Qinming Tan

To address the problem of the fault mode discrimination accuracy of marine rotating machinery under the conditions of noise interference and limited training samples, a fault diagnosis method based on a particle swarm optimization-driven kernelized cascade forest is proposed. This method introduces RBF kernel mapping in the hierarchical structure of the cascade forest and uses PSO to conduct global optimization of the key parameters of the kernel function. Based on the marine fan test platform driven by a three-phase asynchronous motor, the frequency domain features are extracted as the model input, and the predicted probability distribution is utilized during the hierarchical training process. The experimental results show that in a noise-free environment and a noisy environment with Gaussian white noise, the test accuracy of the diagnostic accuracy rate is 98.05% and 97.10%, respectively, with a performance decrease of only 0.95%; it still maintains stable recognition accuracy under the condition of small samples, demonstrating superior small-sample learning ability compared to the benchmark method. Compared with the forest-based baseline method, the proposed method can reduce the performance degradation caused by noise and effectively improve the diagnostic performance of the model under noise and small-sample conditions, and this method may become a potential solution for fault intelligent diagnosis in marine rotating machinery.

Large eddy simulation coupled with a porous medium model and a sliding mesh technique was employed to investigate the flow control mechanism of partially porous blade tips in a pump-jet propulsor (PJP). Two porous tip configurations with 1/3 and 2/3 coverage were examined, and the numerical methodology was validated against published reference data with errors below 5%. The results show that the porous medium suppresses the tip leakage jet by introducing viscous and inertial resistance, thereby weakening the development of the tip leakage vortex (TLV). Compared with the conventional rotor, the TLV core strength is reduced by more than 30%, while the pressure fluctuation amplitudes decrease by 5.07%–12.93% and 27.53%–33.02% for the 1/3 and 2/3 porous tip configurations, respectively. Increasing the porous coverage enhances vortex suppression but also increases thrust loss from approximately 6%–9% to 16%–22%. Under oblique inflow conditions, the 1/3 porous tip configuration achieves the best balance between vortex control and propulsion performance. At θ = 10°, pressure fluctuation amplitudes are reduced by more than 10%. These findings demonstrate the potential of partially porous blade tips as an effective passive flow control strategy for mitigating TLV-induced flow instability in PJPs.

To address the issue of insufficient fault mode discrimination accuracy and stability of rotating machinery under the conditions of noise interference and limited training samples, a fault diagnosis method based on particle swarm optimization-driven kernelized cascade forest is proposed. This method introduces RBF kernel mapping in the hierarchical structure of the cascade forest and uses PSO to globally optimize the key parameters of the kernel function, enhancing the model's ability to express nonlinear and non-stationary vibration features and suppressing noise interference. Based on the test platform of marine fans driven by three-phase asynchronous motors, frequency domain features are extracted using the Fast Fourier Transform as the model input. During the hierarchical training process, the prediction probability distribution is utilized to achieve progressive enhancement of features and adaptive termination. The experimental results show that under noise-free conditions, the diagnostic accuracy of this method on the test set reaches 98.05%; in a Gaussian white noise environment with a signal-to-noise ratio of 20 dB, the performance decline is less than 1%; when the training sample proportion is reduced to 10%, the test accuracy still reaches 90.25%. Comparative results with other algorithms indicate that the collaborative design of kernelized feature mapping and cascade structure can effectively improve the diagnostic robustness of the model under noise and small sample conditions, providing a stable and feasible technical solution for intelligent fault diagnosis of rotating machinery under complex working conditions.

