Marine two-stroke diesel engine rotational flow combustion manifestation model and self-calibration optimization method

Abstract

The invention relates to the field of combustion modeling of internal combustion engines, and discloses a marine two-stroke diesel engine rotational flow combustion manikin and a self-calibration optimization method.The construction method of the model comprises the steps that spraying areas are divided in a layered mode based on the axial direction and the radial direction of the spraying areas, and the spraying penetration distance and the air entrainment mass are dynamically calculated; establishing an interaction model of spray and rotational flow, and calculating spray behaviors through coupling of spray deflection speed and rotational flow speed; performing multi-zone dynamic analysis on the combustion process, and calculating spray crushing, gasification and combustion characteristics through mass conservation and momentum conservation; the dynamic process of premixed combustion and diffusion combustion is simulated, and the heat release rate is calculated by combining superposition of the fuel evaporation capacity and the air entrainment capacity; a heat release correction coefficient based on an air cylinder compression curve is introduced according to the wall impingement effect, and the combustion characteristic of a wall impingement area is corrected. According to the method, the complex combustion process can be accurately represented, the precision and adaptability of the model are remarkably improved, and theoretical support is provided for combustion optimization design and control of the marine diesel engine.

Description

Technical Field

[0001] This invention relates to the field of internal combustion engine combustion modeling technology, specifically to a phenomenological model of swirling combustion in marine two-stroke diesel engines and a self-calibration optimization method. Background Technology

[0002] Shipping plays a crucial role in global trade. Statistics show that approximately 90% of international trade goods are transported by sea, with annual ship fuel consumption reaching 280 million tons. To address the increasingly prominent environmental issues, the International Maritime Organization (IMO) adopted the "IMO Strategy for Reducing Greenhouse Gas Emissions from Ships 2023" at the MEPC80 meeting, explicitly proposing the goal of achieving net-zero greenhouse gas emissions from ships by 2050. Given that two-stroke diesel engines are the primary power source for ships, optimizing their combustion process to reduce brake ratio fuel consumption (BSFC) and harmful gas emissions has become a key research focus.

[0003] Swirling, as a characteristic of the rotational motion of fluid within the cylinder, significantly impacts the combustion performance of two-stroke diesel engines. Due to the characteristics of two-stroke engines—large injection volume and low engine speed—and the limited potential for increasing injection pressure due to injector technology, enhancing the intensity of in-cylinder swirling has become a key technological approach to improving fuel-air mixture quality. Existing research primarily employs experimental methods and three-dimensional numerical simulations to study swirling characteristics. In experimental studies, the effectiveness of a novel swirling loop scavenging (SLS) configuration in enhancing swirling intensity was verified using the tracer gas method. Laser Doppler velocimetry (LDA) was used to measure the axial velocity of the swirling flow and analyze the effects of intake port angle and valve timing on the swirling distribution. Simultaneously, particle image velocimetry (PIV) was used to measure the flow field characteristics in different planes within a single-cylinder optical diesel engine, investigating the influence of parameters such as engine speed, intake port area, and spiral intake port opening area on the swirling flow. However, experimental research is limited by the test conditions and measurement methods, making it difficult to accurately measure the swirling characteristics under all operating conditions. Although the three-dimensional numerical simulation method can obtain detailed flow field information, it has inherent defects such as complex computational mesh division, cumbersome boundary condition setting, slow calculation speed and high computing power requirements, making it difficult to effectively support the overall engine optimization design and control system development.

[0004] Therefore, it is urgent to establish a phenomenological combustion model applicable to all operating conditions of two-stroke diesel engines, so as to provide a theoretical basis for the optimized design and control strategies of marine diesel engines. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a phenomenological model of swirling combustion in marine two-stroke diesel engines and a self-calibration optimization method. This solves the problems of inaccurate descriptions of spray-swirl coupling, combustion dynamics, and wall-impact effects in existing two-stroke diesel engine combustion modeling, which leads to insufficient accuracy and poor adaptability of the combustion model under operating conditions.

[0006] To achieve the above objectives, the present invention provides a method for constructing a phenomenological model of swirling combustion in a marine two-stroke diesel engine, characterized by comprising the following steps: The spray area is divided into axial and radial layers, where the axial layering is determined based on the time step, and the radial layering is based on the uniform division of the spray cone area into multiple layers. An interaction model between the spray and the swirling flow is established. Based on the relative relationship between the swirling flow velocity vector and the spray deflection velocity vector, the spray penetration distance and air entrainment mass are calculated. The combustion process is analyzed dynamically in multiple zones, assuming that the diesel mass is evenly distributed in each zone, and the dynamic characteristics of spray breakup, gasification and combustion are calculated by mass conservation and momentum conservation. The dynamic processes of premixed combustion and diffusion combustion during the simulation are combined with the fuel evaporation rate and air entrainment rate to calculate the heat release rate. To address the wall impact effect, a heat release correction coefficient is introduced through the cylinder compression curve to correct the combustion characteristics in the wall impact region.

[0007] This invention also provides a phenomenological model of swirling combustion in marine two-stroke diesel engines, the model comprising: The axial and radial layered structure of the spray area is used to dynamically calculate the spray penetration distance and air entrainment mass based on the layered spray area; The interaction model between spray and swirl calculates the spray penetration distance and fuel-air mixture ratio by combining swirl velocity and spray deflection velocity; A multi-zone phenomenological model characterizing the dynamic characteristics of premixed combustion and diffusion combustion during the combustion process is used to calculate the fuel evaporation and air entrainment in each spray zone through layered dynamic calculations, and then superimposes the overall heat release rate. A model to correct the impact of the wall impact effect on combustion characteristics is proposed. By introducing a heat release correction coefficient based on the cylinder compression curve, the combustion reaction rate and local heat release characteristics in the wall impact region are dynamically adjusted.

