scholarly journals Engine Combustion System Optimization Using Computational Fluid Dynamics and Machine Learning: A Methodological Approach

2020 ◽  
Vol 143 (2) ◽  
Author(s):  
Jihad A. Badra ◽  
Fethi Khaled ◽  
Meng Tang ◽  
Yuanjiang Pei ◽  
Janardhan Kodavasal ◽  
...  
Keyword(s):  
Machine Learning ◽  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Internal Combustion Engines ◽  
Optimization Problems ◽  
Combustion Engine ◽  
Methodological Approach ◽  
System Optimization ◽  
Operating Conditions ◽  
Attractive Alternative

Abstract Gasoline compression ignition (GCI) engines are considered an attractive alternative to traditional spark-ignition and diesel engines. In this work, a Machine Learning-Grid Gradient Ascent (ML-GGA) approach was developed to optimize the performance of internal combustion engines. ML offers a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. The developed ML-GGA model was compared with a recently developed Machine Learning-Genetic Algorithm (ML-GA). Detailed investigations of optimization solver parameters and variable limit extension were performed in the present ML-GGA model to improve the accuracy and robustness of the optimization process. Detailed descriptions of the different procedures, optimization tools, and criteria that must be followed for a successful output are provided here. The developed ML-GGA approach was used to optimize the operating conditions (case 1) and the piston bowl design (case 2) of a heavy-duty diesel engine running on a gasoline fuel with a research octane number (RON) of 80. The ML-GGA approach yielded >2% improvements in the merit function, compared with the optimum obtained from a thorough computational fluid dynamics (CFD) guided system optimization. The predictions from the ML-GGA approach were validated with engine CFD simulations. This study demonstrates the potential of ML-GGA to significantly reduce the time needed for optimization problems, without loss in accuracy compared with traditional approaches.

scholarly journals Engine Combustion System Optimization Using CFD and Machine Learning: A Methodological Approach

2019 ◽  
Author(s):  
Jihad Badra ◽  
Fethi Khaled ◽  
Meng Tang ◽  
Yuanjiang Pei ◽  
Janardhan Kodavasal ◽  
...  
Keyword(s):  
Machine Learning ◽  
Internal Combustion Engines ◽  
Optimization Problems ◽  
Combustion Engine ◽  
Methodological Approach ◽  
System Optimization ◽  
Merit Function ◽  
Operating Conditions ◽  
Gradient Algorithm ◽  
Attractive Alternative

Abstract Gasoline compression ignition (GCI) engines are considered an attractive alternative to traditional spark-ignition and diesel engines. In this work, a Machine Learning-Grid Gradient Algorithm (ML-GGA) approach was developed to optimize the performance of internal combustion engines. Machine learning (ML) offers a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. The developed ML-GGA model was compared with a recently developed Machine learning Genetic Algorithm (ML-GA). Detailed investigations of optimization solver parameters and variables limits extension were performed in the present ML-GGA model to improve the accuracy and robustness of the optimization process. Detailed descriptions of the different procedures, optimization tools and criteria that must be followed for a successful output are provided here. The developed ML-GGA approach was used to optimize the operating conditions (case 1) and the piston bowl design (case 2) of a heavy-duty diesel engine running on a gasoline fuel with a Research Octane Number (RON) of 80. The ML-GGA approach yielded > 2% improvements in the merit function, compared to the optimum obtained from a thorough computational fluid dynamics (CFD) guided system optimization. The predictions from the ML-GGA approach were validated with engine CFD simulations. This study demonstrates the potential of ML-GGA to significantly reduce the time needed for optimization problems, without loss in accuracy compared to traditional approaches.

scholarly journals THE COMBUSTION ENGINE INTAKE MANIFOLD CFD MODELING DEPENDING ON THE AIRBOX CONFIGURATION

2021 ◽  
Vol 2021 (6) ◽  
pp. 5421-5425
Author(s):  
MICHAL RICHTAR ◽  
◽  
PETRA MUCKOVA ◽  
JAN FAMFULIK ◽  
JAKUB SMIRAUS ◽  
...  
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Combustion Chamber ◽  
Internal Combustion Engines ◽  
Combustion Engine ◽  
Stationary Flow ◽  
Cfd Modeling ◽  
Intake Manifold ◽  
Combustion Engines ◽  
Computational Fluid Dynamics Cfd

