In-cylinder Flow Research Articles

Study of activity stratification-controlled NH3–H2 combustion in a large-bore gas engine utilizing split-channel supercharge and fuel-air mixing technology

In this study, the split-channel supercharge and fuel-air mixing technology (SCS-FAM) is firstly proposed for large-bore ammonia (NH3)-hydrogen (H2) engine, which involves the separation of fuel and air intake ports to mitigate the risk of H2 backfire in intake system. We explored the activity stratification under different intake schemes and H2 blend ratios through simulation, as well as how it affects the NH3–H2 combined combustion characteristics including in-cylinder flow turbulence, temperature, concentration, and key intermediates distribution. The results show that in Scheme 2 when air entered the high swirl ratio intake 1 and fuel entered the low swirl ratio intake 2, a rich mixture was formed near the spark plug, accompanied by localized NH3 thermal reforming phenomena, which improved the activity distribution of H2. Additionally, the strong swirl promoted combustion in the squish zone and significantly shortened the post-combustion period. Scheme 2 achieved a higher indicated thermal efficiency (ITE) than Scheme 1 with a small amount of H2 addition, with the highest ITE of 42.1% obtained at 15 vol% H2 blend ratio. Notably, the addition of H2 promoted NH3 consumption and the generation of active radicals, resulting in reduced N2O emissions during the expansion stroke, but causing an increase in NO emissions.

Influence of swirl, tumble and squish flows on combustion characteristics and emissions in internal combustion engine- review

This study gives an overview of available literature on flow patterns such as swirl, tumble and squish in internal combustion engines and their impacts of turbulence enhancement, combustion efficiency and emission reduction. Characteristics of in-cylinder flows are summarized. Different design approaches to generate these flows such as directed ports, helical ports, valve shrouding and masking, modifying piston surface, flow blockages and vanes are described. How turbulence produced by swirl, tumble and squish flows are discussed. Effects of the organized flows on combustion parameters and exhaust emission are outlined. This review reveals that the recent investigations on the swirl, tumble and squish flows are generally related to improving in-cylinder turbulence. Thus, more experimental and numerical studies including the impacts of this organized flows on turbulence production, combustion behavior and pollutant formation inside the cylinder are needed.

Understanding the role of methanol as a blended fuel on combustion behavior for rotary engine operations

Methanol, being one of the most promising carbon-neutral fuels, has the potential to enhance the fuel mixture homogeneity of gasoline rotary engines, thereby improving engine performance. This study, based on computational fluid dynamics model, investigates the in-cylinder flow, combustion, and emission characteristics under various methanol blending ratios, ignition timings, and equivalence ratios. At the methanol energy blending ratio of 10 %, a more extensive and efficient operating range is observed across different ignition timings and equivalence ratios, attributed to the higher diffusion and combustion rates. An increase in in-cylinder temperature enhances the complete conversion of HC. However, higher proportions of methanol are disadvantageous for the combustion process due to methanol's lower heat value and higher latent heat of vaporization. Thus, the optimal ignition timing was determined to be 30°EA BTDC. At this optimal ignition timing, the cylinder shows higher average turbulent kinetic energy and radical content compared to the ignition timing at 50° EA BTDC. Specifically, the concentrations of OH, H, and O free radicals increase by 12.40 %, 14.77 %, and 13.91 % respectively, leading to more complete initial flame kernel expansion. It is notable that emissions of CO, HC, and NOx are all reduced. The study further explores the influence of the equivalence ratio on the combustion process. At an equivalence ratio of 0.9, there is an earlier peak in the maximum cylinder pressure rise rate and the highest generation of O and OH, with a shorter main combustion period and concentrated heat release. This achieves economically viable operation for rotary engines, indicating a 2.122 % improvement in thermal efficiency and a reduction in fuel consumption to 375.537 g kW−1 h−1.

