With the long-term operation of ships, the performance of marine diesel engines gradually declines due to the wear of internal moving components, increasing the risk of potential failures. Fuel consumption is a critical indicator for assessing engine operating conditions, and accurately predicting baseline fuel consumption under normal operating conditions is essential for evaluating ship energy efficiency and conducting fault diagnosis. To address common issues in marine engine operational data, such as noise pollution, missing values, inconsistent scales, and feature redundancy, a Diesel Engine Data Enhancement and Optimization Framework (DEOF) was developed to systematically improve data quality. Furthermore, to overcome the limitations of existing models, such as insufficient prediction accuracy and poor stability under complex operating conditions, a Meta-learning Diffusion Residual Attention Network (MD-RAN) is proposed. This approach leverages the strengths of diffusion models in nonlinear generative modeling, integrates meta-learning mechanisms to enhance task adaptation speed, employs multi-head attention modules to strengthen feature extraction, and incorporates dynamic residual connections to improve training stability and flexibility. The data used in this study were collected from real-world operations of ocean-going vessels, ensuring high representativeness. This paper systematically benchmarks the proposed model with the traditional learning model. The results are verified to be effective. The MD-RAN algorithm is significantly better than the original model in terms of prediction accuracy, stability, and nonlinear expression ability. The R² value can reach 0.9853, and the RMSE and MAE are as low as 1.5801 and 1.1879, respectively. Its feasibility will be further evaluated in practical applications in the future. This study not only provides a systematic data-driven modeling framework, offering technical insights for constructing high-quality datasets, but also establishes a novel generative modeling approach for marine diesel engine fuel consumption prediction, providing robust support for intelligent engine maintenance and energy efficiency optimization.

This paper has proposed a single objective optimization approach using particle swarm optimization (PSO) algorithm to minimize fuel consumption (CO ₂ emission) or NOx emissions. The proposed optimization approach is conducted based on a zero-dimensional engine model, which is validated with engine experiment data. The compression ratio (CR), start of injection timing (SOI) and exhaust valve closing timing (EC) are selected as optimization variables based on single-parameter sensitivity analysis, as they have significant impact on engine specific fuel oil consumption (SFOC) and NOx emissions. A diesel engine performance optimization program is established, with the minimum SFOC or NOx emissions as the optimization objective, respectively. During the optimization process, constraints on the maximum combustion pressure (P _max ) and the maximum pressure rise rate are considered. A two-stroke diesel engine MAN’s 7G80ME-C is used for case analysis. The optimization results show that a maximum fuel reduction of 2.26% can be achieved at 100% load when SFOC is selected as objective. On the other hand, NOx emissions can be reduced up to 24.61% when minimizing NOx is selected as the objective.

Mass and volume penalties associated with fuel cell integration have hindered the manufacturing and integration of hybrid power systems into civil aircraft. To address this issue, a novel scheme, a turbofan engine integrated with solid oxide fuel cells, is proposed in this paper, which has the advantage of high thermal efficiency and propulsion efficiency, where the electricity produced by the solid oxide fuel cell (SOFC) is used to drive the ducted fan. Then, a multidimensional model that accounts for mass, volume, and thermodynamic performance is established. The equivalence ratio and split ratio have a significantly stronger influence on the volume and mass ratios than the current density. These two parameters directly determine the power ratio between the fuel cell and the gas turbine. An increase in the power ratio leads to a simultaneous rise in both the volume and mass ratios; however, the overall efficiency cannot be continuously improved, with its maximum value being approximately 63% without mass and volume constraints. During the multi-objective optimization process, the weight and volume of the hybrid engine have a linear relationship with the overall efficiency when the efficiency is lower than 50%. However, both mass and volume exhibit exponential increasing trends as the overall efficiency is over 50%. This is caused by the nonlinear change in the fuel cell reaction area; as polarization loss decreases, the thermal efficiency of the hybrid system improves. The maximum overall efficiency is taken as the optimization objective, with the constraints that the mass fraction of the fuel cell does not exceed 0.4 of the total engine mass and that the volume fraction of the fuel cell stack is less than 0.2 of the gas turbine engine volume. Compared with the CFM56 3C1, the thermal efficiency for the novel hybrid engine is increased by 21%, and its propulsion efficiency is increased by 42%.