[0008] This invention also provides a self-calibration optimization method for a swirl combustion phenomenological model of a marine two-stroke diesel engine, the optimization method comprising: The calibration of combustion model parameters is transformed into a constrained multi-input single-output nonlinear optimization problem. The optimization based on the improved adaptive genetic algorithm includes the following steps: Construct an initial population and select individuals based on an elite retention strategy, setting crossover and mutation probabilities; Design a fitness function with engine output torque, fuel economy and emission performance as optimization objectives; During population evolution, the crossover and mutation probabilities are dynamically adjusted to enhance global search capabilities and avoid local convergence. The optimized model parameters are imported into the engine controller, and the engine operating parameters are adjusted through real-time data acquisition and closed-loop feedback to achieve dynamic optimization of the model.

[0009] This invention provides a phenomenological model of swirling combustion in marine two-stroke diesel engines and a self-calibration optimization method. It offers the following advantages: 1. This invention divides the spray area into axial and radial layers and combines a multi-zone phenomenological method to dynamically calculate the spray penetration distance and air entrainment mass, thereby accurately characterizing the dynamic behavior of the spray and the fuel-air mixing characteristics, providing higher precision for the simulation of the combustion process.

[0010] 2. This invention introduces a heat release rate correction coefficient based on the cylinder pressure curve to dynamically adjust the combustion reaction rate and local heat release characteristics of the wall impact region. It can comprehensively characterize the dual regulatory effect of airflow disturbance and fuel evaporation caused by spray wall impact, and effectively optimize the combustion characteristics of the wall impact region.

[0011] 3. This invention reduces the complexity of modeling by simplifying assumptions such as uniform injection rate and fixed spray cone angle. At the same time, it combines genetic algorithm to optimize model parameters, ensuring that the combustion model can adapt to different operating conditions and achieve a good balance between computational efficiency and adaptability.

[0012] 4. This invention optimizes the heat release characteristics within the combustion chamber by performing zoned calculations of the premixed combustion and diffusion combustion processes, combined with fuel evaporation rate and air entrainment efficiency, and precisely superimposing these calculations. This not only improves the engine's thermal efficiency but also significantly reduces harmful gas emissions, providing dual protection for both the combustion performance and environmental performance of diesel engines. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the construction method of the model of the present invention; Figure 2 This is a schematic diagram of the spray area division based on the multi-zone phenomenological method of the present invention; Figure 3 This is a schematic diagram of spray deflection under a swirling flow field according to the present invention; Figure 4 This is a schematic diagram of the adaptive calibration-optimization framework for ship main engines of the present invention. Detailed Implementation

[0014] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see the appendix Figure 1 - Appendix Figure 3 This invention provides a method for constructing a phenomenological model of swirling combustion in marine two-stroke diesel engines. This method, through the analysis and modeling of multiple physical phenomena such as spray, swirling, and wall-impact effects in the combustion process, can achieve accurate characterization of the combustion process and is suitable for optimizing the combustion performance of marine diesel engines.

[0016] like Figure 1 As shown, the method for constructing the phenomenological model of swirling combustion in a marine two-stroke diesel engine may include the following steps: S1. The spray area is divided into axial and radial layers based on the spray area; S2. Establish an interaction model between spray and swirl; S3. Perform multi-zone dynamic analysis of the combustion process; S4. Simulate the dynamic processes of premixed combustion and diffusion combustion during combustion; S5. To address the wall-impact effect, a heat release correction coefficient is introduced through the cylinder compression curve.

[0017] For step S1, as Figure 2 As shown in this embodiment, a method for dividing the spray region based on axial and radial layering is proposed to address the characteristics of the spray region during the combustion process of a marine two-stroke diesel engine. Layering the spray region is the foundation of the combustion model of this invention. Through layering, the spray penetration distance, air entrainment, and fuel distribution during the combustion process can be more accurately and dynamically characterized.

[0018] As an alternative, axial stratification is determined based on time steps, with the dynamic changes in spray penetration distance within each time step used to determine the stratification boundaries. Radial stratification, on the other hand, is based on the spray cone region, dividing the spray area into multiple radially symmetrical layers according to the spray cone angle and spray conditions, and numbering them with the central axis as the reference.

[0019] In one possible implementation, the axial layering is based on the dynamic characteristics of the spray penetration distance. Spray penetration distance The calculation formula can be divided into two cases: before crushing and after crushing.

[0020] in, and These are the characteristic constants of the spraying stage. The specific formula for calculating the spray abrasion time is as follows:

[0021] In some embodiments, and The calculation formula is as follows:

[0022]

[0023] in, The temperature of the gas inside the cylinder; It is the difference between the fuel injection pressure and the engine cylinder pressure; The density of the working fluid inside the cylinder; This is the density of air at normal temperature and pressure. This refers to the density of diesel fuel.

[0024] It should be noted that the dynamic calculation of the spray penetration distance can not only determine the axial layer boundary within the time step, but also characterize the expansion range of the spray region in the combustion process simulation.

[0025] In some embodiments, radial stratification is based on the geometric characteristics of the spray cone region. Specifically, radial stratification uses the spray center axis as a symmetry reference to uniformly divide the spray cone region into multiple radial layers, such as layer 0, layer ±1, layer ±2, etc. The width of each layer is determined by the spray cone angle. The initial conditions for injection are determined, satisfying the following geometric relationships:

[0026] in, For the first The radius of the layer, The radius of the spray center axis, This represents the radial layer width.