The aim of the article is to present the possibilities of application of computational fluid dynamics (CFD) to modelling of air flow in combustion engine intake manifold depending on airbox configuration. The non-stationary flow occurs in internal combustion engines. This is a specific type of flow characterized by the fact that the variables depend not only on the position but also on the time. The intake manifold dimension and geometry strongly effects intake air amount. The basic target goal is to investigate how the intake trumpet position in the airbox impacts the filling of the combustion chamber. Furthermore, the effect of different distances between the trumpet neck and the airbox wall in this paper will be compared.

scholarly journals A zonal approach for estimating pressure ratio at compressor extreme off-design conditions

2018 ◽  
Vol 20 (4) ◽  
pp. 393-404 ◽  
Author(s):  
José Galindo ◽  
Roberto Navarro ◽  
Luis Miguel García-Cuevas ◽  
Daniel Tarí ◽  
Hadi Tartoussi ◽  
...  
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
High Speed ◽  
Internal Combustion Engines ◽  
Pressure Ratio ◽  
Internal Combustion ◽  
Operating Conditions ◽  
Combustion Engines ◽  
One Dimensional ◽  
Performance Map

Zero-dimensional/one-dimensional computational fluid dynamics codes are used to simulate the performance of complete internal combustion engines. In such codes, the operation of a turbocharger compressor is usually addressed employing its performance map. However, simulation of engine transients may drive the compressor to work at operating conditions outside the region provided by the manufacturer map. Therefore, a method is required to extrapolate the performance map to extended off-design conditions. This work examines several extrapolating methods at the different off-design regions, namely, low-pressure ratio zone, low-speed zone and high-speed zone. The accuracy of the methods is assessed with the aid of compressor extreme off-design measurements. In this way, the best method is selected for each region and the manufacturer map is used in design conditions, resulting in a zonal extrapolating approach aiming to preserve accuracy. The transitions between extrapolated zones are corrected, avoiding discontinuities and instabilities.

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

2021 ◽  
pp. 146808742110234
Author(s):  
Opeoluwa Owoyele ◽  
Pinaki Pal ◽  
Alvaro Vidal Torreira ◽  
Daniel Probst ◽  
Matthew Shaxted ◽  
...  
Keyword(s):  
Machine Learning ◽  
Fluid Dynamics ◽  
Genetic Algorithm ◽  
Computational Fluid Dynamics ◽  
Design Optimization ◽  
Internal Combustion Engines ◽  
Bayesian Optimization ◽  
Cfd Simulations ◽  
Engine Design ◽  
Design Optimum

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.

Computational Fluid Dynamics Thermohydrodynamic Analysis of Three-Dimensional Sector-Pad Thrust Bearings With Rectangular Dimples

2013 ◽  
Vol 136 (1) ◽  
Author(s):  
C. I. Papadopoulos ◽  
L. Kaiktsis ◽  
M. Fillon
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Carrying Capacity ◽  
Thermal Effects ◽  
Operating Conditions ◽  
Load Carrying Capacity ◽  
Performance Indices ◽  
Thrust Bearings ◽  
Load Carrying ◽  
Texture Design

The paper presents a detailed computational study of flow patterns and performance indices in a dimpled parallel thrust bearing. The bearing consists of eight pads; the stator surface of each pad is partially textured with rectangular dimples, aiming at maximizing the load carrying capacity. The bearing tribological performance is characterized by means of computational fluid dynamics (CFD) simulations, based on the numerical solution of the Navier–Stokes and energy equations for incompressible flow. Realistic boundary conditions are implemented. The effects of operating conditions and texture design are studied for the case of isothermal flow. First, for a reference texture pattern, the effects of varying operating conditions, in particular minimum film thickness (thrust load), rotational speed and feeding oil pressure are investigated. Next, the effects of varying texture geometry characteristics, in particular texture zone circumferential/radial extent, dimple depth, and texture density on the bearing performance indices (load carrying capacity, friction torque, and friction coefficient) are studied, for a representative operating point. For the reference texture design, the effects of varying operating conditions are further investigated, by also taking into account thermal effects. In particular, adiabatic conditions and conjugate heat transfer at the bearing pad are considered. The results of the present study indicate that parallel thrust bearings textured by proper rectangular dimples are characterized by substantial load carrying capacity levels. Thermal effects may significantly reduce load capacity, especially in the range of high speeds and high loads. Based on the present results, favorable texture designs can be assessed.

