Data-driven enabling technologies in soft sensors of modern internal combustion engines: Perspectives

Li, Ji; Zhou, Quan; He, X. T.; Chen, Wan; Xu, Hongming

doi:10.1016/j.energy.2023.127067

Cited by 8 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Through numerical simulation, the various processes in a hydrogen combustion engine can be easily seen, such as the distribution of the temperature field in the cylinder, the distribution of the concentration of the combustion mass, and the generation pattern of the products. , Meanwhile, in the foreseeable future, it will be a trend to build digital twins based on such methods to achieve the purpose of virtual testing, real-time monitoring, health operation, and maintenance of complex equipment. At present, in the field of the internal combustion engine, scholars have already made attempts to build digital twins on internal combustion engine sensors, internal combustion engine power plants, and other objects. , However, it has not yet been reported in the field of HICEs.…”

Section: Fundamentals Of the Hicesmentioning

confidence: 99%

See 1 more Smart Citation

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Wang,

Li,

Hong

et al. 2023

ACS Omega

View full text Add to dashboard Cite

Energy shortages and environmental problems have brought many challenges to China’s development. In the field of transport, hydrogen energy has become a new type of energy that people pay attention to due to its easy production and nonpolluting generation. The use of hydrogen as a fuel in internal combustion engines can be a good way to make a low-cost and clean conversion of the current internal combustion engine, thus realizing the application of hydrogen energy in vehicles. In this study, hydrogen-fueled internal combustion engines (HICEs) are the focus. A review and analysis of the topical issues encountered in the developmental research of HICE is presented, such as the optimization control method of the combustion process, mechanism and suppression method of abnormal combustion (preignition, backfire, knock, and high pressure rise rate in early stages of combustion), influence regularity of hydrogen injection parameters and hydrogen injection modes on the formation and combustion performance of H 2 –air mixture, emissions and control of NO x , formation of H 2 –air mixture, and combustion cycle variation, among others. Multiyear studies in the hydrogen-fueled engines clearly show that abnormal combustion, as preignition, backfire, and the blockage of intake pipe, is likely to occur at high load, and the H 2 injection parameters and the injection modes possess a prominent effect on the engine’s performance and the blockage of the intake pipe, which has an important impact on the preignition and backfire, and the optimal control of the combustion process is a valuable method for resolving the contradictions among inhibiting abnormal combustion, the blockage of the intake pipe, and enhancing hydrogen-fueled engine power.

show abstract

Section: Fundamentals Of the Hicesmentioning

confidence: 99%

“…At present, in the field of the internal combustion engine, scholars have already made attempts to build digital twins on internal combustion engine sensors, internal combustion engine power plants, and other objects. 14 , 15 However, it has not yet been reported in the field of HICEs.…”

Section: Fundamentals Of the Hicesmentioning

confidence: 99%

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Wang,

Li,

Hong

et al. 2023

ACS Omega

View full text Add to dashboard Cite

show abstract

“…These simulation-based strategies effectively overcome the limitations of traditional laboratory testing methods and serve as a crucial bridge between the two primary estimation techniques. In the current context, machine learning models are increasingly relevant in sectors such as energy and automotive technology [9], [10], [11]. Through integrating engine research with machine learning modeling approaches, it is feasible to optimize engine calibration and reduce the dependency on extensive experimentation and Threedimensional (3D) simulation [12], [13], [14].…”

Section: Introductionmentioning

confidence: 99%

Integrating Machine Learning for Predicting Internal Combustion Engine Performance and Segment-Based CO₂ Emissions Across Urban and Rural Settings

