Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
The automotive industry is increasingly challenged to develop cleaner, more efficient solutions to comply with stringent emission standards. Hydrogen (H2)-powered internal combustion engines (ICEs) offer a promising alternative, with the potential to reduce carbon-based emissions and improve efficiency. However, hydrogen combustion presents two main challenges related to the calibration process: emissions control and measurement of the air excess coefficient (λ). Traditional lambda sensors struggle with hydrogen’s combustion dynamics, leading to potential inefficiencies and increased pollutant emissions. Consequently, the determination of engine performance could also be compromised. This study explores the feasibility of using machine learning (ML) to replace physical lambda sensors with virtual ones in hydrogen-fueled ICEs. The research was conducted on a single-cylinder spark-ignition (SI) engine, collecting data across a range of air excess coefficients from 1.6 to 3.0. An advanced hybrid model combining long short-term memory (LSTM) networks and convolutional neural networks (CNNs) was developed and fine-tuned to accurately predict the air–fuel ratio; its predictive performance was compared to that obtained with the backpropagation (BP) architecture. The optimal configuration was identified through iterative experimentation, focusing on the neuron count, number of hidden layers, and input variables. The results demonstrate that the LSTM + 1DCNN model successfully converged without overfitting; it also showed better prediction ability in terms of accuracy and robustness when compared with the backpropagation approach.
The automotive industry is increasingly challenged to develop cleaner, more efficient solutions to comply with stringent emission standards. Hydrogen (H2)-powered internal combustion engines (ICEs) offer a promising alternative, with the potential to reduce carbon-based emissions and improve efficiency. However, hydrogen combustion presents two main challenges related to the calibration process: emissions control and measurement of the air excess coefficient (λ). Traditional lambda sensors struggle with hydrogen’s combustion dynamics, leading to potential inefficiencies and increased pollutant emissions. Consequently, the determination of engine performance could also be compromised. This study explores the feasibility of using machine learning (ML) to replace physical lambda sensors with virtual ones in hydrogen-fueled ICEs. The research was conducted on a single-cylinder spark-ignition (SI) engine, collecting data across a range of air excess coefficients from 1.6 to 3.0. An advanced hybrid model combining long short-term memory (LSTM) networks and convolutional neural networks (CNNs) was developed and fine-tuned to accurately predict the air–fuel ratio; its predictive performance was compared to that obtained with the backpropagation (BP) architecture. The optimal configuration was identified through iterative experimentation, focusing on the neuron count, number of hidden layers, and input variables. The results demonstrate that the LSTM + 1DCNN model successfully converged without overfitting; it also showed better prediction ability in terms of accuracy and robustness when compared with the backpropagation approach.
<div class="section abstract"><div class="htmlview paragraph">The global push to minimize carbon emissions and the imposition of more rigorous regulations on emissions are driving an increased exploration of cleaner powertrains for transportation. Hydrogen fuel applications in internal combustion engines are gaining prominence due to their zero carbon emissions and favorable combustion characteristics, particularly in terms of thermal efficiency. However, conventional Spark-Ignition (SI) engines are facing challenges in meeting performance expectations while complying with strict pollutant-emission regulations. These challenges arise from the engine's difficulty in handling advanced combustion strategies, such as lean mixtures, attributed to factors like low ignition energy and abnormal combustion events.</div><div class="htmlview paragraph">To address these issues, the Barrier Discharge Igniter (BDI) stands out for its capability to generate non-equilibrium Low-Temperature Plasma (LTP), a strong promoter of ignition through kinetic, thermal, and transport effects. Its surface discharge also facilitates combustion promotion across a wide area, overcoming the limitations of conventional spark systems. The research outlined in this study involves conducting experiments that integrate hydrogen (H<sub>2</sub>) with LTP discharge. Tests were carried out using a single-cylinder research engine by varying the air-fuel mixture and maintaining the same load condition and the same engine speed. Results from the application of BDI, revealed an acceleration in the evolution of the flame front when compared to conventional spark methods. This effect extended the lean stable limit of the engine, leading to reduction in the fuel consumption and emissions and improvements in the delivered power close to the engine lean stable limit. Additionally, adjustment of BDI control parameters played a crucial role in enhancing igniter performance, contributing significantly to a more comprehensive understanding of the innovative approach presented in this study.</div></div>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.