Computational analysis of piston shape effects on in-cylinder gas flow, fuel-charge mixing, and combustion characteristics in a two-stroke rod-less spark ignition opposed-pistons engine

Compared with the traditional in-cylinder direct-injection spark ignition engine, the side-injection and side-spark-ignition characteristics of the two-stroke opposed-piston engine increase the ignition kernel offset and flame propagation distance. Increasing the flame propagation speed can to some extent solve the drawbacks caused by the non-central arrangement of spark plugs. The combustion chamber structure plays a crucial role in gas flow, fuel-charge mixing, and combustion characteristics. Therefore, three pistons were designed and comparatively analyzed in this study. The results show that: The pancake piston is beneficial to maintaining the intake swirl strength due to its simple and smooth spherical arc structure. The swirl strength of the pit and pit-guided piston decreases obviously, and the tumble strength can be maintained well. Compared to pancake and pit-guided pistons, the average TKE for the pit piston increased by approximately 25%, with a more concentrated distribution at the spark timing. The pancake piston exhibits the best scavenging performance, reducing the residual exhaust gas ratio by 2.1% and fresh air loss by 3.3% to the pit piston. A stable ignition core can be formed at the spark timing, but significant differences are observed in the flame propagation process for three pistons. Compared to the pit-guided piston, the pit piston has a 0.3% decrease in the indicated thermal efficiency, but a 13.1% decrease in combustion duration, which reduces knock tendency.

Effects of continuous variable valve timing and duration on fuel/air mixture formation

With increasing global regulations on particulate emissions, the particulate matter and number released under cold start and high-load conditions have become crucial. Under such conditions, the particulate emissions from port fuel injection engines can be significant and comparable to those from gasoline direct injection engines. In this case, fuel evaporation and mixture formation contribute to the reduction in particulate emissions. One method to improve these characteristics is to enhance the in-cylinder flow and turbulence kinetic energy. Therefore, in this study, we investigated the effect of in-cylinder flow enhancement on the particulate emissions reduction in port fuel injection engines. Simulations were conducted with various intake valve timings and durations under cold-start conditions. We then performed engine experiments to confirm the reduction in particulate number. The results showed that the relationship between the intake valve open timing and piston speed was crucial. Opening the intake valve to its maximum value slightly after the point of peak piston velocity resulted in a higher flow intensity and turbulence kinetic energy during the compression stroke. In addition, the experimental results indicated that improving the in-cylinder flow using continuous variable valve timing reduced the particulate number by 27.51 %.

Comparative Analysis Of Cycle-To-Cycle Variabilities And Combustion Development In An Optical Spark-Ignition Engine Fueled By Pure Hydrogen And Propane: Insights From Chemiluminescence and PI

Hydrogen, derived from renewable sources and devoid of carbon emissions, is a pivotal energy carrier for the future. Representing a viable substitute for fossil fuels in internal combustion engines (ICEs), contemporary studies advocate using very-lean and ultra-lean hydrogen-air mixtures, with a fuel-air equivalence ratio below 0.5, as a potent strategy to reduce NOx emissions in hydrogen-fueled ICEs. An experimental setup with an optical spark-ignition engine was devised to investigate the primary factors influencing cycle-to-cycle variations in H2ICEs by optimizing mixture homogeneity and focusing the study on flame interaction with in-cylinder aerodynamics. Simultaneous in-cylinder pressure, chemiluminescence, and PIV analyses were performed to characterize and compare the behaviors of ultra-lean hydrogen-air and propane-air flames, assessing flame development and cyclic variation features. Results indicate that the flame development of hydrogen and propane is significantly different. Propane exhibited increased in-cylinder pressure trace variability at lean and rich flammability limits with a high flame development difference between fast and slow cycles. In contrast, hydrogen-air mixtures under various equivalence ratios (from very-lean to ultra-lean) presented much more stable in-cylinder pressures. Furthermore, fast propane flames were mainly advected to the cylinder's symmetrical axes, while fast hydrogen flames started further away from the spark plug. Horizontal and vertical PIV measurements showed a single structure flow field over the studied conditions with mainly intensity variations over the measured cycles. Fast flames were associated with more intense tumble motion. The differences between fast and slow flames for hydrogen are attributed to the early flame development. However, after initial differences in flame development over several crank angles, there are similarities in all cycles, suggesting that the low variability in in-cylinder pressures is mostly due to the robustness of hydrogen flame ignition and its independence from in-cylinder flow small variations at the early development phase.