[0027] In one possible implementation, the width of the radial spray layer can be dynamically adjusted based on the spray cone angle and cylinder radius to better adapt to the fuel distribution characteristics within the spray penetration area. For example, the radial stratification within the spray area can be adapted based on air entrainment efficiency and fuel vaporization rate, which can improve the resolution and accuracy of combustion simulation.

[0028] Understandably, the stratification of the spray zone needs to be considered in conjunction with the actual kinematic characteristics of diesel spray. Within the spray zone, the distribution of diesel mass is assumed to be equally distributed across each layer. The rationality of this assumption is based on the flow stability of the diesel injection process and the symmetry of the nozzle design.

[0029] In this invention, the spray stratification model can be integrated with subsequent spray dynamic calculation models. For example, by dividing the spray cone region into layers, it can directly support the zonal calculation of spray penetration distance, air entrainment, and fuel vaporization. It should be noted that this stratified structure can effectively adapt to the combustion simulation requirements under different operating conditions.

[0030] In some embodiments, after the spray area is layered, the fuel entrainment air volume in each layer can be calculated according to the following formula:

[0031] in, For the air quality of a single stratified area, For diesel quality, The air entrainment coefficient. For the spray outlet velocity, This represents the average velocity within the current stratified region.

[0032] Specifically, air entrainment efficiency is closely related to the contact area within the spray zone, the air density within the cylinder, and the local velocity field. By dividing the area into layers, the mixing ratio of diesel and air can be calculated separately within each layer, providing partitioned data for the dynamic calculation of the subsequent combustion process.

[0033] For example, the spray zone stratification can also be dynamically adjusted based on the engine's operating conditions. For instance, under low load conditions, the spray penetration distance increases at a lower rate, allowing for a reduction in the time step to improve simulation accuracy. Under high load conditions, the width of the radial stratification can be dynamically adjusted according to a higher spray penetration rate, thus ensuring a balance between simulation accuracy and computational efficiency.

[0034] It should be noted that the layering method for the spray area in this invention is applicable to different types of two-stroke diesel engines. Regardless of changes in nozzle design, cylinder geometry, or injection pressure, it can be flexibly adapted based on the method of this embodiment.

[0035] It is understandable that the spray area layering method in this embodiment can provide a reliable data foundation for the subsequent combustion process model construction and significantly improve the accuracy and adaptability of the model under complex combustion conditions.

[0036] For step S2, in this embodiment, in order to accurately describe the interaction process between the spray and the swirling flow, an interaction model based on the swirling flow velocity vector and the spray deflection velocity vector is established. By analyzing the dynamic coupling relationship between the swirling flow and the spray, this model can accurately calculate the spray penetration distance and air entrainment mass, providing important input parameters for subsequent combustion modeling.

[0037] As an option, such as Figure 3 As shown, the direction of the spray deflection velocity vector is the same as the direction of the swirling velocity vector, and its magnitude is determined by the airflow velocity and fuel injection momentum in the spray area. The actual spray velocity vector is the superposition of the spray velocity vector under swirlless conditions and the spray deflection velocity vector. Specifically, the spray deflection velocity can be calculated using the air entrainment characteristics, swirling intensity, and initial momentum of the spray area.

[0038] In one possible implementation, the calculation of the actual spray velocity vector is based on the superposition of the spray deflection velocity and the undisturbed spray velocity, as described below:

[0039] in, This represents the actual velocity of the spray under swirling action. For non-swirling spray velocity, This is the deflection velocity vector.

[0040] It should be noted that the magnitude of the spray deflection velocity is influenced by the air density, swirling velocity field, and momentum distribution within the spray region. For example, the effects of the tangential and axial velocities in the swirling velocity field on spray deflection dynamically change with the position of the spray layer.

[0041] In some embodiments, the dynamic variation of the spray penetration distance is controlled by the coupling effect of the spray velocity vector and the swirl intensity, specifically through momentum conservation calculations. The spray penetration distance is determined by the following equation:

[0042] in, The mass of air drawn in. The tangential velocity of the swirl is... and These are air quality and spray quality, respectively. This represents the spray deflection speed.

[0043] It should be noted that air mass flow rate Calculated using the following formula:

[0044] in, air density, This refers to the contact area between air and fuel within the spray zone. Let be the swirl velocity. Through the above calculations, the dynamic characteristics of the spray penetration distance as a function of the swirl velocity field can be clearly defined.

[0045] Specifically, the calculation of air entrainment mass is directly related to the air density within the spray area, the spray contact area, and the spray velocity. In this invention, the calculation method and parameter relationships of air entrainment mass have been described in detail in step S1, combining dynamic calculations based on air density, spray area, and spray velocity to ensure an accurate description of the spray and swirling effect.

[0046] As an alternative, in the calculation of air entrainment, air density is considered. The changes can be dynamically adjusted in conjunction with the in-cylinder compression process to adapt to the coupling characteristics of spray and swirl under different operating conditions. For example, under high load conditions, the swirl intensity is greater and the entrained air volume increases, while under low load conditions, the swirl has less impact on spray deflection and the entrained air volume decreases.

[0047] It should be noted that the establishment of the interaction model between the spray and the swirl provides reliable dynamic input parameters for the subsequent fuel-air mixing process. Through precise calculations of the spray penetration distance and air entrainment mass, the characteristics of spray breakup, vaporization, and combustion during the combustion process can be better simulated.

[0048] In some embodiments, the dynamic adjustment of the spray penetration distance also incorporates the non-uniformity of the swirling velocity field within the cylinder, specifically by adjusting for the gradient changes in swirling intensity near the cylinder wall and in the central region. This dynamic adjustment can further improve the accuracy of the spray-swirling interaction model.

