EDP Sciences logo
Submit your paper
Search
Menu
Home All issues Volume 79 (2024) Sci. Tech. Energ. Transition, 79 (2024) 37 Full HTML
Open Access
Issue
Sci. Tech. Energ. Transition
Volume 79, 2024
Article Number 37
Number of page(s) 15
DOI https://doi.org/10.2516/stet/2024031
Published online 27 June 2024

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

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Nomenclatures

BSFC: Brake specific fuel consumption

BTE: Brake thermal efficiency

CNG: Compressed natural gas

CH3OH: Methanol

CO: Carbon monoxide

CO2: Carbon dioxide

HC: Hydrocarbon

L/min: Liter per minute

NaOH: Sodium hydroxide

NOx: Nitrogen oxides

O2: Oxygen

ppm: Particle Per Million

rpm: Revolutions per minute

TDC: Top dead centre

1 Introduction

The high demand for conventional fossil-based fuels leads to excessive environmental pollution [1, 2]. Therefore, environmentally friendly fuels need to be brought to the fore. Biodiesel which is derived from vegetable oils is one of the most popular alternative fuels to conventional fossil-based fuels since it is biodegradable, non-toxic, and renewable [3, 4]. Similar properties of biodiesel compared to diesel generally require little or no modification on the diesel engine. However, the differences in terms of carbon atom numbers and the O2-hydrogen bond of the fuels influence the diesel engine operating characteristics [5, 6].

CNG, which is mainly composed of methane, is another type of fuel that can be used in diesel engines to improve exhaust emissions due to its low-carbon composition. The combination of diesel and CNG lowers the combustion temperature and subsequently reduces NOx emissions [7]. However, the combination of diesel and CNG has some disadvantages such as low combustion efficiency [8].

Many studies have focused on the enlightenment of these differences in biodiesel from various sources and different fuel blends. For instance, Tosun and Özcanlı [9] conducted engine performance test comparisons between a blend of soybean biodiesel with hydrogen enrichment and 20% nanoparticle additives and pure biodiesel. Pure biodiesel delivered the worst performance characteristic due to the higher viscosity and lower energy content. However, both fuels improved the CO emissions. On the other hand, Keskin et al. [10] used crude tall oil with synthesized manganese-based and nickel-based additives in their experiments. It showed that the engine has better performance due to the biodiesel blend’s low sulphur content, aromatic content, low viscosity, higher cetane number, and low cloud point.

Rapid predicting performance, optimizing conditions, and designing structures are accomplished using response surface methodology (RSM) and smart optimization algorithms, which rely on a simulation model enhanced by helical inserts in SMMR [11]. In literature, there are many studies about the optimization process [11–13]. Response surface methodology (RSM) is an effective model for predicting the relationship between desired input and output parameters via mathematical and statistical techniques [14, 15]. RSM is a collection of statistical and mathematical techniques which is useful for prediction, modelling, and optimisation [16, 17]. RSM can be used to determine the optimum point of various parameters. The Box–Behnken design (BBD) is among the various designs offered within RSM commonly employed for designing experiments to analyze the performance and emissions of biofuel-powered engines. BBD requires fewer tests, leading to reduced effort, decreased expenses, and a quicker turnaround compared to the complete factorial design. Additionally, when contrasted with the central composite design, BBD proves to be more efficient regarding the ratio of model coefficients to the total number of trials [18].

Shirneshan et al. [19] implemented RSM to study the effect of biodiesel from waste cooking oil on the operating parameters of four-cylinder direct injection diesel engines such as engine speed, engine load, and exhaust emissions. From their findings, the optimum conditions with a high desirability of 0.98 were obtained at 77.8% biodiesel percentage in fuel, 41.25% engine load, and 2800 rpm engine speed. Simsek et al. [20] used RSM to determine the optimum ratio of liquefied petroleum gas, where the value was 35% at an optimum engine load of 2400 W with less than 4% error ratio.

Singh et al. [21] also used RSM for Spirulina (L.) microalgae and conventional diesel blends in a single-cylinder variable compression ratio engine. BSFC, BTE, PM, NOx, and CO2 were optimised by changing the engine load and the compression ratio. The results showed that maximum BTE, minimal BSFC, and lower emissions could be achieved at an engine load of 63.63%, a compression ratio of 16.5, and a B20 blend (20% biodiesel and 80% diesel). Krishnamoorthy et al. [22] used a diesel-waste cooking oil-alcohol blend to study the relationship between fuel, exhaust gas recycling (EGR) rate, injection timing, and variables such as BTE, BSFC, CO, HC, NOx, and smoke. They concluded that the fuel blend (D50–WCO30–Pe20) injected at a 23° crank angle before TDC with a 15% EGR rate was optimum for exhaust emissions and engine performance at a maximum desirability of 0.974.

