Optimization Of Internal Combustion Engines Research Articles

Experimental and numerical fatigue evaluation of lightweight crankshafts

The development of lightweight components in the automotive industry is an important aspect in designing of products aiming to reducing fuel consumption, optimal use of material resources and sustainability. The optimization of internal combustion engines (ICE) components can be performed for example by mass subtraction procedure and use of different materials. Optimizing crankshafts is challenging due to the applied bending and torsion loads that increase the stress concentrations in critical regions. In this context, the objective of the present work is to propose a numerical approach for fatigue assessment of crankshafts, which must reduce the number of specimens required in experimental tests and the development costs. The stress distribution of a commercial crankshaft model was determined numerically from harmonic response analysis, and correction factors related to surface treatments and failure criterion were applied for estimating the fatigue strength. The fatigue strength estimated numerically was compared to experimental fatigue tests results, and this methodology was also applied to a lightweight model obtained by means of mass reduction. The fatigue strength estimation showed 2 % and 7 % relative error for the original and lightweight crankshaft models, respectively. In addition, both numerical and experimental results achieved about 10 % of mass reduction for the commercial crankshaft without considerable influence on the bending and torsion fatigue strength. The present work indicates that the proposed numerical approach can predict with good accuracy the fatigue strength of crankshafts during the design stage.

Recursive Engine In-Cylinder Pressure Reconstruction Using Sensor-Fused Engine Speed.

The engine in-cylinder pressure is a very important parameter for the optimization of internal combustion engines. This paper proposes an alternative recursive Kalman filter-based engine cylinder pressure reconstruction approach using sensor-fused engine speed. In the proposed approach, the fused engine speed is first obtained using the centralized sensor fusion technique, which synthesizes the information from the engine vibration sensor and engine flywheel angular speed sensor. Afterwards, with the fused speed, the engine cylinder pressure signal can be reconstructed by inverse filtering of the engine structural vibration signal. The cylinder pressure reconstruction results of the proposed approach are validated by two combustion indicators, which are pressure peak Pmax and peak location Ploc. Meanwhile, the reconstruction results are compared with the results obtained by the cylinder pressure reconstruction approach using the calculated engine speed. The results of sensor fusion can indicate that the fused speed is smoother when the vibration signal is trusted more. Furthermore, the cylinder pressure reconstruction results can display the relationship between the sensor-fused speed and the cylinder pressure reconstruction accuracy, and with more belief in the vibration signal, the reconstructed results will become better.

A One-Dimensional Numerical Model for High-Performance Two-Stroke Engines

Computer software that simulates the thermodynamic and gas dynamic properties of internal combustion engines can play a significant role in the design and optimization of internal combustion engines. In the present work, a quasi-dimensional numerical model for two-stroke engines is presented. Particular attention was paid to reporting in-cylinder models, combustion (turbulent with flame development and flame–wall interaction), and turbulence (K-k-ϵ model), with the addition of tumble- and squish-generated turbulence that is quite common in such engines. The aim was to reduce the role of the calibration constants, which are fundamental for correlating the models with the experiments, and relations for calculating the tumble ratio and turbulent scales were reported. A one-dimensional model for manifolds is also presented (solving the Euler equations), using the second-order Roe Riemann solver with some improvements, paying particular attention to the source terms, such as area variation. Additionally, a new approach to the end-pipe boundaries, which would reduce the mass conservation error, is reported. The engines tested were two kart two-stroke engines, used for racing purposes: the IAME X30 engine and the IAME Screamer III KZ engine. A comparison between the model results and the experimental data was made, and good accordance was observed, with a root mean square error of about 0.5 kW and providing good accuracy in evaluating changes, such as the combustion chamber squish area and the exhaust pipe length.