[0049] It is understandable that the interaction model between spray and swirl is not only applicable to conditions with a large spray penetration distance, but also to conditions with a small spray cone angle, by dynamically adjusting the spray deflection velocity vector to adapt to different characteristics of the spray distribution.

[0050] For step S3, in this embodiment, in order to accurately describe the complex dynamic characteristics of the combustion process of a marine two-stroke diesel engine, a modeling method based on multi-zone dynamic analysis is proposed. By assuming that the diesel mass is evenly distributed in each zone, and combining the principles of mass conservation and momentum conservation, the dynamic characteristics of spray breakup, fuel vaporization, and combustion processes are calculated, providing key inputs for the subsequent construction of the combustion model.

[0051] Alternatively, in multi-zone dynamic analysis, a uniform distribution of diesel fuel mass is assumed within the spray strata. This assumption strikes a balance between computational complexity and accuracy while meeting the actual distribution characteristics required by the fuel injection process. Specifically, the uniformity of diesel fuel mass distribution is based on the axisymmetric characteristics of the spray and the stability assumptions of the nozzle design.

[0052] It should be noted that the dynamic analysis of spray breakup is the core step in multi-zone dynamic analysis. The spray breakup process includes a steady-state motion stage and a breakup stage. In the steady-state motion stage, the spray penetration distance increases linearly with time, while the breakup stage exhibits a nonlinear deceleration process influenced by surface tension and air resistance. The specific expression for the spray penetration distance has been explained in step S1. By combining spray time and dynamic stratification characteristics, the axial expansion behavior of the spray can be accurately described.

[0053] In one possible implementation, the critical point for spray breakup is determined by the breakup time. The process of fuel breaking down from a jet state into droplets is dynamically adjusted according to changes in injection pressure and air density in the cylinder. Specifically, higher injection pressure results in a shorter breaking-down time and smaller droplet diameters, thereby improving fuel evaporation efficiency.

[0054] In some embodiments, during the vaporization of the fuel droplets formed after breakup, the fuel evaporation rate is influenced by droplet surface characteristics, local temperature field, and cylinder pressure field. The fuel droplet evaporation process can be calculated using droplet evaporation kinetics equations, and its dynamic changes incorporate the thermodynamic characteristics of fuel stratification.

[0055] Specifically, the calculation of fuel evaporation takes into account the Sauter mean diameter (SMD) of droplets within the spray region and the local temperature field. It should be noted that the fuel evaporation process is not only an important input parameter for the combustion model, but also determines the air-fuel ratio in the premixed combustion and diffusion combustion stages.

[0056] During the combustion stage, the combustion process of fuel is divided into two parts: premixed combustion and diffusion combustion. As an option, the heat release characteristics of premixed combustion are mainly constrained by the fuel evaporation rate and the air entrainment rate. Specifically, in the premixed combustion stage, when the fuel evaporation rate is higher, the combustion rate is faster and the heat release intensity is greater.

[0057] During the diffusion combustion stage, the incompletely evaporated fuel continues to mix and burn with the surrounding air. The reaction rate of diffusion combustion is mainly limited by the amount of unevaporated fuel and the amount of residual air. It should be noted that the heat release rate of the diffusion combustion stage is generally lower than that of premixed combustion, but the combustion duration is longer, which has a significant impact on the engine's emissions performance.

[0058] In one possible implementation, the characteristics of premixed combustion and diffusion combustion are calculated separately for each stratified region. The dynamic heat release characteristics of each stratum are calculated by combining air entrainment, fuel evaporation, and pressure changes within the combustion chamber. Finally, the heat release rates of all strata are superimposed to obtain the overall heat release curve of the combustion chamber.

[0059] For example, dynamic changes in the pressure and temperature fields within the combustion chamber can affect the distribution of combustion rates. Under high load conditions, the higher pressure and temperature within the cylinder can significantly improve fuel evaporation efficiency and combustion reaction rate, while under low load conditions, the proportion of diffusion combustion increases, and the combustion rate decreases.

[0060] It is understood that, through the multi-zone dynamic analysis of the present invention, spray breakup, fuel evaporation and combustion characteristics can be calculated separately in each spray stratification zone, providing higher accuracy for combustion simulation under complex working conditions.

[0061] For step S4, in this embodiment, in order to accurately simulate the dynamic characteristics of premixed combustion and diffusion combustion during the combustion process of a marine two-stroke diesel engine, a dynamic modeling method based on fuel evaporation and air entrainment is adopted. By modeling the two combustion processes in separate zones and superimposing the heat release rate, the heat release characteristics inside the combustion chamber can be characterized more accurately.

[0062] As an alternative, simulations of the premixed combustion stage primarily focus on the mixing ratio of evaporating fuel to air and its impact on the combustion reaction rate. Specifically, the reaction rate of premixed combustion is determined by both the amount of fuel evaporated and the ignition delay time. The ignition delay time can be calculated using the Arrhenius equation:

[0063] in, To delay ignition time, For calibration constant, and These are the cylinder pressure and temperature, respectively. The equivalence ratio of the gas mixture. As the apparent activation energy, and As an experience index, in some embodiments, , .

[0064] It should be noted that during the premixed combustion stage, the amount of fuel evaporated is closely related to the physical properties of the fuel particles, the amount of air entrainment, and the temperature field within the cylinder. For example, smaller droplet sizes and higher local temperatures can significantly increase the fuel evaporation rate, thereby accelerating the premixed combustion reaction process.