Cause-and-effect chain analysis of combustion cyclic variability in a spark-ignition engine using large-eddy simulation, Part I: From tumble compression to flame initiation

Understanding, modeling, and reducing the cycle-to-cycle variability (CCV) of combustion in internal combustion engines (ICE) is a critical challenge to design engines of high efficiency and low emissions. A high level of CCV may contribute to partial burn, misfire, and knock in extreme engine cycles, which affects engine performance and eventually damages the engine. The origins of CCV have been studied both experimentally and numerically, and the variability of in-cylinder aerodynamics is recognized as one of the most important sources of CCV. However, a detailed and quantitative explanation of how in-cylinder flow CCV is generated is not yet clear. The objective of the present study is to develop a methodology to localize inside the chamber of a spark-ignition engine (SIE) the origins of flow variabilities and to identify some driving mechanisms leading to combustion variabilities. Multi-cycle wall-modeled large-eddy simulations (LES) for the TU Darmstadt optical engine under fired conditions are performed using the CFD solver Converge 3.0. The evolution of organized large-scale structures and the small-scale turbulence of the in-cylinder flow are analyzed using a developed methodology that includes the empirical mode decomposition (EMD) method adapted for 2D and 3D flow fields, and a vortex identification tool Γ3p. The contributions of different parts of the flow to CCV are quantified. In Part I of this work, the LES framework is validated against experimental data, and CCV of large-scale structures is characterized at spark timing. In Part II, the overall flow development during compression and intake strokes are quantitatively analyzed, and links are built between different engine phases to establish the cause-and-effect chain. Other CCV factors, such as spray injection and exhaust gas recirculation, are not included in the current study. However, the developed methodology for in-cylinder flow analysis could be used in studies on other engine configurations to improve the development of engine designs.Novelty and significance statementIn this work, the cycle-to-cycle variability (CCV) of combustion in a spark ignition engine is investigated to give a deeper understanding of CCV generation. The present study focuses on CCV caused by the stochastic nature of internal turbulent flow structures. LES approach is chosen due to its ability to capture CCV. The LES methodology was validated in a motored case in Ding et al. (2023). In the present study, it is validated in a reactive case against experimental in-cylinder pressures and velocity fields.A first novelty is the application of EMD methods combined with topology-based techniques to reactive LES results to characterize flow structures of different scales in the three-dimensional domain and to quantify separately their impacts on combustion.A second novelty and important finding is that a link is established between the combustion speed and the tumble formation and destabilization near BDC.Throughout our analyses in Part I and II, starting from the spark-ignition timing and going back to the early intake phase, a cause-and-effect chain is finally established between the development of in-cylinder flow and the combustion variability.

Quantitative analysis of the relationship between charge motion and knocking combustion in spark-ignition natural-gas engines under critical knocking conditions

Increasing the in-cylinder flow intensity is considered to have the potential to restrain engine knock and improve thermal efficiency. However, the effect of charge motion on the knocking combustion in natural gas (NG) engines is unclear. Therefore, swirl, tumble, and squish motions of different intensities were organized by changing the shape of the combustion chamber and initializing the in-cylinder flow velocity. Subsequently, the effect of charge motion on the knocking combustion of a spark-ignition NG engine under critical knocking conditions was investigated via numerical simulations. The results indicate that the knock intensity (KI) first increases and then decreases with an increase in swirl and tumble intensities and linearly increases with an increase in squish intensity. Swirl had the smallest impact on KI, followed by tumble, whereas squish had the greatest impact. The differences in flame propagation resulted in differences in KI under different charge motion patterns. The obvious non-uniformity of the radial and axial flame propagation is a potential cause of the significant increase in KI. Therefore, while increasing the flow intensity, matching a suitable combustion chamber to make the flame propagation more uniform helps achieve anti-knock and rapid combustion. Improving the swirl and tumble intensities can help NG engines achieve anti-knock and rapid combustion, whereas improving the squish intensity does not. This study provides an important reference for determining whether increasing flow intensity can improve NG engines’ knock characteristics and combustion performance.