Validation and Enhancement of a Supermesh Strategy for the CFD Simulation of Four-Stroke Internal Combustion Engines

The present paper describes and validates an efficient CFD implementation to replicate the working fluid-dynamics of a real four-stroke internal combustion engine. To do this, experimental data obtained on a single-cylinder engine are used to validate the proposed computational approach. The engine domain is divided into regions according to each moving zone, and these are coupled using a pseudo-supermesh interface presented in a previous work by the authors. In this work, the original pseudo-supermesh strategy is enhanced by introducing the dual-boundary concept to model the valve opening/closing events to increase the accuracy and simplicity of the simulation procedure. The results produced by the proposed software tool show a good correlation to the experimental measurements of the complete engine cycle. Macroscopic quantities of the in-cylinder flow are accurately replicated as well as the instantaneous evolution of the in-cylinder and intake manifold pressure. Furthermore, the present work shows that the computational efficiency and scalability of the enhanced pseudo-supermesh approach are preserved even when applied to more complex real problems. In this sense, this work contributes to a new engineering tool promoting the enhanced pseudo-supermeshes as an effective tool for the design, development, and optimization of internal combustion engines.

Recent advances and remaining needs for sciencebased optimization of internal combustion engines

Recent advances and remaining needs for sciencebased optimization of internal combustion engines

Advanced Technologies for the Optimization of Internal Combustion Engines

Even in a scenario where electric vehicles gain market share and the sale of internal combustion engines is gradually reduced, at the present time, there are still no real options that can totally replace the internal combustion (IC) engine over the entire range of its applications [...]

Application of an automated machine learning-genetic algorithm (AutoML-GA) coupled with computational fluid dynamics simulations for rapid engine design optimization

In recent years, the use of machine learning-based surrogate models for computational fluid dynamics (CFD) simulations has emerged as a promising technique for reducing the computational cost associated with engine design optimization. However, such methods still suffer from drawbacks. One main disadvantage is that the default machine learning (ML) hyperparameters are often severely suboptimal for a given problem. This has often been addressed by manually trying out different hyperparameter settings, but this solution is ineffective in case of a high-dimensional hyperparameter space. Besides this problem, the amount of data needed for training is also not known a priori. In response to these issues that need to be addressed, the present work describes and validates an automated active learning approach, AutoML-GA, for surrogate-based optimization of internal combustion engines. In this approach, a Bayesian optimization technique is used to find the best machine learning hyperparameters based on an initial dataset obtained from a small number of CFD simulations. Subsequently, a genetic algorithm is employed to locate the design optimum on the ML surrogate surface. In the vicinity of the design optimum, the solution is refined by repeatedly running CFD simulations at the projected optima and adding the newly obtained data to the training dataset. It is demonstrated that AutoML-GA leads to a better optimum with a lower number of CFD simulations, compared to the use of default hyperparameters. The proposed framework offers the advantage of being a more hands-off approach that can be readily utilized by researchers and engineers in industry who do not have extensive machine learning expertise.

Crank-Angle Resolved Flow Measurements in the Intake Duct of a Research Engine Using Novel and Fast Response Aerodynamic Probes

Despite the public debate nowadays on the future of Internal Combustion Engines (ICE), which is impeding their development, one limitation towards further optimization of ICE in terms of fuel consumption and emissions can be seen in the current approach and more specifically in the transient engine operation and its control. The main drawbacks in the current approach source from: 1) complex structure of mechanization including sensors and actuators, 2) low time resolution and accuracy of sensing (cost driven), 3) complex Electronic Control Unit (ECU)-software architecture associated with huge calibration effort and 4) recently, funded research due to unsecure business model of ICE is becoming less. To overcome these difficulties unexploited potential should be utilized. Some of this potential lies in cycle-by-cycle and cylinder-by-cylinder accurate fuel and air control, and in the development of physical based virtual sensors with high time resolution and accuracy. One of the main motivations for this study was to develop a measurement technique that enables crank-angle resolved air mass flow rate measurements during engine operation in a dynamometer test cell. The measurement principle is quite simple and is based on gauging the dynamic pressure in both the intake and exhaust duct at the closest possible positions to the valves. To fulfill these requirements aerodynamic probes have been developed and manufactured utilizing 3D printing. The probes have been integrated in special developed flanges, which correspond exactly to the shape of the air channels in the cylinder head of the engine. Hence, they can be mounted either in front of the valves at the intake or behind the valves at the exhaust duct. Results at different engine operating conditions have been obtained, analyzed and correlated to other sensors like air-flow meter. Those post-processed results can be further used to validate 1-D gas exchange models, or 3-D Computational Fluid Dynamics (CFD) port flow models. The ultimate scope of these measurements is to calibrate fast physical-based gas exchange models that can be directly used in the engine control framework on an embedded system.