[0065] In one possible implementation, diffusion combustion is simulated based on the continuous mixing characteristics of unevaporated fuel and residual air. Specifically, the reaction rate of diffusion combustion is limited by the fuel diffusion rate and the local air volume, and its exothermic characteristics are described by the following formula:

[0066] in, The amount of diesel fuel consumed for diffusion combustion. This represents the total mass of diesel fuel in the spray zone. For the quality of evaporated diesel fuel, To ensure the combustion of the consumed diesel fuel, The stoichiometric air-fuel ratio is the chemical equivalent of diesel fuel.

[0067] It should be noted that the heat release rate of diffusion combustion is usually lower than that of premixed combustion, but its duration is longer, and it plays an important role in the uniformity of the overall temperature field in the combustion chamber.

[0068] In some embodiments, to more accurately simulate the dynamic evolution of the two combustion modes during combustion, the calculations for premixed combustion and diffusion combustion are performed independently within stratified regions. By combining the air entrainment amount, evaporated fuel amount, and pressure changes within the combustion chamber of the spray zone, the premixed combustion rate and diffusion combustion rate of each zone can be calculated separately.

[0069] Alternatively, the zoned calculation results of premixed combustion and diffusion combustion are superimposed to simulate the overall heat release characteristics of the combustion chamber. Specifically, the overall heat release rate... The calculation is as follows:

[0070] in, This represents the total number of spray zones. Because of the low calorific value of fuel oil, For time step.

[0071] It should be noted that the dynamic change of the overall heat release rate can effectively reflect the complex characteristics of the combustion process, and at the same time provide key inputs for the calculation of the temperature field and pressure field in the combustion chamber.

[0072] For example, under high-load conditions, the fuel injection quantity and in-cylinder air density are higher, premixed combustion dominates, and the heat release rate curve exhibits a concentrated peak and a steep rise. Under low-load conditions, however, the proportion of diffusion combustion increases, the heat release rate curve is flatter, and the combustion duration is prolonged.

[0073] It is understood that this invention can accurately calculate the heat release curve in the combustion chamber by dynamically simulating premixed combustion and diffusion combustion, and adapt to combustion characteristics under different loads and operating conditions by adjusting spray and fuel evaporation parameters.

[0074] For step S5, in this embodiment, in order to accurately characterize the effect of the spray wall impact effect on the combustion process of a marine two-stroke diesel engine, a method based on the cylinder compression curve to introduce an exothermic correction coefficient is proposed. By analyzing the interaction process between fuel and cylinder wall in the wall impact region and its regulatory effect on the combustion reaction, the combustion characteristics of the wall impact region are corrected, and this effect is comprehensively considered in the overall combustion model.

[0075] As an alternative, the wall-impact effect mainly includes two aspects: first, the airflow disturbance caused by the spray impinging on the wall may suppress the amount of air entrainment in a local area, thereby slowing down the combustion reaction; second, after the fuel comes into contact with the cylinder wall, the high temperature effect of the cylinder wall may accelerate fuel evaporation and increase the combustion rate. This dual regulation mechanism needs to be described through dynamic modeling.

[0076] Specifically, changes in the cylinder compression profile can reflect the pressure and temperature characteristics of the cylinder wall impact region. By combining the dynamic compression behavior of the gas inside the cylinder, the fuel evaporation rate and air entrainment changes in the impact region can be calculated. The fuel evaporation rate is not only related to the cylinder wall temperature but also directly related to the fuel distribution density after the spray impacts the cylinder wall.

[0077] In one possible implementation, the heat exothermic correction factor for the impact region is described by the following formula:

[0078] in, This is the wall collision correction factor. and These are the cylinder pressure and temperature, respectively. This is the distance from the point where the fuel hits the wall to the center of the cylinder.

[0079] It should be noted that the calculation of the wall impact correction factor needs to consider both the spray penetration distance and the stratified structure of the spray region. After the spray impacts the wall, the termination position and radial distribution of its penetration distance directly determine the fuel concentration distribution in the impact region. By combining this with dynamic modeling of spray stratification, the fuel distribution characteristics in the impact region can be determined more accurately.

[0080] In some embodiments, the effect of the wall impact on the combustion reaction rate can be corrected by the air entrainment efficiency within the combustion chamber. Specifically, the mixing efficiency of fuel and air decreases after the wall impact, but the fuel evaporation rate may increase significantly under the action of the high-temperature cylinder wall, thus creating a positive and negative offsetting effect on the combustion rate.

[0081] For example, the correction to the combustion reaction rate can be expressed by the following formula:

[0082] in, The mass of diesel fuel in the diffusion combustion zone within the impact wall area, The mass of diffusion combustion diesel fuel before correction. This is the wall collision correction factor.

[0083] It should be noted that the fuel evaporation process in the impact zone significantly affects the heat release rate of diffusion combustion. This can be addressed by dynamically adjusting the correction coefficient. This can effectively improve the accuracy of simulating the combustion characteristics of the impact zone.

[0084] In one possible implementation, the heat release correction result for the impact zone needs to be superimposed on the overall combustion chamber heat release curve. The corrected heat release rate is calculated as follows:

[0085] in, This is the corrected total heat release rate. The uncorrected heat release rate, This represents the heat release increment after the wall collision correction.

[0086] Understandably, the dynamic adjustment of fuel-air mixing characteristics and combustion rate in the impact zone can be accurately characterized by the introduction of a correction factor to reflect the dynamic heat release process under actual operating conditions. This is particularly true under high-load conditions, where the higher fuel concentration in the impact zone significantly impacts the overall heat release rate.

