Search Results (4)

Accurate prediction of fuel consumption is critical for achieving efficient and low-carbon ship operations. However, the variability of the marine environment introduces significant challenges, as it leads to dynamic changes in monitoring data, complicating real-time and precise fuel consumption prediction. To address this issue, the authors proposed an incremental learning-based prediction framework to enhance adaptability to temporal dependencies in fuel consumption data. The framework dynamically adjusts a dual adaption mechanism for input features and target labels while incorporating rolling retraining to enable continuous model updates. The effectiveness of the proposed approach was validated using a real-world dataset from an LPG carrier, where it was benchmarked against conventional machine learning models, including Random Forest (RF), Linear Regression (LR), Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP). Experimental results demonstrate that the proposed approach could significantly improve prediction accuracy in both offline and online scenarios. In offline mode, the proposed framework improves the R² of various machine learning models by at least 21.97%. In online mode, the proposed method increases R² by at least 17.97%. This work provides a new solution for real-time fuel consumption prediction in dynamic marine environments. Full article

The present work compares the performance and emissions of a compression ignition (CI) engine using dual-mode LPG at varying flow rates and an oxygenated biodiesel mix (B20). The experimental investigation is carried out on LPG flow rates (0.1, 0.3, and 0.5 kg/h) and replacing the diesel with oxygenated B20, affecting engine performance and emissions under various load circumstances while maintaining engine speed. The study demonstrates the potential of the artificial neural network (ANN) in accurately forecasting the performance and emission characteristics of the engine across different operating conditions. The ANN model’s high accuracy in correlating experimental results with predicted outcomes underscores its potential as a dependable instrument for optimizing fuel parameters. The results show that LPG and oxygenated B20 balance engine performance and emissions, making CI engine functionality sustainable. A biodiesel blend containing diethyl ether (B20 + 2%DEE) exhibits slightly reduced brake thermal efficiency (BTE) at lower brake power (BP); however, it demonstrates advantages at higher BP, with diethyl ether contributing to improved ignition quality. The analysis indicates that the average NO_x emissions for B20 + 2%DEE at flow rates of 0.1 kg/h, 0.3 kg/h, and 0.5 kg/h are 29.33%, 28.89%, 48.05%, and 37.48%, respectively. Consequently, selecting appropriate fuel and regulating the LPG flow rate is critical for enhancing thermal efficiency in a dual-fuel engine. Full article

The present work is focused on the assessment of the performance and fine particulate matter emissions (PM_2.5) of a turbocharged four-cylinder direct injection diesel engine operating under dual-fuel mode with Liquefied Petroleum Gas (LPG). For load levels of 30%, 60% and 100%, measurements were taken, keeping the engine speed constant at 2200, 2500 and 3200 rpm, while the engine knock detonation was detected through a non-invasive internal system. According to experimental measurements, the abnormal knock combustion occurred at full load operation with a maximum LPG energy fraction of ~60%. The brake fuel conversion efficiency increased by 2.6% with an LPG energy fraction of 10%, where a fuel saving of 11.9% was achieved with respect to the diesel-only operation. The reduction of diesel consumption was around 50% with respect to 100% diesel operation at full load operations, where the highest brake fuel conversion efficiency was achieved. The brake fuel conversion efficiency decreased as LPG addition increased for all the engine loads. Regarding emissions, PM_2.5 decreased with the addition of LPG. However, HC and CO emissions increased as LPG injection was higher. NO_x emissions and exhaust gas temperatures were reduced for operation with higher LPG fractions, except for full load levels at 2200 and 2500 rpm. Full article

This work discusses sensing properties of a long-period grating (LPG) and microcavity in-line Mach–Zehnder interferometer (µIMZI) when both are induced in the same single-mode optical fiber. LPGs were either etched or nanocoated with aluminum oxide (Al₂O₃) to increase its refractive index (RI) sensitivity up to ≈2000 and 9000 nm/RIU, respectively. The µIMZI was machined using a femtosecond laser as a cylindrical cavity (d = 60 μm) in the center of the LPG. In transmission measurements for various RI in the cavity and around the LPG we observed two effects coming from the two independently working sensors. This dual operation had no significant impact on either of the devices in terms of their functional properties, especially in a lower RI range. Moreover, due to the properties of combined sensors two major effects can be distinguished—sensitivity to the RI of the volume and sensitivity to the RI at the surface. Considering also the negligible temperature sensitivity of the µIMZI, it makes the combination of LPG and µIMZI sensors a promising approach to limit cross-sensitivity or tackle simultaneous measurements of multiple effects with high efficiency and reliability. Full article