Search Results (2,359)

Against the global carbon neutrality target, predictive maintenance (PdM) of automotive engines represents a core technical strategy to advance the sustainable development of the automotive industry. Conventional single-modal diagnostic approaches for engine abnormal sound defects suffer from low accuracy and weak anti-interference capability. Existing multi-modal fusion methods fail to deeply mine the physical coupling between cross-modal features and often entail excessive model complexity, hindering deployment on resource-constrained on-board edge devices. To resolve these limitations, this study proposes a Physical Prior-Embedded Cross-Modal Attention (PPE-CMA) mechanism for lightweight multi-modal fusion diagnosis of engine abnormal sound defects. First, wavelet packet decomposition (WPD) and mel-frequency cepstral coefficients (MFCC) are integrated to extract time-frequency features from engine audio signals, while a channel-pruned ResNet18 is employed to extract spatial features from engine thermal imaging and vibration visualization images. Second, the PPE-CMA module is designed to adaptively assign attention weights to audio and image features by exploiting the physical coupling between engine fault acoustic and visual characteristics, enabling efficient cross-modal feature fusion with redundant information suppression. A rigorous theoretical derivation is provided to link cosine similarity with the physical correlation of engine fault acoustic-visual features, justifying the attention weight constraint (β = 1 − α) from the perspective of fault feature physical coupling. Third, an improved lightweight XGBoost classifier is constructed for fault classification, and a hybrid data augmentation strategy customized for engine multi-modal data is proposed to address the small-sample challenge in industrial applications. Ablation experiments on ResNet18 pruning ratios verify the optimal trade-off between diagnostic performance and computational efficiency, while feature distribution analysis validates the authenticity and effectiveness of the hybrid augmentation strategy. Experimental results on a self-constructed multi-modal dataset show that the proposed method achieves 98.7% diagnostic accuracy and a 98.2% F1-score, retaining 96.5% accuracy under 90 dB high-level environmental noise, with an end-to-end inference speed of 0.8 ms per sample (including preprocessing, feature extraction, and classification). Cross-engine and cross-domain validation on a 2.0T diesel engine small-sample dataset and the open-source SEMFault-2024 dataset yield average accuracies of 94.8% and 95.2%, respectively, demonstrating strong generalization. This method effectively enhances the accuracy and robustness of engine abnormal sound defect diagnosis, offering a lightweight technical solution for on-board real-time fault diagnosis and in-plant online quality inspection. By reducing engine fault-induced energy loss and spare parts waste, it further promotes energy conservation and emission reduction in the automotive industry. Quantified experimental data on fuel efficiency improvement and carbon emission reduction are provided to substantiate the ecological benefits of the proposed framework. Full article

This study explores an integrated multi-physics design approach for a high-speed Interior Permanent Magnet Synchronous Motor (IPMSM) optimized for marine diesel engine Exhaust Gas Recirculation (EGR) blower systems. To satisfy the rigorous operational demands of marine environments, an IPMSM with a rated output of 150 kW and a base speed of 9000 rpm was developed. The design validity was rigorously verified through a comprehensive multi-physics framework using the Finite Element Method (FEM), ensuring a balance between electromagnetic, thermal, and mechanical performance. The investigation established a mathematical model for the IPMSM driven by a Space Vector Pulse-Width Modulation (SVPWM) inverter, facilitating a detailed analysis of steady-state characteristics within the EGR system. To guarantee long-term reliability at high rotational speeds, the study performed an integrated thermal analysis based on precise electrical loss separation and a rotor-dynamic evaluation focusing on unbalanced vibration responses of the shaft. Finally, the proposed design was validated by integrating the IPMSM into a full-scale EGR blower system. Experimental evaluations across the entire operating range confirm that the integrated design successfully achieves the high power density and mechanical robustness required for marine diesel applications. Full article