[0087] In some embodiments, the fuel distribution and diffusion combustion rate in the impact zone can be further corrected by combining the local temperature field and airflow disturbance characteristics within the combustion chamber. By combining the fuel impact concentration distribution and cylinder wall thermal conductivity characteristics, the parameter range of the impact correction coefficient can be dynamically adjusted.

[0088] It should be noted that the wall-impact effect correction method of the present invention is applicable to different types of diesel engine combustion chamber structures. Regardless of changes in spray penetration distance, nozzle injection angle, or cylinder wall material properties, the correction coefficient can be dynamically adjusted. To achieve an accurate characterization of the wall-collision effect.

[0089] This invention establishes a spray-swirl interaction model by dividing the spray region into axial and radial layers. Combining the principles of mass and momentum conservation, it performs multi-zone dynamic analysis of the spray breakup, vaporization, and combustion dynamics during the combustion process. Simultaneously, it simulates the dynamic processes of premixed combustion and diffusion combustion, calculates the heat release rate based on the superposition of fuel evaporation and air entrainment, and introduces a heat release correction coefficient based on the cylinder compression curve to dynamically correct the combustion characteristics in the wall-impact region. This method can accurately characterize the complex dynamic characteristics of the combustion process, providing theoretical support for the combustion optimization design of marine diesel engines, and exhibits high accuracy and good adaptability to operating conditions.

[0090] Accordingly, the present invention also provides a swirl combustion phenomenological model for marine two-stroke diesel engines formed by the above-described construction method. This model integrates the layered structure of the spray region, the interaction model between the spray and the swirl, the multi-zone combustion phenomenological model, and the wall impact effect correction model, which can accurately describe the multi-physical characteristics and dynamic evolution process of the combustion process.

[0091] In this embodiment, the model first dynamically calculates the spray penetration distance and air entrainment mass through the axial and radial layering structure of the spray region. The spray layering structure, combined with the time step and spray cone angle, can capture the dynamic expansion characteristics of the spray and the fuel distribution characteristics, ensuring the accurate calculation of the spray penetration distance and air entrainment.

[0092] As an alternative, the spray-swirl interaction model establishes a dynamic relationship between spray penetration distance and fuel-air mixing ratio by combining swirl velocity and spray deflection velocity vectors. The model dynamically calculates the spray velocity and swirl coupling characteristics of each layer on a layered basis, while simultaneously evaluating the impact of swirl on air entrainment efficiency.

[0093] In one possible implementation, a multi-zone combustion phenomenological model characterizes the dynamic properties of premixed combustion and diffusion combustion. This model calculates the fuel evaporation and air entrainment rates independently within each stratified zone, then superimposes the combustion reaction rates and heat release characteristics of each zone to obtain the overall heat release rate curve of the combustion chamber. The model can adapt to changes in dynamic combustion characteristics under different operating conditions.

[0094] It should be noted that, in order to more accurately characterize the impact of the spray-impact effect on the combustion process, the model dynamically adjusts the combustion reaction rate and local heat release characteristics in the impact region using a heat release correction coefficient based on the cylinder compression curve. This correction model combines the changes in fuel evaporation rate and air entrainment efficiency after fuel impact to dynamically compensate for the combustion characteristics in the impact region, and the results are superimposed on the overall combustion model.

[0095] For example, the combustion phenomenology model of the present invention is applicable to the simulation of combustion processes in marine two-stroke diesel engines under various operating conditions. Whether under low or high load conditions, the model can accurately describe the combustion process by dynamically adjusting spray stratification and combustion correction coefficients. Simultaneously, this model can provide a reliable theoretical basis for the optimized design and control strategies of diesel engines.

[0096] It is understandable that by integrating the spray layering structure, the interaction model between spray and swirl, the multi-zone combustion model, and the wall impact effect correction model, this invention can comprehensively characterize the multi-physical coupling characteristics of the combustion process of marine two-stroke diesel engines and significantly improve the accuracy and applicability of combustion simulation.

[0097] Please see the appendix Figure 4Accordingly, the present invention also provides a self-calibration optimization method for a swirl combustion phenomenological model of a marine two-stroke diesel engine. By transforming the parameter calibration of the combustion model into a multi-input single-output nonlinear optimization problem with constraints, and combining an improved adaptive genetic algorithm, the combustion model parameters are accurately optimized. Furthermore, by combining real-time data acquisition and closed-loop feedback, the engine operating parameters are dynamically optimized, thereby improving the adaptability and accuracy of the combustion model.

[0098] In this embodiment, the parameter calibration problem of the combustion model is first transformed into a nonlinear optimization problem. Specifically, the goal of this problem is to optimize the model parameters to meet the engine's performance requirements, including output torque, fuel economy, and emissions performance, while being constrained by the boundary conditions of the engine's operating conditions, such as injection quantity, injection time, and load range. Through the optimization process, the combustion model parameters reach global optimum while satisfying the constraints.

[0099] Alternatively, an improved adaptive genetic algorithm is used for optimization. Specifically, the optimization process first constructs an initial population and selects individuals with high fitness through an elite retention strategy to improve the initial search quality of the algorithm. The initial population setting can be based on historical running data or an empirical range of the combustion model to ensure that the optimization search can cover all possible parameter spaces.

[0100] In one possible implementation, the fitness function is designed with engine output torque, fuel economy, and emissions performance as the core optimization objectives, and the weight coefficients of each objective are dynamically adjusted according to actual operating conditions. For example, under high-load operating conditions, the weight of output torque can be increased, while under low-load or environmentally friendly conditions, the weights of fuel economy and emissions performance can be increased. This dynamic adjustment of the fitness function can flexibly adapt to different operating conditions, improving the practical applicability of the optimization results.