Niroomand,

Bach

2024

IEEE Access

View full text Add to dashboard Cite

The assessment of artificial intelligence (AI) application for prediction of internal combustion engine (ICE) performance and its impact on CO2 emissions is conducted in this paper. Three machine learning techniques (Random Forest, Support Vector Regression, and Semi-supervised Deep Fuzzy C-means) are developed to analyze inputs from an engine simulation software package database. By employing these sophisticated mathematical techniques, we successfully assess the influence of engine power range on CO2 emissions in this paper. Moreover, the framework facilitates segment-based analysis, which enables segmentspecific assessment of CO2 emission based on metrics such as average traveled distance and average daily trips in urban and rural settings. The DFCM model seems promising to predict engine performance, with high predictive accuracy and a coefficient of determination (R 2 ) approaching unity. The results indicate that integration of inter-class and intra-class distinctions, along with considering the interquartile range of engine power provides invaluable insights for the formulation of strategies aimed at overhauling the passenger vehicle fleet and advancing decarbonization efforts. By implementing the proposed innovative techniques, we aspire to enrich the precision of ICE emission models, leading to more reliable calculations and an enhanced understanding of the environmental implications associated with vehicles.

show abstract

“…Soft sensing technology can be broadly classified into two main types: model‐driven (fundamental, first‐principle) and data‐driven (empirical, black box). [ 12 ] Model‐driven methods are based on mathematical models of the system and require a deep understanding of the chemical and physical principles involved in the process, which can be challenging to achieve in complex industrial applications. In contrast, data‐driven methods do not rely on system models but directly utilize historical data for modelling and prediction, reducing the need for specific industrial process knowledge.…”

Section: Introductionmentioning

confidence: 99%

Soft sensor based on multi‐phase stacking ensemble model with self‐selected primary learner for batch processes

Zhao,

Yang,

et al. 2023

Can J Chem Eng

View full text Add to dashboard Cite

In batch processes, existing soft sensing methodologies encounter substantial challenges when confronted with nonlinearity and multi‐phase issues. In response to these challenges, an innovative soft sensing framework known as the multi‐phase stacking ensemble model with self‐selected primary learner is proposed. The main innovation of this framework lies in the solution to the primary learner selection issue within the stacking model. To commence, the batch process is divided into multiple phases employing a Gaussian mixture model, thereby establishing local ensemble models for each phase. Subsequently, the iterative self‐selection of primary learners strategy is proposed, which iteratively selects suitable primary learners for these models, optimizing their combination of primary learners for each local model. This primary learner selection strategy effectively enhances the accuracy of predictions in the stacking ensemble model. To further enhance the performance, Bayesian optimization is utilized to tune the hyperparameters of each local ensemble model. This step guarantees optimal performance of the model across diverse phases. Extensive simulation experiments are conducted on an industrial penicillin fermentation process to validate the effectiveness of the proposed framework. According to the findings, the model demonstrated superior performance compared to existing single‐learner soft sensing methods and commonly utilized ensemble‐based soft sensing methods in terms of both R2 score (0.97584) and RMSE (0.0513). Overall, this framework offers a novel approach for selecting primary learners in stacking ensemble models and enhancing the predictive performance in batch processes for soft sensing.

show abstract

Data-driven enabling technologies in soft sensors of modern internal combustion engines: Perspectives

Cited by 8 publications

References 35 publications

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Integrating Machine Learning for Predicting Internal Combustion Engine Performance and Segment-Based CO₂ Emissions Across Urban and Rural Settings

Soft sensor based on multi‐phase stacking ensemble model with self‐selected primary learner for batch processes

Contact Info

Product

Resources

About

Data-driven enabling technologies in soft sensors of modern internal combustion engines: Perspectives

Cited by 8 publications

References 35 publications

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Research and Development of Hydrogen-Fueled Internal Combustion Engines in China

Integrating Machine Learning for Predicting Internal Combustion Engine Performance and Segment-Based CO2 Emissions Across Urban and Rural Settings

Soft sensor based on multi‐phase stacking ensemble model with self‐selected primary learner for batch processes

Contact Info

Product

Resources

About

Integrating Machine Learning for Predicting Internal Combustion Engine Performance and Segment-Based CO₂ Emissions Across Urban and Rural Settings