© The Author(s), published by EDP Sciences, 2024

Nomenclature

BTE: Brake Thermal Efficiency

BSFC: Brake-Specific Fuel Consumption

CO: Carbon Monoxide

HC: Hydrocarbon

NO_x: Nitrogen Oxides

RSM: Response Surface Methodology

ANOVA: Analysis of Variance

CCD: Central Composite Design

DOE: Design of Experiments

PE: Polyethylene

LDPE: Low-Density Polyethylene

HDPE: High-Density Polyethylene

PVC: Polyvinylchloride

PS: Polystyrene

PU: Polyurethane

PP: Polypropylene

PET: Polyethylene Terephthalate

XLPE: Crosslinked Polyethylene

1 Introduction

The limited availability and dispersed distribution of petroleum resources and the environmental damage caused by combustion end products have prompted researchers to concentrate on enhancing fuel properties, identifying new and cleaner energy sources, and enhancing efficiency [1]. For use in diesel engines, alternative fuels such as alcohols, compressed or liquefied natural gas, liquefied petroleum gas, vegetable and animal oils, or the biodiesel derived from them, tires, waste plastics and packaging, waste lubricants, agricultural residues, and wastes containing cellulose are utilized. The appropriate disposal of waste plastics is particularly interesting [2]. Although 8300 million metric tons (Mt) of raw plastics have been manufactured so far, in 2015, 6300 Mt of plastic garbage was created, of which 12% were burned, 79% were landfilled or deposited in the environment, and only 9% were recycled [3]. If existing production and waste management patterns persist, the worldwide output of plastic is projected to rise by 3.8% by 2030 [4], ultimately reaching a staggering 12000 million tons by 2050 [3]. The most often used plastics in use today are polyethylene (PE), polyvinyl chloride (PVC), polystyrene (PS), polyurethane (PU), polypropylene (PP), and polyethylene terephthalate (PET), even though there are around 20 distinct types of plastics.

Because of its degree of polymerization and suitability for various uses, PE is the plastic type with the largest percentage of plastic waste. High-density polyethylene (HDPE) and low-density polyethylene (LDPE) are the two most prevalent types of PE. They vary in linear chain lengths, branching, and crystallinity [5]. Cross-linked polyethylene (XLPE) is one other type of PE that is commonly used for the insulation of electrical wires and cables. However, XLPE is reported to contribute significantly to the total amount of PE waste [6]. XLPE waste is difficult to recycle due to its limited fluidity and inability to be formed in molds. For this reason, the recycling rate is much lower compared to other polymers, and most of the XLPE waste is kept in landfills [7]. In general, it is desirable to prevent waste pollution by incineration or landfill. Nevertheless, the incineration of these pollutants results in the consumption of energy and pollution of the atmosphere, whereas burial increases the need for landfills and causes soil pollution. For this reason, properly using these resources is believed to contribute to building a sustainable future, both environmentally and economically [1, 8]. Plastic waste can be recycled mechanically or chemically. Mechanical recycling involves sorting, separating, washing, and granulating to obtain the final product. However, this method is laborious and costly. The pyrolysis process [9], which is part of chemical recycling and has recently attracted the attention of researchers, eliminated the sorting and separation processes, and converted waste plastics into valuable hydrocarbons, fuels, and monomers without harming the environment [10]. Pyrolysis is a thermal decomposition process that takes place in the absence of oxygen without combustion. Pyrolysis can transform waste plastics into three distinct products: liquid, vapor, and solid. In internal combustion engines, the liquid product produced by pyrolysis of waste plastics can be utilized as a fuel due to its similarity to petroleum-derived fuels [1]. Because numerous researchers have investigated the chemical recycling of plastics [11–13], they have reported that waste plastics possess fuel qualities comparable to diesel fuel and can be utilized as fuel in internal combustion engines [14–16].