Subbaiah et al. [23] studied the performance and emission attributes of a Common Rail Direct Diesel Injection (CRDI) engine fueled by a blend of waste plastic oil and biodiesel, incorporating exhaust gas recirculation (EGR). This investigation utilized the Box–Behnken Design (BBD) matrix within the framework of Response Surface Methodology (RSM). They were chosen because the engine’s input parameters were IP, load and EGR ratio, and the responses were BSFC, BTE, NOx and HC. The optimum output engine responses were 21.25% BTE, 0.411 kJ/kW h BSFC, 20.31 ppm HC and 883.29 ppm NOx, respectively.

Kumar et al. [24] implemented artificial neural networks and RSM techniques to predict and optimise palm biodiesel and decanol proportion in the blend for a single-cylinder, four-stroke compression ignition engine. All samples have 10%, 20%, and 30% decanol by volume, and 50% diesel. The study indicated that the optimal proportion was 30% decanol in the fuel blend at a 0.754% desirability rate. Sharma et al. [25] performed the Taguchi method and RSM in a diesel engine test with blends of diesel-jojoba biodiesel. Several input parameters were used such as injection pressure, injection timing of fuel, and percentage blends of jojoba-based biodiesel, and engine load as the input parameters. The RSM optimiser showed that the optimum values of corresponding engine responses such as BTE, peak cylinder pressure, exhaust gas temperature, and HC were 34.83%, 5.11 MPa, 444.74 °K, and 38.85 ppm respectively. These engine responses were determined at 25° before TDC injection timing, 21.52 MPa fuel injection pressure, and 24.11% blending of jojoba biodiesel at 80% engine load condition with 0.9024 composite desirability.

From the literature, most works were focused on the optimisation of exhaust emissions of diesel engines, to the best of the authors’ knowledge. Therefore, this paper presents the optimisation method on fuel to identify engine vibration and noise (VN) attributes of a diesel engine in addition to exhaust emissions. A Box–Behnken method based on RSM was used to determine the minimum values of exhaust emissions and engine VN of CNG-biodiesel blends by optimising 1) CNG addition, 2) biodiesel blend ratio, and 3) engine speed.

2 Materials and method

In this work, canola oil was used as a feedstock in the transesterification reaction. Canola oil was obtained from local markets. NaOH and CH3OH were used as a catalyst and an alcohol, respectively.

2.1 Biodiesel production

The production of biodiesel consisted of a 250 ml spherical glass reactor equipped with a condenser, a contact thermometer, and a magnetic stirrer. The production of biodiesel has six processes. 1) 50 ml of oil was placed in the glass reactor and heated up to 60 °C for each run. 2) 6:1 of CH3OH to oil molar ratio and 0.5 wt NaOH were used to obtain a sodium methoxide mixture. 3) The solution was added to the preheated canola oil. The transesterification reaction was carried out at 60 °C reaction temperature for 90 min by stirring. 4) After the completion of the transesterification reaction, the crude biodiesel was transferred into a separation funnel and held on for 8 h inside it. 5) The crude glycerine was removed from the methyl ester, and then washed and heated up to 110 °C to eliminate the residuals and water contents. 6) The filtering operation was carried out to separate small impurities inside the methyl ester.

2.2 Preparation of test fuels for engine tests

The obtained biodiesels with transesterification reaction were mixed with a neat diesel at different volumetric ratios (20% and 40%).

The properties of CNG which were used in the engine tests contained 97.37% methane, 2.33% nitrogen, 0.23% ethane, and 0.07% propane. CNG was injected through the intake manifold into intake air with different flow rates; 5 litre per minute (L/min) (CNG5), 10 L/min (CNG10), and 15 L/min (CNG15). The CNG regulator employs a two-stage regulation mechanism. Initially, it lowers the CNG pressure from 200 bar to 12 bar, then further decreases it to 0–1 bar in the second stage. Subsequently, gas flow is managed by a needle vane following the regulator and the flow rate is gauged using a flowmeter positioned after the needle vane.

The fuel properties of diesel and canola biodiesel are given in Table 1.

Table 1

Fuel properties of diesel, canola biodiesel, and methane.