Exhaust after-treatment (EAT) thermal management remains a critical challenge for diesel engines operating under low-load conditions, where low exhaust temperatures delay catalyst light-off and reduce emission control efficiency. This operating regime is common in marine auxiliary engines and onboard diesel generator sets during hoteling, maneuvering, and partial-electrical-load conditions. Conventional strategies such as late fuel injection or exhaust throttling can increase exhaust temperature but often result in significant fuel consumption penalties. This study numerically investigates the combined use of late exhaust valve opening (LEVO) and cylinder deactivation (CDA) to enhance EAT thermal management with a reduced fuel penalty. A six-cylinder diesel engine is analyzed at a low-load condition (1200 RPM, 2.5 bar BMEP) using a calibrated one-dimensional engine simulation model. LEVO applied to all cylinders increases exhaust temperature to approximately 250 °C, but with a considerable increase in fuel consumption. When two cylinders are deactivated and the remaining cylinders operate with LEVO, airflow and pumping losses decrease, enabling higher exhaust temperatures at comparable fuel consumption levels. Despite a 30% reduction in exhaust mass flow rate, the higher exhaust temperature dominates EAT heat transfer. Consequently, the combined strategy increases EAT heat transfer by up to 143% and achieves exhaust temperatures approaching 295 °C. These results indicate that combined valve timing and load redistribution through CDA can improve the exhaust temperature–mass flow trade-off, providing a potential pathway for enhanced EAT warm-up during low-load operation within the limitations of the numerical model. Full article

With rising global energy demand and the gradual depletion of petroleum-based resources, interest in alternative fuels for internal combustion diesel engines (ICDEs) has increased. In ICDEs, firing-related and mechanical excitations may result in adverse vibration and noise responses. This study examines whether incorporating sour cherry pit pyrolysis oil (SCPO) with n-butanol and 2-ethylhexyl nitrate (2-EHN) may reduce vibration and noise under constant-load, steady-state operating conditions compared with neat diesel (D100). For the experimental tests, five fuel types were prepared: one neat diesel fuel and four blended fuels with a constant diesel fraction of 40% and a fixed 2-ethylhexyl nitrate (2-EHN) content of 5%, while the SCPO and n-butanol fractions were varied (D40/SCPO0/B55/2-EHN5, D40/SCPO5/B50/2-EHN5, D40/SCPO10/B45/2-EHN5, and D40/SCPO15/B40/2-EHN5). Experiments were performed using a single-cylinder ICDE at a fixed load of 10 Nm under steady-state conditions at engine speeds of 1500, 1800, 2400, 3000, and 3600 rpm. For each operating condition, vibration and noise data were recorded over a 10.4 s window. Experimental findings indicate that D40/SCPO10/B45/2-EHN5 yielded the lowest mean overall RMS vibration, with a 37.5% reduction relative to neat diesel (D100), and the lowest equivalent sound level (LAeq) among the tested fuels. Under the investigated steady-state constant-load conditions, the D40/SCPO10/B45/2-EHN5 fuel blend demonstrates the potential to achieve lower measured vibration and noise levels than neat diesel. Full article

Accurate and physically consistent state prediction is essential for shipboard power systems (SPS) operating under dynamic conditions. However, purely data-driven models often exhibit degraded robustness and physically inconsistent outputs when exposed to transient disturbances or limited data coverage. To address these limitations, this paper proposes a physics-constrained hybrid prediction model that integrates a convolutional neural network–bidirectional long short-term memory (CNN–BiLSTM) architecture with wide residual connections (WRC) and a physics-constrained loss (PCL). The proposed modeling approach combines real operational measurement data with high-resolution simulation data to enhance data diversity and improve generalization capability. The CNN–BiLSTM structure captures nonlinear temporal dependencies, while the WRC preserves critical low-level transient electrical features during deep temporal modeling. In addition, multiple physical constraints, including power balance, voltage conversion relationships, and battery state-of-charge (SOC) dynamics, are incorporated into the training process to enforce physically consistent predictions. The model is validated using charging and discharging experiments on a laboratory-scale SPS under both steady-state and transient conditions. Comparative results demonstrate that the proposed approach achieves higher prediction accuracy, improved dynamic stability, and faster recovery following disturbances compared with conventional data-driven models. These results indicate that physics-constrained deep learning provides an effective and interpretable modeling framework for SPS state prediction, supporting digital twin-oriented monitoring and real-time prediction applications. Full article