Sosetyo et al. produced LDPE plastic oil in a diesel engine with a single cylinder. They mixed it with diesel fuel at 10%, 20%, 30%, 40%, and 50% and analyzed it regarding engine power, torque, BSFC, and BTE. They stated that the addition of WPO to diesel fuel resulted in a 5% increase in engine power and torque and a reduction in BSFC and BTE. Additionally, they suggested that waste plastic oil could be utilized as an alternative fuel for diesel engines [17]. Mangesh et al. conducted a comparative analysis of the characteristics of oil obtained by pyrolysis HDPE, LDPE, PP, and PS to assess their viability as fuel for diesel engines. They found that PP pyrolysis oil is like diesel fuel regarding physicochemical properties. They blended 5%, 10%, and 15% PPO (polypropylene pyrolysis oil) by volume with diesel fuel to produce fuels. During engine tests at 25%, 50%, 75%, and 100% load levels, they observed a decrease in BTE as the blend ratio increased. They also reported an increase in CO, HC, and NO_x emissions. They reported that specific fuel consumption tended to decrease as the level rises all load fuels but increased with increasing blend ratios. They concluded that pressure inside the cylinder and heat dissipation increased with increasing blend ratios compared to diesel fuel [18].

The study conducted by Yaman et al. examined the impact of varying concentrations of 1-heptanol (0%, 10%, and 20%) at various compression ratios (10:1, 8:1, and 6:1) and varied engine loads (12, 8, and 4 kg) on a gasoline engine. They utilized RSM to determine the optimum operating parameters. Researchers identified the most favorable operating conditions to obtain optimal performance and emission levels, which include using 8% 1-heptanol, a compression ratio of 10.0:1, and an engine load of 6 kg. The researchers discovered that the BSFC, BTHE, CO₂, CO, HC, and NO_x values were 0.32 kg/kWh, 26.03%, 15.07%, 0.56%, 182.54 ppm, and 676.16 ppm, respectively. Furthermore, the validation research indicated that the discrepancy rates between the optimal and experimental outcomes were within the acceptable thresholds of 0.74% and 8.96% [19]. The effects of spray pressure (245 bar, 225 bar, and 205 bar) and engine load (100%, 75%, and 50%) variations in a single-cylinder diesel engine were experimentally investigated by Uslu. ALPY-diesel fuel blends (ALPY20, ALPY40, and ALPY60) were prepared by mixing with diesel fuel at varying ratios and analyzed using RSM. The data was analyzed using RSM. The analytical findings identified the optimal operating conditions as a 30.51% ALPY ratio, a spray pressure of 225 bar, and an engine load of 62.12%. The findings achieved under these optimal circumstances were as follows: 26.89% for BTE, 399.31 °C for exhaust gas temperature, 0.19% for CO, 9.09 ppm for HC, 474.73 ppm for NO_x, and 24.40% for smoke. According to reports, RSM is a very successful and dependable approach for forecasting the performance and emissions of a low-power diesel engine while employing various fuel mixes [20].

Investigations into engine performance and emissions in research may be time-consuming and expensive when using input factors like engine load, engine speed, fuel mixture, etc. As a result, researchers have turned to computer-based analysis programs to obtain results with the same accuracy with fewer experiments. Various methods, such as RSM, linear regression, artificial neural networks, and mathematical methods, are used for this purpose [21]. RSM is a proven and powerful application that is preferred for optimization in many scientific fields, reducing the number of tests, saving labor, and being economical compared to classical methods [22]. Several researchers have used RSM to investigate the impact of input elements on desired output parameters in studies related to alternative fuels [19, 20, 23].

Within the scope of the literature, various plastic waste materials have been subjected to pyrolysis using different techniques. Experimental research has focused on investigating the effect of using pyrolyzed materials as fuel in internal combustion engines on both engine performance and exhaust pollutants. However, research on the pyrolysis of non-mechanically recyclable XLPE cables is scarce, and their applicability in internal combustion engines has not been tested. This research has two objectives. First, to investigate the feasibility of blends of WPO and D100 derived from waste XLPE cables as fuel for a single-cylinder diesel engine. Second, an inquest was conducted using the RSM to optimize performance and reduce exhaust emissions.

2 Material method

In the experimental studies, WPO obtained from XLPE cable wastes was utilized in the facility established in partnership with Karabük University and Karabük Municipality, which was supported by BAKKA (Western Black Sea Development Agency) with the project “Disposal and Recovery of Plastic Wastes by Pyrolysis Method” with the contract number TR 81/20/SANAYİ/0016. The process of pyrolysis waste XLPE cables is depicted in Figure 1. In waste management for XLPE cables, the cables are initially broken into small pieces. The shredded plastics are then conveyed into the main reactor, subjected to incineration, producing gas, oil, and solid carbon black. The resulting pyrolytic gas is re-combusted in the system to obtain more energy, while the pyrolytic oil is distilled to obtain waste plastic oil.