In-cylinder flow evolution in the horizontal plane of a motoring compression ignition engine

This study used the two-dimensional particle image velocimetry technique to quantify the in-cylinder horizontal plane velocity field evolution in a swirl-supported light-duty single-cylinder diesel engine. The data were acquired at a constant engine speed of 1600 revolutions per minute. For each case, the distance of the laser sheet from the fire deck was varied (z = 5, 10, and 20 mm) to investigate the axial variations in the flow field during the flow evolution in the compression stroke. A vortex identification algorithm was used to detect the swirl center and its deviation from the rigid body rotation. A Bessel fit was obtained using the experimental data. The result revealed that the in-cylinder flow was not axisymmetric. The swirl center approached the geometrical center as the piston approached the top dead center. The flow evolved at the farthest plane from the fire deck. The axial diffusion of angular momentum resulted in the formation of the swirl flow structure in the plane closer to the fire deck. Angular momentum analysis of a simplified geometry has been presented to explain the swirl amplification. The estimated results were compared with the experimental results to show the momentum stratification in the engine cylinder later in the compression stroke.

Characterization of near-wall flow and in-cylinder free-stream turbulence during the intake phase of a direct-injection engine: A flow bench study

This study investigates the characteristics of the near-wall flow and pressure-induced wall-influenced turbulence inside a direct-injection engine during the intake phase by utilizing a wall-resolved Large Eddy Simulation. An engine flow bench operating under stationary conditions is used to reduce the complexity of engine flows and to shed light on the in-cylinder flow during the intake phase. The investigation focuses on in-cylinder turbulence on symmetry (SP) and valve planes (VP), particularly emphasizing the flow close to the intake valves and the impingement region on the liner wall. An experimental dataset acquired through 2D high-speed particle image velocimetry (PIV) facilitates validation of the simulation of both time-averaged velocity field and associated turbulence structure. Turbulence anisotropy analysis shows that, in complex wall-bounded flows, the IC engine-related in-cylinder turbulence structure is characterized by the anisotropic states that cope with both axisymmetric expansion and contraction straining events. Due to the strong influence of the intake flow, the turbulence anisotropy level, expressed in terms of the two-componentality parameter F, retains its value of less than 0.85, and no turbulence occurs in accordance with the fully three-component isotropic state, characterized by the F-parameter value of 1. In the vicinity of the intake valve near wall area, the turbulence anisotropy state transitions from that resembling a canonical channel flow characteristic to curvature-induced acceleration after impingement, which aligns with axisymmetric expansion in the anisotropy invariant map. The presence of intake flow sustains turbulence anisotropy. Limitations in applying wall-function treatments for near-wall flow in engine contexts due to simplifications, such as the zero-pressure gradient assumption, are highlighted. In this respect, the interplay of non-equilibrium effects in the near-wall region needed to be further explored. Accordingly, analysis of the viscosity-affected layer near the intake valve and the flow impingement area of the liner wall underscores the importance of the wall-tangential pressure gradient in shaping the near-wall behavior. A budget analysis of the turbulent kinetic energy equation at the intake valve emphasizes the dominance of fluctuating pressure-induced diffusive transport in turbulence generation.

Influence of baffle height on the in-cylinder mixing and performance characteristics of a gasoline direct injection engine - a CFD investigation

In modern GDI engines, in-cylinder flows play a crucial role in the performance and emission characteristics. Adding baffles to the cylinder head is one way to enhance these flows, as baffle effectiveness is linked to its height. This study aims to investigate the influence of baffle height on the in-cylinder flows and its impact on the performance and emission characteristics of a four-stroke, four-valve, spray-guided GDI engine using computational fluid dynamics (CFD) analysis. The maximum baffle height is determined based on the geometry of the engine. Throughout the analysis, the engine operates at a constant speed of 1000 rev/min., under part-load conditions. From the results, the baffle heights of 2 mm and 3 mm significantly improve mixture stratification compared to the base engine. Furthermore, the indicated mean effective pressure (IMEP) for the engine with 2 mm and 3 mm baffle height improves by about 3% and 4%, respectively, compared to the base engine. In addition, with the increase in the baffle height, there is a significant reduction in the HC emissions with a marginal increase in NOX emissions.