Open source extensions applied to meshing problems for KIVA 4

ABSTRACT KIVA is a very successful software that was released around 1985. One of its advantages is the capacity to include computer routines made by the users. That is why it is mostly used for research and development and as a tool for simulation driven design and optimization of internal combustion engines. As with many CFD software, mesh generation with KIVA is time consuming when its own meshing program is used. This can be exacerbated in the case of complex geometries with a special mesh quality, or when unstructured meshes are needed. Commercially, there are several meshing tools that have features and are available for a price. But this study explores mesh generation for KIVA in four cases of difficult geometries using open source tools. This allows to control all the meshing steps and parameters like mesh refinement, anisotropy and directionality. The first case is a Venturi tube with diesel spray injected in the middle of the throat. The second one an unstructured mesh for an eccentric diesel piston bowl. The last two cases are hybrid meshing methods for an engine with two valves and a pre-chamber, and an engine with two valves and a non-axisymmetric bowl. All the cases were validated in KIVA. This study extends the applicability of KIVA in terms of possibilities and ease of mesh generation. Users can use their preferred CAD/mesh software and convert it to KIVA if the format is compatible with OpenFOAM. Accordingly, the presented methodology would reduce the time devoted to meshing.

An evaluation of combustion aspects with different compression ratios, fuel types and injection systems in a single-cylinder research engine

The growing commercialization of flex-fuel vehicles in Brazil demands the optimization of internal combustion engines for the operation with ethanol (E100) and gasohol (E22), in an attempt to reduce the fuel consumption and minimize the pollutant emissions into the atmosphere. In this sense, this work proposes the study of different volumetric compression ratios in a single-cylinder research engine in order to conclude about its fuel conversion efficiency and, in particular, a more detailed combustion investigation for two of the most common Brazilian fuels. Dynamometric bench tests were performed for distinct compression ratios, injection systems and fuel types, which demanded a specific piston design to meet the requirements for each combustion chamber configuration. The use of ethanol was the most suitable when compared to gasohol, especially at high load conditions, in which was observed a knock tendency for E22 but not for E100, due to its improved physicochemical properties, resulting in enhanced combustion aspects. The proposed methodology proved effective in allowing fuel conversion efficiency gains for the tested fuels, injection systems and piston designs when compared to the engine baseline operation, with up to 6.1% improvement when using the most appropriate compression ratio. Finally, the ethanol direct injection confirmed the potential of this Brazilian biofuel as one of the most promising renewable options for internal combustion engines in current and future sustainable energy directives.

Dynamics of spray impingement wall film under cold start conditions

Fuel film that adhered on engine walls from spray impingement is considered a primary source of harmful combustion emissions. However, the physics of the wall film formation, propagation, and breakup is not fully understood yet because of its multiphase nature. Existing literature has revealed that the mass transportation within the fuel film takes a wave propagation form. This article aims to identify the dynamics of the wall film during spray impingement via high-speed laser diagnostics. In this work, a single-hole injector was used and the spray impinged onto a stage made of sapphire glass for diagnostics purposes. Iso-octane was used as the fuel and 10% ethanol was blended to dope rhodamine 6G as the fluorescent species for laser excitation. Simultaneous optical measurements, such as laser-induced fluorescence and Mie scattering, are performed to obtain the characteristics of the wall film quantitatively. Various aspects of the wall film, including the frequency of the wave, wave speed, and wave height, are inspected. The impact of fuel temperature and wall temperature under typical cold-start conditions are also studied to investigate the temperature dependence of the wall film dynamics. The research is also intended to provide quantitative experimental data for numerical models for impingement prediction and thus help the process of emission reduction and combustion optimization of internal combustion engines.

Model based preliminary design and optimization of Internal Combustion Engine and Fuel Cell hybrid electric vehicle

The present paper analyses and compares two optimized solutions for a HEVs: a spark ignition Internal Combustion Engine (ICE) based one and Fuel Cell (FC) based one. A mathematical model to simulate the performance and the energy consumption of the two proposed solutions has been developed using the WLTP, US06, Highway and Urban driving cycles for the comparison. The control strategy assumed for the power sharing in both the configuration is a hybrid SOC based load following one, since that it enables to decouple the APU and the battery pack size.