Fig. 1 Schematic view of WPO production.

In the experimental study, the reference fuel D100 (0% WPO + 100% D100), WPO10 (10% WPO + 90% D100), WPO20 (20% WPO + 80% D100), WPO30 (30% WPO + 70% D100), WPO40 (40% WPO + 60% D100), and WPO50 (50% WPO + 50% D100) fuel mixtures including 10%, 20%, 30%, 40%, and 50%, respectively, WPO by volume were used. Diesel fuel (D100) and waste plastic oil (WPO) were analyzed at the “Energy and Chemistry Laboratory of Tübitak Marmara Research Centre” to determine the properties of the fuels used in the experimental study. Some chemical and physical properties of WPO and D100 used in the engine experiments are shown in Table 1.

The experiments were done with a generator with a diesel engine that was direct injection, single-cylinder, air-cooled, four-stroke, and a steady speed of 3000 rpm. Figure 2 provides a schematic representation of the experimental setup, while Table 2 provides full technical details for the test engine and generator.

Fig. 2 Schematic view of the experimental setup. 1. Computer, 2. Programmable Logic Controller (PLC), 3. Experiment set control panel, 4. Load resistors, 5. 4–20 mA converter, 6. Precision Balance, 7. Emission device, 8. Injector, 9. Experiment motor generator group.

The experiments were conducted at 220 bar standard spray pressure and 750, 1500, 2250, 3000, 3750, and 4500-W loads. No loading was performed at higher loads due to the irregular operation of the engine. Prior to commencing the testing, the engine was powered using a test fuel for a certain duration. Once it reached a stable state, the trials were started. Bilsa MOD 2210 WINXP-K exhaust gas meter with technical specifications in Table 3 was used to measure emissions.

To facilitate the loading of the generator, an adjustable load unit (0–6 kW) was designed to accommodate the alternator capacity. The loading of the diesel engine was achieved by connecting resistance groups to the generator’s output, which was connected to the shaft of the diesel engine. A load unit comprising resistors capable of loading up to 6000 W in 250 W increments was employed. The diesel engine was loaded by controlling the power drawn from the generator. A control panel was designed to switch the loads on and off and monitor and record the generator and engine assembly data. A Vipa CPU 313SC PLC carried out these operations. An interface was designed using LabVIEW software to control and monitor the variables during the experiment. The communication between the interface and the PLC was achieved via the TCP/IP communication protocol. The loading value of the generator in watts is transmitted to the PLC via the interface program, whereupon the PLC determines the resistance stages according to the incoming load value and activates the contactors, driving the resistors via their digital outputs. The experimental investigation measured fuel consumption using a Radwag WLC X2 electronic scale with 0.1 g accuracy. A Radwag brand 4–20 mA converter was connected to the scale’s output, and the PLC read the data in mA. The equivalent of the fuel values in grams at the beginning and end of each experiment and the time elapsed during the experiment in seconds were recorded in Excel format with LabVIEW software. Figure 3 depicts the LabVIEW block diagram of the fuel consumption measurements.

Fig. 3 LabVIEW block diagram image of fuel consumption measurements.

2.1 Response surface methodology

Due to its capacity for conducting multi-objective optimization and handling a wide range of inputs and response parameters, RSM is a valuable technique in experiments involving internal combustion engines. The RSM, or RSM, is a powerful statistical model used in engineering to improve and understand complex systems. RSM is to determine the influence of several factors on the response, elucidate the connection between these variables, and minimize the required number of experimental trials to optimize the response surface [24]. The primary goal of this optimization investigation is to determine the optimal operational conditions for a diesel engine by reducing CO, NO_x, HC, and smog emissions, increasing BTE, and minimizing BSFC. The experimental findings in RSM are examined using a second-order polynomial model to accurately anticipate outcomes using response regression [25].$$y={\beta }_0+\sum _i^k{\beta }_i{x}_i+{\beta }_0+\sum _{i=1}^k{\beta }_i{x}_i\sum _{j\ge 1}^k{\beta }_{\mathrm{ij}}{x}_i{x}_j+\mathcal{E}$$(1)where β₀ represents a fixed value, β_i represents a coefficient that varies linearly, and β_ij is a coefficient that represents the interaction between variables i and j. The variables i and j represent the coefficients for linear and quadratic terms, respectively. ℰ represents a random test error, k denotes the number of factors, y represents the predicted response, and x_i and x_j are independent factors [19].

