DoctorateOpen Access

Experimental analysis of performance, emission and combustion characteristics of a diesel engine at different operating parameters of fuel obtained from waste industrial oils and waste plastics and optimisation with artificial neural network method

2023
0 views
0 downloads
Advisor: Doç. Dr. Selman Aydın

Abstract (EN)

In this study, the aim is to produce diesel-like fuel by pyrolytic distillation method from waste industrial and waste plastic oils, which cause permanent damage to the environment and nature for many years and which can be recycled back to energy by recycling, and to examine the performance, combustion, and emission parameters of diesel fuel mixtures at different ratios and to model them with artificial neural networks (ANN). As a result of the tests performed on both fuels produced, a two-stage desulfurization process was carried out due to the high sulfur content, especially in diesel-like plastic (DL-PF) fuel. In the first stage, no change was observed in diesel-like industrial (DL-IF) fuel as a result of the application of 10% perlite, 10% calcium oxide, and 10% zeolite in both fuels in the basic desulphurization process, while in DL-PF, 25%. 3% of the sulfur content was removed in the 10% perlite application and used as a catalyst in this fuel. While the sulfur value of diesel-like industrial fuel is within the limit values of standard diesel fuel, since DL-PF is much higher than the limit values, the acidic desulfurization process was applied to both fuels as the second stage. In the desulphurization process applied at different ratios of 1/2 Acetic Acid-H2O2, 1/2 Formic Acid-H2O2, and Sulfuric Acid (5 gr–7, 5 g–10 g) to both fuels produced, it was determined that there was no decrease in the sulfur ratio; on the contrary, it was determined that it caused increases in both fuels. At the end of the preliminary tests, ethylhexyl nitrate, which was used as an improving additive in DL-PF, which has a low cetane index, was added to DL-IF and applied to this fuel at the same rates to measure its reactions to combustion and emissions. Since these diesel-like fuels are within and close to the diesel fuel standards used in internal combustion fuels, tests were carried out in a direct injection, single-cylinder, water-cooled diesel engine. The engine performance characteristics, combustion, and emission values of the test fuels were compared with those of diesel fuel (DF). In the performance analysis of DL-IF, DL-PF, DL-IF (EN), and DL-PF (EN) fuel blends with DF at different ratios, it was observed that the specific fuel consumption (SFC) of DF fuel was higher than the other industrial-plastic blended test fuels and lower on average than the Ethylhexyl Nitrate (EN) blended test fuels. The combustion and emission data under different loads were analyzed in detail by comparing the blended fuels and EN-doped experimental fuels with DF fuel in combustion parameters, and the results are given comparatively. While the blended fuels presented similar and close values to DF fuel, it was found that there were decreases in the parameters (cylinder gas pressure, net heat release, pressure increase rate) of blended fuels with the additive (EN). It was observed that NOx emissions were slightly higher than blended test fuels but lower than all blended test fuels, HC emissions were lower than blended test fuels and higher than blended fuels, CO2 emissions were higher in blended and blended test fuels except for high loads (7.5 bar Bmep), and smoke darkness was higher than DF fuel on average for all test fuels. The optimum results were predicted by ANN modeling using the data obtained in these experimental studies using the matlab program.

Author

Dr. Abdulkerim Yıldız

How to Cite

Abdulkerim Yıldız (Doctorate thesis). Experimental analysis of performance, emission and combustion characteristics of a diesel engine at different operating parameters of fuel obtained from waste industrial oils and waste plastics and optimisation with artificial neural network method, 2023, Batman University.

Keywords

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Batman University