Comparative study of the artificial neural network with three hyper-parameter optimization methods for the precise LP-EGR estimation using in-cylinder pressure in a turbocharged GDI engine

Jo, Yuhyeok; Min, K.; Jung, Donghyuk; Sunwoo, Myoungho; Han, Manbae

doi:10.1016/j.applthermaleng.2018.12.139

Cited by 19 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jo et al [ 72 ] studied the turbocharged gasoline direct engine, which involves the low-pressure cooled exhaust gas recirculation to curtail NOx emissions. The long path of LP-EGR leads to the transport delay, which leads to an inaccurate estimation of LP-EGR flow.…”

Section: Modeling Of Internal Combustion Enginesmentioning

confidence: 99%

Application of Artificial Neural Network for Internal Combustion Engines: A State of the Art Review

Bhatt

Shrivastava

2021

Arch Computat Methods Eng

View full text Add to dashboard Cite

The automotive industry is facing a crucial time. The transformation from internal combustion engines to new electrical technologies requires enormous investment, and hence the IC engines are likely to serve as a means of transportation for the coming decades. The search for sustainable green alternative fuel and operating parameter optimization is a current feasible solution and is a critical issue among the scientific community. Engine experiments are complicated, costly, and time-consuming, especially when the global economy is drastically down due to the COVID-19 pandemic and putting the limitation of social distancing. Industries are looking for proven computational solutions to address these issues. Recently, artificial neural network has been proven beneficial in several areas of engineering to reduce the time and experimentation cost. The IC engine is one of them. ANN has been used to predict and analyze different characteristics such as performance, combustion, and emissions of the IC engine to save time and energy. The complex nature of ANN may lead to computation time, energy, and space. Recent studies are centered on changing the network topology, deep learning, and design of ANN to get the highest performance. The present study summarizes the application of ANN to predict and optimize the complicated characteristics of various types of engines with different fuels. The study aims to investigate the network topologies adopted to design the model and thereafter statistical evaluation of the developed ANN models. A comparison of the ANN model with other prediction models is also presented.

show abstract

Section: Modeling Of Internal Combustion Enginesmentioning

confidence: 99%

Application of Artificial Neural Network for Internal Combustion Engines: A State of the Art Review

Bhatt

Shrivastava

2021

Arch Computat Methods Eng

View full text Add to dashboard Cite

show abstract

“…The optimisation of the EGR system settings for automotive applications were studied in [42][43][44][45][46][47][48][49][50]. Hiroyasu [42] carried out an optimisation study for a single-cylinder truck engine by employing the neighbourhood cultivation genetic algorithm (NCGA) with the objective to minimise the NO x and soot emissions along with the fuel consumption by controlling the fuel injection shape, the boost pressure, the EGR rate, the start of injection, the injection duration angle, and the swirl ratio.…”

Section: Exhaust Gas Recirculation System Settings Optimisationmentioning

confidence: 99%

“…Jaliliantabar et al [46] numerically investigated a small single-cylinder four-stroke diesel engine equipped with an EGR system, aiming to optimise the EGR rate and the biodiesel fuel percentage using the nondominated sorting genetic algorithm NSGA-II. Jo et al [47] proposed an artificial neural network (ANN) using multiple combustion parameters, calculated from the in-cylinder pressure, to estimate the LP-EGR rate of a turbocharged GDI engine. This ANN was trained and validated using experimental data at steady-state conditions.…”

Section: Exhaust Gas Recirculation System Settings Optimisationmentioning

confidence: 99%

Multiobjective Optimisation of a Marine Dual Fuel Engine Equipped with Exhaust Gas Recirculation and Air Bypass Systems

Stoumpos

Theotokatos

2020

Energies

View full text Add to dashboard Cite

Dual fuel engines constitute a viable solution for enhancing the environmental sustainability of the shipping operations. Although these engines comply with the Tier III NOx emissions regulations when operating at the gas mode, additional measures are required to ensure such compliance at the diesel mode. Hence, this study aimed to optimise the settings of a marine four-stroke dual fuel (DF) engine equipped with exhaust gas recirculation (EGR) and air bypass (ABP) systems by employing simulation and optimisation techniques, so that the engine when operating at the diesel mode complies with the ‘Tier III’ requirements. A previous version of the engine thermodynamic model was extended to accommodate the EGR and ABP systems modelling. Subsequently, a combination of optimisation techniques including multiobjective genetic algorithms (MOGA) and design of experiments (DoE) parametric runs was employed to identify both the engine and the EGR/ABP systems settings with the objective to minimise the engine brake specific fuel consumption and reduce the NOx emissions below the Tier III limit. The derived simulation results were employed to analyse the EGR system involved interactions and their effects on the engine performance and emissions trade-offs. A sensitivity analysis was performed to reveal the interactions between considered engine settings and quantify their impact on the engine performance parameters. The derived results indicate that EGR rates up to 35% are required, so that the investigated engine with EGR and ABP systems, when operating at the diesel mode, achieves compliance with the ‘Tier III’ NOx emissions, whereas the associated engine brake specific fuel consumption penalty is up to 8.7%. This study demonstrates that the combination of EGR and ABP systems can constitute a functional solution for achieving compliance with the stringent regulatory requirements and provides a better understating of the underlined phenomena and interactions of the engine subsystems parameters variations for the investigated engine equipped with EGR and ABP systems.

show abstract

“…The results showed that the method performs with higher learning speed with lower network complexity. Jo et al [16] conducted a comparative study of optimisation methods to search for the optimal values of the ANN's hyper-parameter. The goal is obtaining a precise model of the LP-EGR rate in order to improve the fuel efficiency and reduce exhaust emissions in turbocharged GDI engines.…”

Section: Related Workmentioning

confidence: 99%

Using artificial neural networks to represent a diesel–biodiesel engine

Lage

Hanriot

Zárate

2020

J Braz. Soc. Mech. Sci. Eng.

View full text Add to dashboard Cite

In this work, six computational models based on artificial neural networks were developed to simulate an operating diesel engine fuelled with 8% biodiesel in order to predict performance and emissions of a diesel-biodiesel engine in a group generator. The ANN models were used to simulate a diesel-biodiesel engine that has four cylinders with a volume of 3.9 l, a compression ratio of 17:1, direct injection, and a rated power of 49 kW. The models were validated against experimental data for 10 kW, 20 kW, and 30 kW loads. The models were capable of accurately predicting the output power, thermal efficiency, and emissions of CO 2 , CO, NO, and NOx. Their comparison with experimental results showed a satisfactory agreement and the reliability of their predicted results for new operating conditions.

show abstract

Comparative study of the artificial neural network with three hyper-parameter optimization methods for the precise LP-EGR estimation using in-cylinder pressure in a turbocharged GDI engine

Cited by 19 publications

References 53 publications

Application of Artificial Neural Network for Internal Combustion Engines: A State of the Art Review

Application of Artificial Neural Network for Internal Combustion Engines: A State of the Art Review

Multiobjective Optimisation of a Marine Dual Fuel Engine Equipped with Exhaust Gas Recirculation and Air Bypass Systems

Using artificial neural networks to represent a diesel–biodiesel engine

Contact Info

Product

Resources

About