This research used RSM to calculate the optimal engine load, WPO ratio, engine load, BSFC, BTE, CO, HC, NO_x, and soot emission responses. Central Composite Design (CCD) was selected for its precise forecast findings [20]. Table 4 contains the inlet components and levels, while Table 5 contains the settings of the RSM model. The MINITAB program was utilized to construct the RSM model. Thirteen experiments were conducted, each with a different combination of inlet factors. The RSM model was constructed using BTE (%), BSFC (g/kWh), NO_x (ppm), HC (ppm), CO (%), and smoke (%) as egress parameters, while WPO mixing ratio and engine load were selected as inlet parameters.

3 Results and discussion

3.1 Statistical analysis

In addition to RSM, ANOVA was used to see whether factor means changed significantly. ANOVA is an accepted statistical method for determining study component interactions. The F-value and p-values dominate the ANOVA table. Higher F-values imply model dependability, whereas lower p-values suggest model significance [26]. For a factor to affect the response, the p-value must be less than 0.05 [27]. In this research, ANOVA tests were conducted with 95% confidence. The percentage contribution of the elements shows their proportional influence on the response.

It is more appropriate to interpret these factors together, as they will yield linear results with obtained Pareto graphs [28]. A Pareto chart is a type of graph used to determine the effect of a factor on response and the importance of this effect. The vertical line on this graph represents the boundary line, indicating whether the input factors are significant. It can be concluded that the factors to the right of this boundary line are effective on the selected response [27]. Conversely, R² values are employed to ascertain the degree of fit between the experimental data and the model [20, 24, 26, 28]. This value is between 0 and 1, with a value of 1 signifying perfect agreement between the experimental data and the model. Table 6 presents the ANOVA findings for exhaust emission and engine performance responses.

Significant − (0,000 < p ≤ 0,05). (A: WPO ratio, B: Load, A² : WPO ratio * WPO ratio, B² : Load * Load, A*B: WPO ratio * Load)

Table 6 presents the R² values and shows how the factors’ impact on the replies, as shown by the p-value, is significant. A represents the engine load, and B represents the WPO ratio. If the p-value is greater than 0.05, the component under study does not significantly affect the response, in line with the 95% confidence level used in the experimental investigation. The WPO ratio and engine load significantly affect each response, as shown by the p-values in Table 6. Furthermore, the regression equations for the egress parameters BTE, BSFC, CO, HC, NO_x, and smoke emissions, as a function of the inlet factors, are provided in Table 7.

3.2 The combined effect of the WPO ratio and the load on engine performance

BSFC measures the fuel consumption rate per power unit during a certain time period. Figures 4a and 4c show the BSFC values as a surface and contour plot, respectively, as a function of changing WPO ratio and engine load. The inclusion of WPO results in an elevation in BSFC. The BSFC values that were the lowest were recorded when the engine load was 3750 W. At 3750 W engine load, while the reference diesel fuel was 330.67 g/kWh, it was 333.86 g/kWh for the WPO10 blend, 337.06 g/kWh for the WPO20 blend, 340.5 g/kWh for the WPO30 blend, 342.50 g/kWh for the WPO40 blend and 360.26 g/kWh for the WPO50 blend, which is approximately 8.94% higher than the reference diesel fuel. This may be attributed to the decreased calorific contents of the fuel mixtures. However, unlike the effective efficiency, a small increase in specific fuel consumption was observed at 4500 W. The decreased calorific values of the test fuels are a factor that directly affects the BSFC. Table 1 shows that the calorific value of the reference diesel fuel is somewhat higher than that of the pyrolysis fuel. Fuels with a lower calorific value need more fuel to be fed into the combustion chamber to get an identical level of effective potency [29]. This leads to a rise in fuel consumption [1, 30–32].

Fig. 4 (a) BSFC surface plot, (b) BTE surface plot, (c) BSFC contour plot, and (d) BTE contour plot.

Figures 4a and 4c demonstrate that the BSFC reduces as the engine load rises when using mixes of diesel fuel and waste plastic oil. The observed outcome may be ascribed to the significant heat transfer occurring at the walls of the combustion chamber, along with the considerably elevated friction loss seen under low engine load conditions. Moreover, when the engine load escalates, the quantity of fuel injected into the combustion chamber amplifies since more fuel is necessary to achieve higher loads [33]. The increased fuel quantity results in more fuel burning, leading to elevated temperatures. These higher temperatures, especially under heavier loads, cause the test fuels to ignite more quickly [34]. Fuel blends show lower specific fuel consumption due to increased combustion efficiency and higher cylinder pressure at higher loads, as specific fuel consumption increases at low engine loads and decreases as engine load increases.