Inclusion of entropy generation minimization in multi-objective CFD optimization of diesel engines

This paper describes the optimization of operating conditions of a direct injection diesel engine by the response surface method targeting the first and second law efficiencies and a merit function accounting for emission. To the best of our knowledge, this is the first time that entropy generation minimization (EGM) is considered as one of the objectives of 3D CFD optimization of internal combustion engines. Therefore, the effects of different design parameters, i.e. SOI (start of injection), DOI (duration of injection) and spray angle, on the entropy generation response surface are investigated in detail. Various methods of optimization such as MOGA (multi-objective Genetic algorithm), MOSA (multi-objective simulated annealing), MOPSO (multi-objective particle swarm optimization) and NSGA (non-sorted Genetic algorithm) are used and compared. It is observed that among the design parameters, SOI is the most important parameter influencing the first law efficiency and emission while spray angle is the least important one. Also, it is found that DOI has the strongest impact on the second law efficiency. The most fitting error relates to response surface of the second law efficiency since, compared to the other objectives, it shows a much more complex behavior and changes nonlinearly with the change in design parameters.

Multi-objective optimization of internal combustion engine by means of 1D fluid-dynamic models

The definition of an efficient optimization methodology for internal combustion engine design using 1D fluid dynamic simulation models is presented. This work aims at discussing the fundamental numerical and fluid dynamic aspects which can lead to the definition of a best practice technique, depending on the complexity of the problem to be dealt with, on the number of design parameters, objective variables and constrains. For these reasons, both single-and multi-objective problems will be addressed, where the former are still of relevant interest (i.e. optimization of engine performances), while the latter have a much wider range of applications and are often characterized by conflicting objectives. The Mesh Adaptive Direct Search (MADS) was chosen among the class of direct search methods and compared with the Genetic Algorithms to solve single-objective problems, and similarly two different algorithms were chosen and compared to solve multi-objective problems: the ε-constraint method and the NSGA-II (Non-Dominated Sorting Genetic Algorithm) A single cylinder spark ignition engine, used in a motorbike application, was chosen as test case, to allow reduced computational times, without any loss of generality of the results. The analysis evaluate the convergence and efficiency of each methodology for the different problems which are solved. The achieved goal is not the definition of an ever valid mathematical strategy, but here focus is given on the parallel application of a detailed fluid dynamic analysis and automated optimization techniques to suggest a best practice technique to be employed depending on the characteristic of the optimization problem to be solved.

An efficient generic method for calculating the properties of combustion products

A simple method for calculating the properties of combustion products is presented. The method is generic: the same equations apply to any fuel for any fuel—air ratio and temperature. The equations are suitable for both hand calculation and computer applications requiring repetitive property evaluations. The present equations offer accuracy comparable with the most elaborate method available but at a significantly reduced computational cost. The theory should find direct use in design-point and off-design performance calculations and optimization of internal combustion engines, e.g. gas turbine plants, calculation of the three-dimensional flowfield within a multistage turbine, and other applications requiring properties of combustion products.

A methodology for engine design using multi-dimensional modelling and genetic algorithms with validation through experiments

A methodology for internal combustion engine design has been formulated which incorporates multidimensional modelling and experiments to optimize and simulate direct injection diesel engine combustion and emissions formation. The computer code KIVA-GA performs full-cycle engine simulations within the framework of a genetic algorithm (GA) global optimization code. The methodology is applied to optimize a heavy-duty diesel truck engine. The study simultaneously investigated the effects of six engine input parameters on emissions and performance for a high-speed medium-load operating point. The start of injection (SOI), injection pressure, amount of exhaust gas recirculation (EGR), boost pressure and split injection rate shape were optimized. The convergence of the GA optimization process is demonstrated and the results were compared to those of the experimental optimization study employing a response surface method (RSM), which uses statistically designed experiments to determine an optimum design. In addition, the parameters of the computationally predicted optimum were run experimentally and good agreement was obtained. The potential for ultra-low emissions levels was assessed through additional computational GA runs that included higher maximum EGR levels (up to 50 per cent). The predicted optimum results in significantly lower soot and NOx emissions together with improved fuel consumption compared to the baseline design. The present results indicate that an efficient design methodology has been developed for optimization of internal combustion engines, one that allows simultaneous optimization of a large number of parameters.