Analysis of In-Cylinder Flow in a Small-Bore Spark-Ignition Engine Using Computational Fluid Dynamics Simulations and Zero-Dimensional-Based Modeling

Abstract The evolution of in-cylinder flow involves large- and small-scale structures during the intake and compression strokes, significantly influencing the fuel–air mixing and combustion processes. Extensive research has been conducted to investigate the flow evolution in medium- to large-sized engines using laser-based diagnostic methods, computational fluid dynamics (CFD) simulations, and zero-dimensional (0D) based modeling. In the present study, we provide a detailed analysis of the evolution of flow fields in a small-bore spark ignition (SI) engine with a displacement volume of 110 cm3. This analysis employs a unique methodology, where CFD simulation is performed and validated using measured particle image velocimetry (PIV) data. Subsequently, the validated CFD results are utilized to develop and validate a 0D-based model as it is computationally more efficient. The validated CFD simulation and 0D-based model are then used to evaluate the quantified strength of the flow by calculating the tumble ratio and turbulent kinetic energy (TKE). The streamlines and velocity vectors of the flow fields obtained from CFD simulations are utilized to explain the evolution of these parameters during intake and compression strokes. The study is further extended to analyze the effect of engine speed on the evolution of flow fields. With an increase in engine speed, relatively higher values of tumble ratio and TKE at the end of the compression stroke are observed, which is expected to improve the fuel–air mixing and combustion efficiency.

Spray formation and spray-wall-flow interaction in a gasoline direct-injection (GDI) engine under early-injection conditions: A flow bench study

The work presented in this study aims to understand the spray-wall-flow interaction within a gasoline direct-injection (GDI) engine flow bench under simulated early-injection conditions. The Engine Combustion Network (ECN) Spray G injector is installed in the Darmstadt optically accessible engine flow bench. Under simulated early-injection conditions, the formation of a multi-hole spray and the interaction with characteristic intake flows, such as the intake jet and central tumble flow, are extensively discussed. By reducing the complexity in the number of variables inherent in engine flow and whole-engine simulation, an engine flow bench operating under various mass flow rates is applied in this study. The numerical simulation is carried out using Large Eddy Simulation (LES) under the Eulerian-Lagrangian framework for spray simulation. Experimental data, acquired through particle image velocimetry (PIV) measurements, provides 2-D flow fields on both the central tumble and valve planes, facilitating the validation of in-cylinder flow fields. Furthermore, experimental data obtained through Mie scattering is utilized to investigate spray formation and evolution within the GDI engine, providing the liquid penetration length and liquid spray angle. Comparison between the numerical and experimental data demonstrates several agreements. Moreover, the variation of different spray plumes under different mass flow rates is observed in the case of both experimental and numerical data. Increasing the mass flow rate distorts the overall plume shape and shifts it away from the intake port. This phenomenon is examined by extracting the liquid volume fraction and vapor fields of each plume. Spray plumes encounter different convective disturbances and evaporation due to their local characteristic in-cylinder flow. Furthermore, spray-wall-flow interaction and wall film deposits are observed during the injection. Lastly, the influence of the spray-induced turbulence is analyzed under different mass flow rates.

Effect of multi-injection strategy on characteristics of methanol-fueled direct injection spark ignition engine

Present paper numerically investigates the effect of injection strategy and start of injection (SOI) timing on in-cylinder flow, air–fuel mixing, fuel distribution near spark plug, engine performance, and exhaust emissions for highly stratified methanol-fueled, multi-injection, direct injection spark ignition engine having high compression ratio. SOI is kept constant at −23° crank angle (CA) after top dead center (ATDC) with a spark timing (ST) of −2° crank angle (CA) ATDC. Mass of fuel is divided into pilot and main injection ports having pilot to main fuel injection mass ratio of 1:1, 1:2, and 1:3 at 0° and 2° dwell times between main and pilot injections. As the quantity of fuel in main injection increases, pressure rise rate increases, which results in higher in-cylinder pressure and higher rate of burning that gives higher apparent heat release. Due to higher peak pressure rise rate and faster burning in the case of 2° crank angle (CA) dwell time, shorter combustion duration is achieved compared to 0° crank angle (CA) dwell time. In the case of multi-injection, faster burning rate enhanced in-cylinder temperature; therefore, nitrogen oxides (NOx) emissions are higher. Pilot to main fuel mass ratio of 1:3 has resulted highest indicated thermal efficiency, lowest specific fuel consumption, lowest soot, and carbon monoxide (CO) emissions.