Figures 4b and 4d illustrate the BTE values concerning the varying WPO ratio and engine load, presented as surface and contour plots, respectively. Thermal efficiency, a key indicator of the efficiency with which energy is utilized, refers to the proportion of heat energy produced because of the combustion reaction of the fuel converted into useful energy. An increase in the WPO content of the test fuels resulted in a reduction in thermal efficiency. This is believed to be because the energy released during the afterburning period is not converted into work [35]. Moreover, due to the lower cetane number of WPO compared to diesel fuel, an elevated proportion of WPO in the fuel mixture results in a more pronounced rise in the ignition delay, ultimately leading to a reduction in thermal efficiency [36, 37].

Upon examination of the influence of engine load on BTE, it was determined that an increase in load up to 4500 W engine load increased BTE values. As the load increased, combustion efficiency increased, and thermal losses decreased. At a load of 3750 W and with diesel fuel entirely, the BTE value reached its maximum point of 25.09%. For the same engine load, the following BTE values were obtained for the WPO10, WPO20, WPO30, WPO40, and WPO50 fuels: 24.90%, 24.71%, 24.65%, 24.40%, and 23.24%. The inclusion of waste plastic oil in diesel fuel at ratios of 10%, 20%, 30%, 30%, 40%, and 50% led to a reduction in BTE values of 0.75%, 1.51%, 1.75%, 2.75%, and 7.37%, respectively, when compared to pure diesel fuel. Thermal losses fall, and combustion efficiency rises with increasing load. However, at 4500 W engine load, there is a slight decrease in the BTE value. It is postulated that this is due to decreased combustion efficiency and increased BSFC. Other research in the literature has also shown similar findings using combinations of waste plastic and oil [30, 38–40]. Figure 5 displays the Pareto plot, which visually represents the efficiency of the components for the BTE and BSFC responses. Upon analyzing the Pareto plots, it is evident that both engine load and WPO ratio significantly influence BTE and BSFC. However, the engine load effect is more apparent than the WPO ratio.

Fig. 5 Pareto plots of BFSC (a) and BTE (b).

3.3 The combined effect of the WPO ratio and the load on the exhaust emission responses

CO emission is a crucial parameter representing lost chemical energy in engines due to the low A/F ratio of fuel. If combustion is complete and efficient, CO is converted to CO_2. Figures 6a and 6b show the surface and contour plots illustrating the impact of the combined variables of WPO % and engine load on CO emissions.

Fig. 6 (a) CO emission surface plot, (b) CO emission contour plot.

In regions of a lean mixture, higher CO concentrations are observed due to the formation of extinction zones around the combustion chamber and partial oxidation of unburned HC [41]. When Figure 6 is examined, it is evident that CO emissions decrease for all fuel mixtures with increasing load and increase slightly at 4500 W load. As the concentration of WPO in diesel fuel rose, there was a corresponding drop in CO emissions compared with pure diesel fuel. The CO emission level reached its minimum when the engine load was raised from 3000 W to 3750 W. At a load of 3750 W, the CO emissions were measured as 0.022% for diesel fuel, 0.020% for the WPO10 blend, 0.019% for the WPO20 blend, 0.018% for the WPO30 blend, 0.016% for the WPO40 blend and 0.014% for the WPO50 blend, respectively. This was accompanied by a 36.36% reduction in emissions when the WPO ratio was increased from 0% to 50%. One of the greatest essential parameters affecting fuel atomization is viscosity. It has been demonstrated that high viscosity has a negative effect on fuel atomization [42]. Diesel fuel exhibits higher viscosity compared to WPO fuel. As the concentration of WPO in the WPO and diesel mixture rises, the emission of CO decreases because of its greater volatility and lower viscosity, as well as enhanced fuel atomization and vaporization. This also ensures optimal fuel and air mixture throughout the ignition delay period while the engine is operating under high loads, resulting in reduced CO emissions [43]. The surface and contour plots in Figures 7a and 7b depict the combined influence of WPO % (weight percent of oil) and engine load on HC emissions.

