Scholarlink

Purpose <p>This study investigated the impact of antioxidants diphenylamine (DPA) and nanoparticles ceria (CeO<sub>2</sub>) on engine block vibration in a B30 biodiesel blend. The influence of critical input parameters such as compression ratio (CR), fuel injection pressure (FIP), load, and exhaust gas recirculation (EGR-HOT) on vibration behavior was analyzed. Response Surface Methodology (RSM) and Machine Learning (ML) algorithms were employed to predict experimental root mean square (RMS) acceleration values.</p> Method <p>Experiments were conducted using diesel and biodiesel blends (B30, B30+DPA100, and B30+DPA50+CeO<sub>2</sub>50) under varying conditions of CR, FIP, load, and EGR-HOT. ML algorithms, including Support Vector Machines (SVM), Multi-Layer Perceptron (MLP), and k-Nearest Neighbors (k-NN), were employed to predict RMS acceleration. Model performance was evaluated using metrics such as coefficient of determination (R<sup>2</sup>), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and relative RMSE (rRMSE).</p> Results <p>The study finds significant reductions in mean RMS acceleration for B30, B30+DPA100, and B30+DPA50+CeO<sub>2</sub>50 compared to diesel across load, CR, FIP, and EGR-HOT variations, demonstrating the effectiveness of these fuel blends in minimizing engine vibration. For load variation, the reductions were 12.32% for B30, 7.15% for B30+DPA100, and 9.12% for B30+DPA50+CeO250 as compared to diesel. Similarly, with CR variation, the decreases were 12.14% for B30, 7.04% for B30+DPA100, and 8.99% for B30+DPA50+CeO<sub>2</sub>50 as compared to diesel. For FIP variation, the mean RMS acceleration reductions were 10.28% for B30, 5.97% for B30+DPA100, and 7.61% for B30+DPA50+CeO<sub>2</sub>50 as compared to diesel. Finally, with EGR-HOT, the mean RMS acceleration reductions were 10.33% for B30, 5.99% for B30+DPA100, and 7.65% for B30+DPA50+CeO<sub>2</sub>50 as compared to diesel. Furthermore, findings show that the k-NN model outperforms SVM and MLP models, exhibiting superior R<sup>2</sup> values and consistently lower error metrics.</p> Conclusion <p>Overall, these findings emphasize the efficacy of DPA and CeO<sub>2</sub> additives in mitigating engine vibrations in biodiesel blends and k-NN algorithm emerged as a reliable tool for predicting RMS acceleration.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Experimental Investigation and Prediction of Vibrational Characteristics in Diesel Engines with Diphenylamine Antioxidant and Ceria Nanoparticle Enriched Biodiesel

  • 期刊论文

  • 发表日期：

    2025年1月9日

查看全文
  • Vijay Kumar,
  • Akhilesh Kumar Choudhary

摘要

Purpose

This study investigated the impact of antioxidants diphenylamine (DPA) and nanoparticles ceria (CeO2) on engine block vibration in a B30 biodiesel blend. The influence of critical input parameters such as compression ratio (CR), fuel injection pressure (FIP), load, and exhaust gas recirculation (EGR-HOT) on vibration behavior was analyzed. Response Surface Methodology (RSM) and Machine Learning (ML) algorithms were employed to predict experimental root mean square (RMS) acceleration values.

Method

Experiments were conducted using diesel and biodiesel blends (B30, B30+DPA100, and B30+DPA50+CeO250) under varying conditions of CR, FIP, load, and EGR-HOT. ML algorithms, including Support Vector Machines (SVM), Multi-Layer Perceptron (MLP), and k-Nearest Neighbors (k-NN), were employed to predict RMS acceleration. Model performance was evaluated using metrics such as coefficient of determination (R2), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and relative RMSE (rRMSE).

Results

The study finds significant reductions in mean RMS acceleration for B30, B30+DPA100, and B30+DPA50+CeO250 compared to diesel across load, CR, FIP, and EGR-HOT variations, demonstrating the effectiveness of these fuel blends in minimizing engine vibration. For load variation, the reductions were 12.32% for B30, 7.15% for B30+DPA100, and 9.12% for B30+DPA50+CeO250 as compared to diesel. Similarly, with CR variation, the decreases were 12.14% for B30, 7.04% for B30+DPA100, and 8.99% for B30+DPA50+CeO250 as compared to diesel. For FIP variation, the mean RMS acceleration reductions were 10.28% for B30, 5.97% for B30+DPA100, and 7.61% for B30+DPA50+CeO250 as compared to diesel. Finally, with EGR-HOT, the mean RMS acceleration reductions were 10.33% for B30, 5.99% for B30+DPA100, and 7.65% for B30+DPA50+CeO250 as compared to diesel. Furthermore, findings show that the k-NN model outperforms SVM and MLP models, exhibiting superior R2 values and consistently lower error metrics.

Conclusion

Overall, these findings emphasize the efficacy of DPA and CeO2 additives in mitigating engine vibrations in biodiesel blends and k-NN algorithm emerged as a reliable tool for predicting RMS acceleration.