Transport decarbonization and electrification are the current concepts of sustainable logistics. The European Green Deal aims to remove internal combustion engine vehicles from the roads and make the continent climate neutral by 2050. However, there is much debate about the means to achieve this goal and the rivalry between diesel and electric vehicles. This article aims to analyze the impact of the energy efficiency of diesel and electric vehicles on the sustainability of urban logistics and the benefits for the average transport user—the driver. The study uses scientific literature, statistical, comparative, SWOT analysis methods, and experimental research methods. In addition, a qualitative study was conducted with the help of experts, and the problematic relationships between diesel and electric vehicles were analyzed. The results of the study showed that even an old diesel vehicle is not inferior to a new electric vehicle in terms of energy efficiency and operation for the average user but does not meet the theoretical sustainability standards for urban logistics. Therefore, broader apolitical discussion and practical experiments are needed to ensure that the results of future research are unbiased. Full article

In this study, a novel “Physics-Informed Hybrid Machine Learning” framework was developed to model and optimize the complex combustion and carbon-based emission characteristics of Cu₂O nano-additive doped diesel fuel. To reduce reliance on purely empirical correlations, the proposed framework integrates alterations in fuel physical properties into the prediction loop, thereby enhancing physical consistency and model generalizability. The methodology comprises data pre-processing, modeling via Gaussian Process Regression (GPR) with an Automatic Relevance Determination (ARD) kernel, and multi-objective optimization using NSGA-II. Experimental tests were conducted at a constant engine speed of 2000 rpm under varying load conditions. The developed hybrid model exhibited high predictive accuracy, particularly for performance metrics and gaseous emissions (e.g., R² > 0.95 for BSFC and CO). ARD-based feature importance analysis confirmed that nano-additive dosage plays a critical role in the fine-tuning of emissions. Crucially, the optimization algorithm identified a nano-additive dosage of ~29 ppm and an engine load of 15.5 Nm as the optimal operating point for the simultaneous improvement of performance and carbonaceous emissions. This finding, exploring the unmeasured design space, demonstrates the framework’s capability to discover optimal conditions beyond discrete experimental points. Full article

This study evaluates the effects of Al₂O₃ and CeO₂ nanoparticles as additives to standard diesel and biodiesel fuels on the combustion and emissions characteristics of a CR diesel engine with split injection (pilot and main injections). Three nanoparticle dosing levels (50 ppm, 100 ppm, and 150 ppm) were compared with undoped standard diesel and biodiesel fuels. The results showed that the presence of both Al₂O₃ and CeO₂ in biodiesel increased the ignition delay of the pilot fuel by about 8.0% at low load and about 3.5% at high load. The addition of both nanoparticles to diesel and biodiesel fuels had an insignificant effect on the main injection fuel’s ignition delay, MBF50 position and combustion duration. The thermal efficiency was up to 1.0% lower. Al₂O₃ additive in diesel had no significant effect on NO_x emissions. CO emissions were higher by 4.4–7.5% in most cases. The Al₂O₃ additive in biodiesel reduced NO_x emissions by an average of 38%, 17.1%, and 9.4% at low, medium, and high engine loads, respectively. The reduction in CO emissions averaged 15%. The addition of CeO₂ nanoparticles to diesel fuel reduced NO_x emissions by 22.5%, 8.5%, and 3.1% on average across the corresponding load ranges. When the engine was operated on CeO₂-doped biodiesel, NO_x emissions were lower by an average of 25.7%, 9.6%, and 2.5% at low, medium, and high loads, respectively. Adding CeO₂ nanoparticles to diesel fuel increased CO emissions, whereas adding them to biodiesel significantly reduced CO emissions. Full article

Deep neural networks (DNNs) have demonstrated remarkable capability in accurately predicting equilibrium combustion products and thermodynamic properties of diesel combustion. However, the lack of awareness of uncertainty and interpretability has limited their scientific credibility and practical application. In this work, an enhanced DNN framework with uncertainty quantification and explainability is developed. The model achieves high accuracy across all outputs, with R² values exceeding 0.99 for major thermodynamic variables. In this model, Monte Carlo dropout sampling is used to estimate epistemic uncertainty, and prediction confidence intervals are analyzed across all species and thermodynamic outputs, revealing strong correlations for major components. Model explainability is further explored using Shapley additive explanations (SHAP), which attribute the influence of equivalence ratio, temperature, and pressure on each predicted species and combustion characteristics. The combined uncertainty quantification and explainability framework not only enhances confidence in DNN combustion models but also provides physical insight into the relationships between input conditions and equilibrium thermochemistry that are learned by the DNN. Full article