Fig. 7 (a) HC emission surface plot, (b) HC emission contour plot.

Increasing engine load results in a direct and proportionate increase in HC emissions for all test fuels. More fuel being injected into the cylinder is the cause of the higher HC emissions seen with increasing engine load [16]. Consequently, the fuel/air ratio increased, creating rich mixing zones, which caused the sprayed fuel to travel to the cylinder walls, further increasing the fuel concentration in the core of the fuel beam. Another potential reason for the observed increase in HC emissions compared to diesel fuel with increasing engine load is fuel accumulation on the cylinder walls [44]. The highest HC emission values were obtained at an engine load of 4500 W. At an engine load of 4500 W, HC emissions were 29 ppm for diesel fuel, 32 ppm for the WPO10 blend, 39 ppm for the WPO20 blend, 41 ppm for the WPO30 blend, 46 ppm for the WPO40 blend, and 54 ppm for the WPO50 blend, respectively. The absence of saturated HC in WPO that is not broken down during combustion explains the greater HC concentration for WPO and diesel fuel mixes than pure diesel fuel [45]. Additionally, the lower viscosity of WPO and diesel blends than diesel fuel may contribute to fuel leakage from the injector nozzle [46] and uneven fuel injection into the combustion chamber [13], increasing HC emissions.

As engine load increased, NO_x emissions climbed for all test fuels. Figures 8a and 8b illustrate the combined effect of WPO percentage and engine load on NO_x emissions, as presented in surface and contour plots. As the engine load rises, there is a corresponding increase in the quantity of fuel entering the cylinders and the fuel-to-air ratio. Augmenting the fuel/air ratio results in heightened heat production inside the combustion chamber, leading to an escalation in temperatures. Consequently, a rise in the fuel/air ratio results in a corresponding increase in NO_x emissions. WPO blends demonstrated reduced NO_x emissions compared to diesel fuel across all loads. Diesel fuel had NO_x emissions of 833 ppm at 4500W load, whereas blends of WPO10, WPO20, and WPO30 had emissions of 809 ppm, 808 ppm, 716 ppm, and WPO50 mix had emissions of 703 ppm. At 4500 W engine load, a 15.6% reduction occurred by increasing the WPO ratio from 0% to 50%. It was observed that NO_x emissions decreased as the WPO ratio in diesel fuel increased. The low calorific value of WPO leads to a decrease in in-cylinder temperature and pressure, which is thought to cause WPO and its blends to produce fewer NO_x emissions. Similar results have been reported in different studies with WPO [38, 47, 48].

Fig. 8 (a) NOx emission surface plot, (b) NOx emission contour plot.

In diesel engines, smoke indicates incomplete combustion resulting from an overly rich A/F ratio or partially vaporized fuel particles [49]. The combined effects of engine load and WPO% on smoke emissions are shown using surface and contour plots in Figures 9a and 9b.

Fig. 9 (a) Surface plot of smoke emission, (b) Contour plot of smoke emission.

As the load increases, more fuel is introduced into the cylinders, leading to elevated temperatures in the regions with a high fuel concentration. As a result, the amount of oxygen available for combustion decreases during the diffusion combustion phase. This prolongs the length of the diffusion combustion process, hence resulting in an augmentation of smoke emissions [50]. The incorporation of a greater WPO content in the test fuels resulted in a reduction in smoke emissions at all engine loads. This phenomenon can be attributed to the fuel blends’ low viscosity and high volatility [43]. However, another contributing factor to reducing smoke emissions is the low cetane number of WPO fuels. Low cetane number fuels generally shorten the diffusion-controlled combustion period and produce lower smoke emissions [51, 52]. Furthermore, it was observed that the impact of low and mid-engine loading engine loads on smoke emissions was negligible. But when engine loads were high, smoke emissions significantly increased. The smoke emissions across all fuel types experienced a nearly 2.5-fold rise when the engine load was increased from 3750 to 4500 W. The greatest smoke emission was noted at the 4500 W motor load. At an engine load of 4500 W, smoke emissions were 6.732% for diesel fuel, 6.44% for the WPO10 blend, 6.164% for the WPO20 blend, 5.876% for the WPO30 blend, 5.192% for the WPO40 blend and 4.804% for the WPO50 blend. Figure 10 presents the Pareto plots, illustrating the relative importance of various factors in influencing the response. It can be observed that both engine load and WPO ratio exert a significant influence on the emission responses. But engine load is the most relevant factor in all responses.

