclusion, this research highlights the potential of VVT and split injection strategies to enhance HDI engine performance while mitigating environmental impact, supporting the transition to sustainable energy solutions.