Metal Temperature Prediction of a Dry Low NOx Class Flame Tube by Computational Fluid Dynamics Conjugate Heat Transfer Approach

2015 ◽  
Vol 138 (3) ◽  
Author(s):  
Riccardo Da Soghe ◽  
Cosimo Bianchini ◽  
Antonio Andreini ◽  
Lorenzo Mazzei ◽  
Giovanni Riccio ◽  
...  
Keyword(s):  
Heat Transfer ◽  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Operating Conditions ◽  
Metal Temperature ◽  
Thermal Loads ◽  
Flame Tube ◽  
Numerical Predictions ◽  
Partial Load ◽  
Critical Components

Combustor liner of present gas turbine engines is subjected to high thermal loads as it surrounds high temperature combustion reactants and is hence facing the related radiative load. This generally produces high thermal stress levels on the liner, strongly limiting its life expectations and making it one of the most critical components of the entire engine. The reliable prediction of such thermal loads is hence a crucial aspect to increase the flame tube life span and to ensure safe operations. The present study aims at investigating the aerothermal behavior of a GE Dry Low NOx (DLN1) class flame tube and in particular at evaluating working metal temperatures of the liner in relation to the flow and heat transfer state inside and outside the combustion chamber. Three different operating conditions have been accounted for (i.e., lean–lean partial load, premixed full load, and primary load) to determine the amount of heat transfer from the gas to the liner by means of computational fluid dynamics (CFD). The numerical predictions have been compared to experimental measurements of metal temperature showing a good agreement between CFD and experiments.

Performance Analysis of Gas-Expanded Lubricants in a Hybrid Bearing Using Computational Fluid Dynamics

2015 ◽  
Author(s):  
Brian K. Weaver ◽  
Gen Fu ◽  
Andres F. Clarens ◽  
Alexandrina Untaroiu
Keyword(s):  
Fluid Dynamics ◽  
Computational Fluid Dynamics ◽  
Phase Behavior ◽  
Ambient Pressure ◽  
Operating Conditions ◽  
Full Potential ◽  
Dissolved Carbon ◽  
Bearing Stiffness ◽  
Multi Phase ◽  
Stiffness And Damping

Gas-expanded lubricants (GELs), tunable mixtures of synthetic oil and dissolved carbon dioxide, have been previously shown to potentially increase bearing efficiency, rotordynamic control, and long-term reliability in flooded journal bearings by controlling the properties of the lubricant in real time. Previous experimental work has established the properties of these mixtures and multiple numerical studies have predicted that GELs stand to increase the performance of flooded bearings by reducing bearing power losses and operating temperatures while also providing control over bearing stiffness and damping properties. However, to date all previous analytical studies have utilized Reynolds equation-based approaches while assuming a single-phase mixture under high-ambient pressure conditions. The potential implications of multi-phase behavior could be significant to bearing performance, therefore a more detailed study of alternative operating conditions that may include multi-phase behavior is necessary to better understanding the full potential of GELs and their effects on bearing performance. In this work, the performance of GELs in a fixed geometry journal bearing were evaluated to examine the effects of these lubricants on the fluid and bearing dynamics of the system under varying operating conditions. The bearing considered for this study was a hybrid hydrodynamic-hydrostatic bearing to allow for the study of various lubricant supply and operating conditions. A computational fluid dynamics (CFD)-based approach allowed for a detailed evaluation of the lubricant injection pathway, the flow of fluid throughout the bearing geometry, thermal behavior, and the collection of the lubricant as it exits the bearing. This also allowed for the study of the effects of the lubricant behavior on overall bearing performance.

Sign in / Sign up

Export Citation Format

Share Document