A comprehensive RSM approach for modeling and optimization of performance and emission characteristics of a diesel engine fueled with waste tire pyrolysis oil and transformer oil blends


Creative Commons License

Öner İ. V., Arpa O., Doğan T. H., Yeşilyurt M. K., Kandemir L.

Journal of the Energy Institute, cilt.127, ss.1-23, 2026 (Scopus)

Özet

In this study, waste transformer oil (WTRO) and waste tire pyrolysis oil (WTPO) were mixed with standard diesel
fuel in different volumetric ratios to prepare four fuel samples (D100, WTRO25WTPO25, WTPO50, WTRO50). In
the first stage, the fuels were characterized physicochemically thru density (832.3862.3 kg/m3), kinematic
viscosity, cloud/flow point, lower heating value, and FT-IR spectroscopy; all mixtures except for WTPO50
(1.77 mm2/s) meet the ASTM D975 viscosity limits (1.96.0 mm2/s). In the second phase, performance and
exhaust emission tests were conducted under full load conditions in a single-cylinder, four-stroke, air-cooled
compression ignition (CI) engine within the 12002400 rpm range. Experimental data were analyzed using
second-order multi-term Response Surface Methodology (RSM) models; high statistical consistency was achieved
for the response variables BTE, BSFC, CO, and NOx with R2 = 0.9640.978 (p < 0.001). ANOVA results revealed
the dominant determining position of engine speed in performance parameters, while Pareto analyzes showed
that the SpeedWTRO interaction term was statistically significant in emission parameters. In terms of engine
performance, the WTRO50 blend increased torque and brake power by 8.9% and 8.7% respectively at 1200 rpm
compared to D100, while reducing BSFC by 4.3%. WTPO50, due to its low lower heating value (40.74 MJ/kg)
and high aromatic content, reached the highest BSFC value (486.01 g/kWh) at 2400 rpm. The WTRO25WTPO25
triple mixture exhibited a balanced emission profile while producing performance values close to D100. Using
the DerringerSuich suitability function and the Differential Evolution algorithm, scenario-based weighted multicriteria optimization was performed under four priority scenarios (Base, Performance, Economy, EmissionFocused). Overall fitness values remained within a narrow band (D = 0.3460.535; ΔD = 0.189), however, the
optimal engine speed exhibited a statistically significant shift of 832.6 rpm between scenarios (Performance
Scenario S-2: 1542.9 rpm; Emission-Focused Scenario S-4: 2375.5 rpm). In all optimization scenarios, the optimal
WTRO ratio was determined to be 50% and WTPO = 0%; this finding statistically proves the inadequacy of raw
and unprocessed WTPO compared to WTRO in terms of performance-emission balance. The study presents three
original engineering contributions: (i) the first RSM-based study evaluating WTRO and WTPO as simultaneous
dual diesel replacements; (ii) the first quantitative demonstration of an optimal speed shift of 832.6 rpm between
performance-prioritized and emission-prioritized scenarios; (iii) the statistical proof that raw WTPO requires preprocessing under all four scenarios. The findings show that WTRO offers a technically viable waste valorization
approach as a diesel substitute up to 50% and provides measurable CO and NOx emission benefits. HC, PM, and
soot emissions could not be measured as they were beyond the measurement capacity of the Testo 350 analyzer
used in this study; a comprehensive environmental assessment (HC/PM/soot measurement and life cycle analysis) has been left for future studies.