This study analyzes Diesel Oxidation Catalyst (DOC) and Carbon Diesel Particulate Filter (CDPF) after-treatment systems integrated into a WARTSILA W20DF marine dual-fuel engine. The CDPF was coated with a non-precious metal catalyst whose catalytic redox performance improved with increasing temperature. The carbon particulate matter combustion reached up to 12.5 mg/s at 800 K and over 20 mg/s at 900 K. Then, the W20DF running at 230 kW, 450 kW, 680 kW, and 810 kW with 1000 rpm; a Tisch 10-8xx; and an AVL SPC 478 were used to sample and analyze the carbon particulate matter (CPM) before and after DOC + CDPF. The gaseous emissions (O₂, CO₂, CO, HC, NOx, and NO₂) were analyzed with the flue gas analyzer AVL i60. The results show that the collected carbon particulate matter simultaneously became darker as the load decreased. This study finds that the maximum amount of CPM per unit volume of exhaust gas occurs under 50% working conditions and the lowest amount under 90% working conditions. After DOC + CDPF treatment with a non-precious metal coating, the CPM was reduced by about 50%. Furthermore, this type of catalyst’s efficiency rises with the temperature increase. The CPM combustion efficiency reached up to 20 mg/s at 900 K. The other gas components in the exhaust gas before and after DOC + CDPF also changed. These research results have a significant reference value for DOC + CDPF optimization to decrease the carbon particulate matter in marine engines.

Particulate matter (PM) is a major pollutant in the exhaust of marine diesel engines, which seriously endangers human health and the atmospheric environment, and how to reduce particulate matter emissions from marine engines has become a key research direction in the field of environmental protection and diesel engines. In this study, we analyzed the components and sources of PM from marine engines and conducted tests on the performance of Wärtsilä 20DF Diesel Particulate Filter (DPF) catalysts to verify the capture efficiency, gaseous pollutant removal rate, regeneration effect and the relationship between carbon loading and pressure loss of DPF catalysts in the context of Tier III emission regulations. The results showed that PM emissions of 20DF in diesel mode after adding the DPF system meet the requirements of the regulatory limit, but the pressure drop of the engine increases after adding the DPF system. Therefore, numerical simulation was used to optimize the DPF structure by evaluating the system velocity field, flow field distribution uniformity and system pressure drop to improve the pressure drop.

The influence of gas intake temperature, composition and the volume concentration of each gas component on diesel engine combustion, emission and the output power was studied by building a calculation model of the B&W 6S35ME-B9 marine two-stroke low-speed diesel engine, followed by a comprehensive optimization exploration. The results showed that under 295 K and 18.5% O₂ of intake gas, the engine’s NOx emission is only 4.5 g/kWh and reduced to 58% from the normal air gas intake condition. Moreover, their power output is very similar. In addition, the effect of CO₂ or H₂O added into the intake of the diesel engine on the performance of the diesel engine can be compensated by reducing the intake temperature. At the intake temperature of 295 K, the engine’s NOx emission with 20.58% O₂, 77.42% N₂ and 2% H₂O is 8.62 g/kWh, and 9.06 g/kWh under 20.79% O₂, 78.21% N₂ and 2% CO₂. It is lower than 11.77 g/kWh, which is under normal intake conditions (315 K, 21%O₂ and 79%N₂). The power output is also similar to the normal intake condition. Therefore, the comprehensive optimization of gas intake temperature, composition and concentration can effectively optimize the diesel engine’s performance in terms of combustion, emission and power output. The research results have an important reference value for the optimization of diesel engine performance.

This paper studies the combustion performance of diesel in constant volume combustion vessels under different conditions of mixed low-nitrogen (O₂ and N₂) or non-nitrogen (O₂ and CO₂) in varying proportions. The high-speed camera is used to shoot the combustion flame in the constant volume combustion vessel. The process and morphology of the combustion flame are amplified in both time and space to study and analyze the effects of different compositions and concentrations in gases on the combustion performance of diesel and conduct a study on the contributory factors in the performance of diesel with no nitrogen. According to the study, in the condition of low nitrogen, the O₂ concentration is more than 60%, the ignition delay period is shortened, the combustion flame is bright and slender, it spreads quickly, and the blue flame appears when the O₂ concentration reaches 70%; While for nitrogen-free combustion, only when the O₂ concentration reaches 30% is the combustion close to the air condition; when the O₂ concentration reaches 40%, the combustion condition is optimized obviously and the combustion flame is relatively slender compared to the air working condition. Similarly, with the increase of the O₂ concentration, the ignition delay period of nitrogen-free diesel is shortened, the duration is extended, and the combustion performance is optimized. In addition, when the O₂ concentration reaches 50%, with the decrease of the initial temperature, the ignition delay period is prolonged, and the duration is shortened obviously. When the temperature is lower than 700 K, there is no ignition. The increase of the diesel injection pressure is beneficial to optimize the ignition performance of diesel non-nitrogen combustion and shorten its ignition delay period and combustion duration. Related research has important guiding significance to optimize nitrogen-free combustion technology, which produces no NOx of the diesel engine.