Fig. 10 Pareto plots for CO (a), HC (b), NOx (c), and smoke (d).

The principal goal of this investigation is to minimize the BSFC by optimizing the WPO ratio with RSM optimization and simultaneously maximize the BTE while minimizing all emissions this investigation is to minimize the BSFC by optimizing the WPO ratio with RSM optimization and simultaneously maximize the BTE while minimizing all emissions. The optimization study with RSM determined that the optimum WPO ratio is 19.6%, and the optimum engine load value is 2600 W, as shown in Figure 11. The highest BTE value achieved when the optimal WPO ratio and engine load are used is 22.3%. The lowest values for BSFC, CO, HC, NO_x, and smoke emissions are 332.3 g/kWh, 0.033%, 31.5 ppm, 397.9 ppm, and 1.64%, respectively. The experimental investigation’s findings, derived from the ideal engine load and WPO ratio, are compared to the optimization results. Table 8 illustrates the error rates and outcomes.

Fig. 11 Optimization plot for BTE, BSFC, CO, HC, NOx, and smoke.

4 Conclusions

This study’s optimization research used RSM to recognize the ideal WPO to engine load ratio, providing the most power and the fewest emissions. Experiments were conducted at five different WPO ratios (0%, 10%, 20%, 30%, 40%, and 50%) and six different engine loads (750, 1500, 2250, 3000, 3750, and 4500 W) to gather the data required for the RSM optimization. The results are shown below:

• The optimal WPO ratio is 19.6%, with an engine load of 2600 W.

• BTE, BSFC, CO, HC, NO_x, and smoke responses have R² values of 99.95%, 97.76%, 98.10%, 99.74%, 99.79%, and 95.67%, respectively. The findings indicate the RSM technique can forecast the impact of the WPO ratio on the responses of a low power single cylinder diesel engine at various engine load levels.

• The highest BTE value, achieved at the optimal WPO and engine load, was determined to be 22.3718%. Additionally, the lowest values for BSFC, CO, HC, NO_x, and smoke were measured at 332.33 g/kWh, 0.0334%, 31.5258 ppm, 397.9181 ppm, and 1.6394%, respectively.

• The results indicated that adding WPO increased HC emissions, decreased CO, NO_x, and smoke emissions, and slightly negatively impacted BTE and BSFC.

• As the engine load rose, it has been seen that BTE rose and BSFC reduced. Additionally, the engine load rose, leading to a reduction in CO emissions and a rise in all other emissions.

• RSM models have been seen to efficiently optimize engine performance and emission characteristics for a low-power diesel engine using WPO-D100 mixtures, with an error rate of less than 5%, according to research. This research can help predict and optimize optimal operating factors with fewer tests, improving engine performance and emissions.

Acknowledgments

This research was funded by the Coordination Unit for Scientific Research Projects at Karabuk University. (Project No: KBÜBAP-21-DS-059 and KBÜBAP-21-DR-060). We thank the BAP unit for their support.

References

All Tables

All Figures

	Fig. 1 Schematic view of WPO production.
In the text

	Fig. 2 Schematic view of the experimental setup. 1. Computer, 2. Programmable Logic Controller (PLC), 3. Experiment set control panel, 4. Load resistors, 5. 4–20 mA converter, 6. Precision Balance, 7. Emission device, 8. Injector, 9. Experiment motor generator group.
In the text

	Fig. 3 LabVIEW block diagram image of fuel consumption measurements.
In the text

	Fig. 4 (a) BSFC surface plot, (b) BTE surface plot, (c) BSFC contour plot, and (d) BTE contour plot.
In the text

	Fig. 5 Pareto plots of BFSC (a) and BTE (b).
In the text

	Fig. 6 (a) CO emission surface plot, (b) CO emission contour plot.
In the text

	Fig. 7 (a) HC emission surface plot, (b) HC emission contour plot.
In the text

	Fig. 8 (a) NOx emission surface plot, (b) NOx emission contour plot.
In the text

	Fig. 9 (a) Surface plot of smoke emission, (b) Contour plot of smoke emission.
In the text

	Fig. 10 Pareto plots for CO (a), HC (b), NOx (c), and smoke (d).
In the text

	Fig. 11 Optimization plot for BTE, BSFC, CO, HC, NOx, and smoke.
In the text