Al Awadh Mohammed, Gulbarga Mohammad Imtiyaz
Department of Industrial Engineering, College of Engineering, King Khalid University, P.O. Box 394, Abha, 61421, Saudi Arabia.
Center for Engineering and Technology Innovations, King Khalid University, Abha, 61421, Saudi Arabia.
Sci Rep. 2025 Apr 28;15(1):14854. doi: 10.1038/s41598-025-98201-x.
Artificial intelligence-based technologies are rapidly advancing and significantly influencing the engineering sector, particularly in the automotive industry, through AI-driven neural network tools and Sankey diagrams. Meanwhile, the depletion of fossil fuels and rising emissions have pushed global efforts towards renewable and clean fuel solutions. Hydrogen, as a key clean fuel, has garnered considerable research interest. Combining hydrogen with biomass-derived fuels has gained attention due to its dual benefits of addressing biomass waste disposal and alleviating hydrogen storage and safety concerns. This study focuses on the production of aquatic plant oil (duckweed bio-oil) and its combination with hydrogen gas, evaluating their effects on the performance of a Reactivity Controlled Compression Ignition (RCCI) engine. The results revealed that the H40 blend demonstrated a 1% higher brake thermal efficiency (BTE) than diesel, along with emission reductions of 40% for HC, 6% for NOx, 27% for CO, and 14% for smoke. The results were further validated using an Artificial Neural Network (ANN) and a Sankey diagram. The ANN achieved low RMSE values (0.9965-0.9996) and MPAE values within 4%, while the Sankey diagram effectively illustrated energy distribution with minimal loss. These findings highlight the potential of hydrogen-enriched fuels for future internal combustion engines.
基于人工智能的技术正在迅速发展，并通过人工智能驱动的神经网络工具和桑基图对工程领域，特别是汽车行业产生重大影响。与此同时，化石燃料的枯竭和排放量的增加推动了全球对可再生和清洁燃料解决方案的努力。氢气作为一种关键的清洁燃料，已引起了相当大的研究兴趣。将氢气与生物质衍生燃料相结合，因其在解决生物质废物处理以及缓解氢气储存和安全问题方面的双重好处而受到关注。本研究着重于水生植物油（浮萍生物油）的生产及其与氢气的结合，评估它们对反应控制压缩点火（RCCI）发动机性能的影响。结果表明，H40混合燃料的制动热效率（BTE）比柴油高1%，同时碳氢化合物（HC）排放量减少40%，氮氧化物（NOx）排放量减少6%，一氧化碳（CO）排放量减少27%，烟雾排放量减少14%。使用人工神经网络（ANN）和桑基图对结果进行了进一步验证。人工神经网络实现了较低的均方根误差（RMSE）值（0.9965 - 0.9996）和4%以内的平均绝对百分比误差（MPAE）值，而桑基图有效地展示了能量分布且损失最小。这些发现突出了富氢燃料在未来内燃机中的潜力。