0/1 and 3 - dimensional cold flow analysis of a diesel engine: A case study

Gas motion in the engine cylinder plays a critical role in diesel engines' air-fuel mixing and combustion processes. Moreover, it influences the engine's performance, emissions, and heat transfer. The intake air motion regulates the main phases of the flow in the cylinder, which is characterized by swirl, squish, and turbulence. Inducing swirl and tumble in the intake process provides high turbulence levels at ignition, resulting in more effective flame speeds and better combustion for lean air-fuel ratios or with EGR. Computational fluid dynamics (CFD) software is used to enhance in-cylinder flow characteristics. In this study, a cold flow simulation of a naturally aspirated, direct injection diesel engine was conducted with different piston bowls using AVL Fire M R2022.2. Swirl, tumble, and TKE parameters were investigated to make a detailed analysis of the in-cylinder flow for the relevant engine. Contrary to the expectation, the swirl ratio for the Piston B configuration is less than that for the original piston configuration. It causes a decrement of swirl ratio compared to the initial piston geometry and the maximum decrement is about 19%. In both cases, the second peak of TKE corresponding to the reverse-squish is around at the -30 CA and the difference between the curves is about ±8%.

Computational Fluid Dynamics of Four Stroke In-Cylinder Charge Behavior at Distinct Valve Lift Opening Clearance in Spark Ignition Reciprocating Internal Combustion Renault Engine

In-cylinder flow dynamics in internal combustion Renault engine is complex, expensive and difficult to compute experimentally. The present study attempts to emulate the in-cylinder charge behaviour at distinct valve lift opening clearance in four stroke spark ignition internal combustion engine using computational fluid dynamics. Considering the complexity of the geometry and in-cylinder fluid motion, governing equations for unsteady, three dimensional, compressible turbulent flow were computed with continuity equations (conservation of mass), Navier-Stokes equations (conservation of momentum) and RNG k-ε turbulence model. Assumed to be an inline spark ignition (SI) operating on a four stroke cycle, the engine was modelled with SolidWorks 2019 version while the in-cylinder charge behaviour was simulated using ANSYS Fluent 14.5. Increase in cylinder temperature enhanced the thermal properties of air-fuel mixture during combustion. Increase in valve lift opening clearance led to more charge quantity being ingested through the intake valve opening into the cylinder, thereby causing increase in temperature of in-cylinder charge as well as significant improvement in the volumetric and mechanical efficiency of the cycle. It was also observed that the rate of heat retention in the cylinder may be optimum at lower valve lift which may be characterised by minor or zero loses, while significantly high cylinder charge temperature may be prone to reduction of the intake charge density. Based on Particle Image Velocimetry (PIV), in-cylinder velocity vectors, vorticity magnitudes and distributions of turbulence kinetic energy (TKE) increased with increasing valve lift opening clearance, thereby, improving combustion efficiency, increasing torque and power output for effective engine performance.

Research on Scavenging Flow Dynamics of Marine Two-Stroke Engines With a CFD-Derived Quasi-Dimensional Model

The complex in-cylinder gas state and flow significantly affect fuel-air mixture, combustion efficiency and emissions. However, the scavenging CFD model of large bore marine engines in digital twin system requires substantial computational resources. Therefore, a fast-run phenomenological model of the scavenging process is built in this study. The model simulates the thermal states and flow dynamics of inlet and exhaust gas based on the energy and momentum conservation principles, considering the ideal in-cylinder swirl velocity profile with effects of air mass loss, wall friction, and swirl shear. The model’s accuracy is confirmed by comparing it with CFD simulations on different engines, showing an average relative error less than 3.5%. It also analyzes the impacts of intake pressure. The model provides accurate boundary conditions for subsequent fuel spray, combustion, and emission simulations and can be combined with these models in the future, thereby applied to engine design, diagnostics, and control.