[0101] It should be noted that, to avoid the genetic algorithm getting trapped in local optima, the crossover and mutation probabilities are dynamically adjusted during the optimization process. Specifically, the crossover probability is dynamically adjusted as the population fitness changes. When the population fitness distribution is relatively uniform, the crossover probability is increased to enhance global search capabilities; when the fitness converges, the crossover probability is decreased to strengthen local search capabilities. The mutation probability, on the other hand, is randomly adjusted using a Gaussian distribution to enhance population diversity and further improve global search capabilities.

[0102] For example, the optimized combustion model parameters are imported into the engine controller through real-time data acquisition and a closed-loop feedback mechanism to dynamically adjust the engine's operating state. Real-time data acquisition includes monitoring operating parameters such as engine speed, load, fuel injection quantity, and ignition delay time, and dynamically adjusting the combustion model parameters based on this real-time data. Through the closed-loop feedback mechanism, the real-time optimized model parameters can continuously iterate according to the actual operating conditions of the engine, further improving the model's accuracy.

[0103] In one possible implementation, the closed-loop feedback mechanism can automatically correct for abnormal data in the engine's operating state. For example, when the engine output torque is low, the system can automatically optimize fuel injection parameters or ignition delay time to ensure stable engine output and meet performance requirements. Through this real-time adjustment strategy, the optimization method of this invention can maintain the dynamic adaptability of model parameters under complex and changing operating conditions.

[0104] It should be noted that the optimization method of this invention can be widely applied to combustion models of different types of marine two-stroke diesel engines. Regardless of changes in engine load, speed range, or injection characteristics, the combustion model parameters can be dynamically adjusted through optimization algorithms and real-time feedback mechanisms to ensure the applicability and accuracy of the model.

[0105] It is understood that the self-calibration optimization method of the present invention, by combining an improved genetic algorithm, dynamic adjustment of the fitness function, and real-time closed-loop feedback, can not only efficiently optimize combustion model parameters, but also achieve dynamic optimization and continuous updating of the model through coupling with the actual operating conditions of the engine, providing an efficient and intelligent technical means for optimizing the combustion performance of marine diesel engines.

[0106] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a phenomenological model of swirling combustion in a marine two-stroke diesel engine, characterized in that, Includes the following steps: The spray area is divided into axial and radial layers, where the axial layering is determined based on the time step, and the radial layering is based on the uniform division of the spray cone area into multiple layers. An interaction model between the spray and the swirling flow is established. Based on the relative relationship between the swirling flow velocity vector and the spray deflection velocity vector, the spray penetration distance and air entrainment mass are calculated. The combustion process is analyzed dynamically in multiple zones, assuming that the diesel mass is evenly distributed in each zone, and the dynamic characteristics of spray breakup, gasification and combustion are calculated by mass conservation and momentum conservation. The dynamic processes of premixed combustion and diffusion combustion during the simulation are combined with the fuel evaporation rate and air entrainment rate to calculate the heat release rate. To address the wall impact effect, a heat release correction coefficient is introduced through the cylinder compression curve to correct the combustion characteristics in the wall impact region.

2. The method for constructing the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 1, characterized in that, The step of dividing the spray area into axial and radial layers based on the spray area includes: In the axial direction, the spray area is divided according to the time step, based on the dynamic change characteristics of the spray penetration distance, to ensure that the mass distribution of the spray area in each time step satisfies the conservation relationship. In the radial direction, the spray cone region is uniformly divided into multiple layered structures along the spray center axis, including a central layer based on the spray center axis and symmetrically distributed outer layer regions. The spray direction and velocity of each layer region are dynamically adjusted according to the initial spray cone angle and swirling velocity. The mass of sprayed diesel fuel within each layer is assumed to be evenly distributed, and the air-fuel mixing ratio in each layer is determined by calculating the entrained air mass and spray penetration distance. By combining the swirling velocity field with the axial and radial stratification of the spray area, a dynamic distribution relationship between air entrainment and fuel mass within the spray area is established.

3. The method for constructing the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 1, characterized in that, The steps for establishing the interaction model between the spray and the swirling flow include: By analyzing the directional relationship and dynamic variation law between the swirling velocity vector and the spray velocity vector, the direction and magnitude of the spray deflection velocity under the action of swirling flow are determined. Based on the stratification characteristics of the spray area, the actual velocity vector of the spray at each stratification location is calculated. The velocity vector is the superposition result of the spray deflection velocity vector and the undisturbed jet velocity vector. Based on the dynamic motion state of the spray area, the spray penetration distance is determined according to the principle of momentum conservation. The change in the spray penetration distance is adjusted with the velocity and swirl intensity at the spray position. By combining the spray deflection velocity and the entrained air mass, the entrainment ratio of air to fuel in each layered spray area is calculated. The entrained air mass is related to the spray velocity, the area of ​​contact, and the local air density. Finally, a dynamic distribution model of air entrainment and a spray penetration distance characteristic model under the action of spray and swirl were established to describe the influence of the interaction between spray and swirl on the fuel-air mixing process.