Biodiesel composition largely affects its combustion and emission performance in diesel engines, while the pilot–main injection strategy has the potential to simultaneously improve engine efficiency and alleviate the soot–NO_x trade-off. Accordingly, soybean oil (SO), animal fat (AF), and waste cooking oil (WCO) biodiesels were numerically investigated under single injection and pilot–main double injection strategies with pilot energy ratios of 5.6%, 10%, and 15% over a range of main injection timings. The CFD framework was validated against the experimental in-cylinder pressure and AHRR, showing good agreement across the tested operating conditions. The results show that advancing the injection timing increases the AHRR peak and MPRR, whereas the pilot–main injection strategy reduces the AHRR peak and generally advances combustion phasing. A stable CA10 at injection timing ranges of 350–358 °CA persists across biodiesels and injection strategies. Emissions correlate with MICT, where NO_x increases while soot decreases with the rise in MICT, suggesting an intermediate MICT window for a balance between efficiency and emissions. Furthermore, at the respective highest gross ITE of biodiesels, SO provides the most favorable combination of MPRR and efficiency, whereas WCO shows lower NO_x but stronger indicators of incomplete combustion. Full article

This paper presents the design of a sensor system for remote measurements of exhaust emissions from automotive combustion engines. The system’s purpose is to determine the likelihood of a given vehicle’s potential harmfulness to the environment. This system, if implemented, could detect vehicles posing a threat to the environment in road traffic. A remote measurement system can be installed in the front of a measuring vehicle driving behind the vehicle being diagnosed. This approach allows for rapid road testing of multiple vehicles while they are operating in real-world conditions where engines can emit the highest levels of undesirable pollutants. Exceeding emission standards may be related to modifications made to the vehicle’s exhaust gas aftertreatment systems, engine wear, or malfunctions of engine-related systems such as the diesel particulate filter (DPF) or catalytic converter. Toxic and undesirable substances include carbon monoxide (CO), hydrocarbons (HC), nitrogen oxides (NO_x), carbon dioxide (${\mathrm{C}\mathrm{O}}_2$), and particulate matter (PM) particles. The main goal of the measurements is to identify vehicles that potentially pose a threat to the environment during normal operation. The sensor system consists of several types of sensors utilizing various physical and chemical phenomena, with particular emphasis on their low cost and easy availability. The measurement unit utilizes MEMS technology, photoacoustic spectroscopy, electrochemical methods, light absorption and scattering, spectrophotometry, and electro-optical detection. Full article

Dual-fuel combustion is often proposed for diesel engines as a means to partially replace conventional diesel with cleaner and/or more sustainable alternatives, such as those derived from green hydrogen. However, the low reactivity of these fuels (i.e., methane, hydrogen, and ammonia) often leads to prolonged ignition delay (ID) and combustion instability. This challenge could potentially be overcome using nanomaterials, which are additives that could improve reactivity and compensate for autoignition deficiencies. Thus, this study evaluates the effect of carbon nanotubes (CNTs) dispersed in diesel fuel on the autoignition process under dual-fuel operation. CNTs were dispersed at a concentration of 100 mg/L and stabilized with surfactant sodium dodecylbenzene sulfonate (SDBS). The resulting nanofuels were then tested in a constant volume combustion chamber (CVCC) using methane, hydrogen, and ammonia as secondary fuels across various energy substitution ratios and temperatures (535 °C, 590 °C and 650 °C). The results show that the impact of CNTs on ID is negligible, especially at high temperatures. At the lowest tested temperature (535 °C) and 40% methane substitution ratio, only slight reductions in ID were obtained. Nevertheless, this effect is less significant at higher temperatures (590 °C and 650 °C). Regarding pressure gradient, the addition of CNTs and SDBS generally induced a decrease in pressure-peak of up to 15%. This trend is attributed to the trapping of fuel droplets within the CNT structures, which creates a physical barrier that delays vaporization. Results confirm that autoignition, which is expected to be the main phenomenon influenced by CNT addition, is not enhanced. Full article