With the implementation and expansion of international sulfur emission control areas, effectively promoted the marine low sulfur diesel fuel (MLSDF) used in marine diesel engines. In this study, a large low-speed, two-stroke, cross-head, common rail, electronic fuel injection marine diesel engine (B&W 6S35ME-B9) was used for the study. According to diesel engine’s propulsion characteristics, experiments were launched respectively at 25%, 50%, 75%, 100% load working conditions with marine low sulfur diesel fuel to analyze the fuel consumption, combustion characteristics and emissions (NOx, CO ₂ , CO, HC) characteristics. The results showed that: Marine diesel engine usually took fuel injection after top dead center to ensure their safety control NOx emission. From 25% to 75% load working condition, engine’s combustion timing gradually moved forward and the inflection points of pressure curve after top dead center also followed forward. While it is necessary to control pressure and reduce NOx emission by delaying fuel injection timing at 100% load. Engine’s in-cylinder pressure, temperature, and cumulative heat release were increased with load increasing. Engine’s CO ₂ and HC emissions were significantly reduced from 25% to 75% load, while they were increased slightly at 100% load. Moreover, the fuel consumption rate had a similar variation and the lowest was only 178 g/kW·h at 75% load of the test engine with MLSDF. HC or CO emissions at four tests’ working conditions were below 1.23 g/kW·h and the maximum difference was 0.2 or 0.4 g/kW·h respectively, which meant that combustion efficiency of the test engine with MLSDF is good. Although the proportion of NOx in exhaust gas increased with engine’s load increasing, but NOx emissions were always between 12.5 and 13.0 g/kW·h, which was less than 14.4 g/kW·h. Thus, the test engine had good emissions performance with MLSDF, which could meet current emission requirements of the International Maritime Organization.

This study adopts 25 kHz ultrasonic wave and 0.4T magnetization field in online optimization of Marine Heavy Fuel Oil (MHFO). It aims at studying the feasibility of optimizing the combustion and emission performance of two-stroke marine diesel engine with one kind of MHFO with ultrasonic wave and magnetization field. The results show that engine’s fuel consumption could reach 222 g/kW· h under 25% of engine rated power, while this consumption reduces to 188 g/kW· h under 90% of engine rated power. It can be seen that the rate of fuel consumption decreased by about 15% between 25% and 90% of engine rated power. The improved combustion performance under high operating load conditions produced low carbon monoxide (CO) and carbon dioxide (CO₂) emissions. The rise of in-cylinder temperature promoted the formation of nitrogen oxides (NOx), in which nitrogen oxide (NO) accounted for about 99.5% of the NOx composition. Through the optimization of online MHFO with ultrasonic waves and magnetization field device under 25% of engine rated power, the engine’s fuel consumption rate decreased by about 1.5%. This shows that CO and CO₂ emissions have been effectively reduced. However, the NOx emissions increased slightly. The CO, CO₂ and NOx emissions decreased when the engine’s power increased form 50% to 75%, In this operating range, its fuel saving decreased from 0.4% to 0.08%. At 90% of engine rated power the volume concentration of CO emissions and fuel consumption rate increased. However, its CO₂ emissions decreased, while the NOx emissions remained unchanged.

Nowadays, steam turbine propulsion system is no longer the only one for LNG carriers. With the rapid rise of fuel price, the selection of LNG carrier propulsion systems from the profit point of view has become more and more important. With a purely mathematical solution and ¿the charter nomogram¿, the paper analyses and compares the profitability of two different LNG propulsion systems ¿A¿ and ¿B¿. It can supply some theoretical guidance for the future building of our country LNG carriers.