4. The method for constructing the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 1, characterized in that, The steps for multi-zone dynamic analysis of the combustion process include: Assuming that the diesel mass is evenly distributed in each layer of the spray, a model of the distribution of fuel mass in time and space is established by combining the layer characteristics of the spray area. Based on the principle of mass conservation, the dynamic changes of diesel mass and air entrainment mass within the spray area are calculated. The diesel mass distribution is related to the evaporation rate after spray breakup and the air entrainment rate. Based on the principle of momentum conservation, and combining the velocity changes before and after spray breakup and the interaction between the spray and the swirling flow, the spray penetration distance and the fuel evaporation rate within the region are calculated. During the spray fragmentation stage, the interaction between the liquid fuel fragmentation process and the airflow in the combustion chamber is dynamically analyzed, and the fuel fragmentation rate is adjusted with the changes in spray penetration distance and injection pressure. During the vaporization stage, the rate of change of the mass of evaporated fuel is calculated by combining the physical properties of fuel particles during evaporation and the high-temperature environment inside the cylinder, and a stratified combustion characteristic model is established. During the combustion stage, premixed combustion and diffusion combustion are treated separately. The premixed combustion rate is limited by the amount of evaporated fuel and the amount of residual air, while the diffusion combustion rate is limited by the amount of unevaporated fuel and the amount of residual air. Finally, the dynamic heat release rate of the overall combustion is calculated by superimposing these factors.

5. The method for constructing the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 1, characterized in that, The steps of the dynamic process of premixed combustion and diffusion combustion in the simulated combustion process include: In the premixed combustion stage, the mixing ratio of fuel and air is determined based on the matching relationship between fuel evaporation rate and entrained air rate, and the premixed combustion rate is calculated based on the fuel evaporation rate and ignition delay time. During the diffusion combustion stage, for incompletely evaporated fuel, the diffusion combustion rate is calculated based on the matching relationship between fuel diffusion rate and residual air amount, and the characteristics of diffusion combustion are determined by combining the dynamic changes of local temperature field and airflow field in the combustion chamber. A zonal processing method was adopted to independently calculate the heat release characteristics of premixed combustion and diffusion combustion in each stratified spray area, and the stratified calculation results were superimposed to obtain the overall heat release rate of the spray area. In the calculation of heat release rate, the combustion reaction rate is simultaneously constrained by the amount of fuel evaporation, the amount of air entrainment, and the pressure change in the combustion chamber, and the ratio of premixed combustion to diffusion combustion is dynamically adjusted. Finally, the dynamic evolution of premixed combustion and diffusion combustion during the simulation process is carried out, and the overall heat release curve is corrected by combining the combustion characteristics at the fuel injection termination stage to ensure that the heat release characteristics in the combustion chamber are consistent with the actual operating conditions.

6. The method for constructing the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 1, characterized in that, The step of introducing a heat release correction coefficient into the cylinder compression curve to correct the combustion characteristics of the impact zone in response to the wall-impact effect includes: Based on the contact behavior between fuel and cylinder wall after spray impact, the dynamic relationship between fuel evaporation rate and cylinder wall temperature in the impact area is analyzed to determine the impact of impact on combustion rate. By analyzing the inhibitory effect of airflow disturbance caused by wall impact on local air entrainment and combustion reaction rate, the change in air entrainment in the wall impact area is calculated, and the air-fuel mixing ratio is dynamically adjusted in conjunction with the spray penetration distance. A heat release correction factor is introduced to characterize the combustion characteristics of the wall impact region. The correction factor is dynamically determined by the cylinder compression curve and spray distribution characteristics within the wall impact region. Within the impact zone, the combustion reaction rate is dynamically adjusted based on the combined effects of fuel evaporation rate, air entrainment, and cylinder wall temperature to calculate the local heat release rate of the impact zone. Finally, by superimposing the combustion heat release characteristics after the heat release correction of the impact zone, a corrected overall combustion heat release curve is generated to match the cylinder compression characteristics and actual working conditions.

7. A phenomenological model of swirling combustion in a marine two-stroke diesel engine, constructed by the method described in any one of claims 1-6, characterized in that, The model includes: The axial and radial layered structure of the spray area is used to dynamically calculate the spray penetration distance and air entrainment mass based on the layered spray area; The interaction model between spray and swirl calculates the spray penetration distance and fuel-air mixture ratio by combining swirl velocity and spray deflection velocity; A multi-zone phenomenological model characterizing the dynamic characteristics of premixed combustion and diffusion combustion during the combustion process is used to calculate the fuel evaporation and air entrainment in each spray zone through layered dynamic calculations, and then superimposes the overall heat release rate. A model to correct the impact of the wall impact effect on combustion characteristics is proposed. By introducing a heat release correction coefficient based on the cylinder compression curve, the combustion reaction rate and local heat release characteristics in the wall impact region are dynamically adjusted.

8. A self-calibration optimization method for a swirl combustion phenomenological model of a marine two-stroke diesel engine, applicable to the model as described in claim 7, characterized in that, The optimization method includes the following steps: The calibration of combustion model parameters is transformed into a constrained multi-input single-output nonlinear optimization problem. The optimization based on the improved adaptive genetic algorithm includes the following steps: Construct an initial population and select individuals based on an elite retention strategy, setting crossover and mutation probabilities; Design a fitness function with engine output torque, fuel economy and emission performance as optimization objectives; During population evolution, the crossover and mutation probabilities are dynamically adjusted to enhance global search capabilities and avoid local convergence. The optimized model parameters are imported into the engine controller, and the engine operating parameters are adjusted through real-time data acquisition and closed-loop feedback to achieve dynamic optimization of the model.

9. The self-calibration optimization method for the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 8, characterized in that, The fitness function dynamically adjusts the weight coefficients of the optimization targets based on the engine operating conditions. The optimization targets include output torque, fuel economy and emission performance, and the target performance to be optimized is determined according to the specific operating conditions.

10. The self-calibration optimization method for the swirl combustion phenomenological model of a marine two-stroke diesel engine according to claim 8, characterized in that, The real-time data acquisition and closed-loop feedback include real-time monitoring of operating parameters such as engine speed, load, fuel injection quantity and ignition delay time, and dynamic optimization of engine operating status based on model parameters.

Images