This study explores the viability of condensate petroleum, an ultra-light hydrocarbon derived from natural gas production, as an alternative diesel engine fuel. The researchers tested six different fuel blends, increasing the condensate volume by 10% increments, in a compression ignition engine under three distinct load conditions (25%, 50%, and 75%) to evaluate both combustion characteristics and emission performance. The results demonstrate that condensate blends significantly enhance key combustion parameters. The heat release rate, in-cylinder pressure, and in-cylinder temperature all increased, with the highest heat release rate improvement of 35.6% observed at a 75% load using a 60% condensate petroleum blend. However, increasing the condensate ratio also extended ignition delay times and raised the ringing intensity, which peaked with a 34.7% increase at a 25% load. Brake thermal efficiency improved at lower and medium loads—achieving a maximum 11.2% increase with the 50% condensate petroleum blend at 50% load—but decreased when the engine reached 75% load. In terms of environmental impact, the condensate blends proved largely beneficial. Carbon monoxide emissions dropped by 57.9% (at 75% load, 60% condensate petroleum), smoke opacity decreased by 72.6% (at 25% load, 40% condensate petroleum), and hydrocarbons fell by 34.4% (at 50% load, 60% condensate petroleum). The primary drawback was that nitrogen oxide emissions worsened, increasing by 20.4% at 75% load with the 50% condensate petroleum blend. Overall, the study concludes that the effects of condensate petroleum are highly acceptable, making it a promising alternative fuel and additive for diesel engines. Full article

This work investigates the influence of hybrid NiO–SiO₂ nanoparticles on the engine behavior of biodiesel derived from waste sunflower oil and evaluates the experimental outcomes using a data-driven modeling approach. Biodiesel was produced via transesterification and doped with nanoparticles at concentrations of 50, 75, and 100 ppm. Performance and emission tests were conducted on a single-cylinder diesel engine operating at constant speed under varying loads. Specific fuel consumption, brake thermal efficiency, CO, HC, NO_x, smoke opacity, and exhaust gas temperature were recorded and analyzed. The incorporation of nanoparticles improved combustion quality and contributed to substantial reductions in harmful emissions. The WSOB20 blend containing 100 ppm NiO–SiO₂ provided the most balanced results, decreasing CO, HC, and smoke emissions by 39.50%, 39.40%, and 35.20%, respectively, relative to diesel fuel, while preserving competitive thermal efficiency. A linear regression model developed for CO prediction produced a low mean squared error (1.08 × 10⁻⁵), indicating strong predictive capability. The findings confirm that hybrid nanoparticle additives can enhance biodiesel performance while supporting accurate emission forecasting. Full article

This study proposes a method to recover liquefied natural gas (LNG) cold energy from the fuel gas supply system (FGSS) of a two-stroke ME-GI dual-fuel (DF) marine engine to enhance energy utilization efficiency. LNG cold energy was employed to reduce the scavenging air temperature (SAT) through a CaCl₂-based secondary refrigerant loop integrated into the engine cooling system. Thermodynamic analysis showed that approximately 12.3% of the required scavenging air cooling heat flux can be recovered at full load. Transient crank-angle-resolved CFD simulations, validated against experimental data (maximum deviation < 8%), were conducted to evaluate combustion and emission impacts under varying SAT conditions. Reducing SAT from 37 °C to 17 °C in DF mode increased indicated mean effective pressure (IMEP) by approximately 3.8%, reduced specific gas consumption by 3.7%, and significantly decreased NO emissions by up to 36.5% and soot emissions by 47.6%, while CO₂ emissions decreased by 1.8%. Considering both performance enhancement and emission reduction, operating the engine in DF mode with SAT controlled at approximately 17 °C is recommended. The proposed system demonstrates a practical pathway for improving thermal efficiency and reducing greenhouse gas (GHG) emissions in LNG-fueled marine propulsion systems. Full article