Swirl-induced motion prediction with physics-guided machine learning utilizing spatiotemporal flow field structure

PurposeHigher energy conversion efficiency of internal combustion engine can be achieved with optimal control of unsteady in-cylinder flow fields inside a direct-injection (DI) engine. However, it remains a daunting task to predict the nonlinear and transient in-cylinder flow motion because they are highly complex which change both in space and time. Recently, machine learning methods have demonstrated great promises to infer relatively simple temporal flow field development. This paper aims to feature a physics-guided machine learning approach to realize high accuracy and generalization prediction for complex swirl-induced flow field motions.Design/methodology/approachTo achieve high-fidelity time-series prediction of unsteady engine flow fields, this work features an automated machine learning framework with the following objectives: (1) The spatiotemporal physical constraint of the flow field structure is transferred to machine learning structure. (2) The ML inputs and targets are efficiently designed that ensure high model convergence with limited sets of experiments. (3) The prediction results are optimized by ensemble learning mechanism within the automated machine learning framework.FindingsThe proposed data-driven framework is proven effective in different time periods and different extent of unsteadiness of the flow dynamics, and the predicted flow fields are highly similar to the target field under various complex flow patterns. Among the described framework designs, the utilization of spatial flow field structure is the featured improvement to the time-series flow field prediction process.Originality/valueThe proposed flow field prediction framework could be generalized to different crank angle periods, cycles and swirl ratio conditions, which could greatly promote real-time flow control and reduce experiments on in-cylinder flow field measurement and diagnostics.

Effects of injection timing on mixture formation, combustion, and emission characteristics in a n-butanol direct injection spark ignition engine

The impact of injection timings on the in-cylinder flow field intensity, mixture formation process, combustion, and emission characteristics are analyzed. The range of injection timing is set between −320°CA and −260°CA. The research results indicate: Delaying the injection timing causes the spray plumes increasingly align with the tumble direction, consequently enhancing the in-cylinder tumble intensity and the turbulence near the spark timing. The mixture homogeneity is influenced by both the formation time and the in-cylinder flow field intensity. Moreover, combustion performance is linked to the homogeneity of the mixture and turbulence intensity near the spark timing. Despite the fact that postponed injection may augment both turbulence intensity and the equivalent ratio around the spark plug at the moment of spark, achieving higher explosion pressure, but the higher viscosity of n-butanol impedes droplet break-up, and its low volatility decelerates the evaporation rate, leading to poorer mixture homogeneity, reduced BMEP, and increased CO and soot emissions. However, at −300°CA start of injection, the extended mixing time optimizes the fuel-air mixture homogeneity, thus improving combustion quality and reducing CO, soot, and HC emissions, albeit leading to increased NOx emissions.

Statistical analysis of flow field variations via independent component analysis

The link between in-cylinder flow and subsequent combustion in a single-cylinder gasoline spark-ignition engine is analyzed via independent component analysis (ICA). Experimentally, the two in-plane components of the velocity are measured in the central cylinder plane by high-speed particle image velocimetry (PIV) with the engine running slightly lean at 1500 rpm in skip-fired mode. In 213 cycles, measurements are made during the late compression stroke before ignition with approximately 1° crank-angle temporal resolution. ICA then decomposes the set of 213 flow fields at each time step, yielding a set of “source” flow patterns—the independent components (IC). The temporal coherence between the ICs is then examined in a persistence analysis, comparing each IC with the one from the previous time step starting at ignition timing and going backwards in time. The results show which ICs persist how long throughout the compression stroke. To investigate the link between the ICs and combustion, the crank angle at which 10% of the fuel are burned (CA10) in each cycle is correlated with the extent to which a given IC can be found in each flow field. The most persistent IC can be traced over more than half of the 70 degrees crank angle over which images were acquired. The IC that correlates best with CA10 visually more resembles some of the flow features found in conditional averaging of fast-burning versus slow-burning cycles.