2.3 Test setup

The engine tests were based on a four-stroke, four-cylinder, direct-injection diesel. The specifications of the engine are given in Table 2 and Figure 1, the schematic of the experimental set-up was illustrated. Engine experiments were performed at 1 atm and between 20–25 °C. The engine was brought to a consistent operating temperature and the fuel line was purged before introducing it with a different fuel. It was then allowed to stabilize before collecting experimental data. The various fuels were tested at engine speeds of 1500 (±3 rpm), 1800 (±3 rpm), and 2100 rpm (±3 rpm) by adjusting the throttle, without any load in order to eliminate the noise and vibration generated by the dynamometer. Moreover, the periphery of the dynamometer was covered with a wooden box which has sound absorber panels inside it. After the engine stabilized, experimental data was collected for 10 s and an average of them was used.

thumbnail Fig. 1

Schematic view of engine test setup.
Table 2

Engine technical specifications.

MRU Delta 1600-V exhaust gas analyser was used to measure exhaust emissions values. Exhaust emissions were analysed via MRU Delta 1600-V gas analyser which is able to measure CO, carbon dioxide (CO2), oxygen (O2), nitric oxide (NO), and nitrogen dioxide (NO2) emissions. The range and accuracy of them are; 0–4000 ppm and ±20 ppm; 0%–20% and ±0.5%; 0%–22% and 0.01%; 0–1000 ppm and ±5 ppm; and 0–200 ppm and ±5 ppm, respectively. The device calculates NOx emissions according to the sum of NO and NO2 emissions.

Acoustic and vibration data were recorded via SoundbookTM universal portable measuring system with SAMURAI v2.6 software from SINUS Messtechnik GmbH. ToughbookTM CF-19 from a Panasonic portable PC was used to record the VN data of the engine. VN data was filtered according to ISO 10816, ISO 7919, and ISO 2954 standards. The vibration measurement range was from 2 Hz to 20 kHz. The sound level meter (SLM) of the software meets the Class 1 SLM according to IEC 60651, IEC 60804, and DIN EN 61672-1:2003 standards. The vibration of the engine was measured in three orthogonal axes (longitudinal, vertical, and lateral). The total engine vibration acceleration (atotal) was calculated via these values using equation (1).(1)

Triaxial ICP® accelerometer sensor from PCB electronics model 356A33 and GRAS 46AF 1/2″ LEMO half-inch Free-field Standard Microphone Set were used in the engine test setup. The accelerometer was adhered to the engine support with quick bonding gels to measure the data with high accuracy. The microphone was placed one meter away from the engine.

2.4 Uncertainty analysis

In this work, the experimental combined uncertainty was computed by equation (2). According to the study of How et al. [28] and the combined uncertainty was found as 4.80%.(2)

2.5 Modelling and optimisation

RSM involves employing a blend of mathematical and statistical techniques to model and analyze situations where a desired outcome is influenced by multiple variables, to optimize this outcome. Design of Experiment (DOE) serves as a commonly employed tool for attaining the most favourable optimization outcome through an optimal number of experiments. In the optimization process, several DOE techniques are utilized, including fractional design, Full Factorial Design (FFD), the Taguchi method, Box Behnken Design (BBD), and Central Composite Design (CCD) [29]. In this study, Box–Behnken of RSM was used to design and analyse the independent variables and their responses. A BBD is a type of experimental design for response surfaces that does not incorporate a built-in factorial or fractional factorial design. BBDs are employed to generate higher-order response surfaces with fewer required runs compared to a typical factorial approach. The critical parameters are assessed through a BBD experiment. This experimental setup enables achieving a satisfactory fit using a quadratic model with the gathered data points [30].

Three input parameters were used in the design of the experiment using Box–Behnken, namely CNG addition (X1), biodiesel ratio (X2), and engine speed (X3). All input parameters were arranged at three levels (−1, 0, +1) with equally spaced intervals. MINITAB statistical software was used to enhance the RSM model. Table 3 depicts the input parameters and their levels.

Table 3

Input factors and their levels.

A total of 17 engine tests were performed, where these numbers were determined by equation (3) [26].(3)where N is the number of engine tests required, k is the number of input parameters and Cp is the number of central points. Statistical 3-D plots, analysis of variance (ANOVA), and coefficient of determination (R2) were used to interpret the results. For the factors, the polynomial equation (Eq. (4)) is obtained from [27].(4)

where Y is the predicted response, β0 is model constant, X1, X2, and X3 are input parameters; β1, β2, and β3 are linear coefficients; β12, β13, and β23 are cross-product coefficients; and β11, β22, and β33 are the quadratic coefficients.

The process flow diagram of the applied RSM model is shown in Figure 2. The corresponding NOx, CO, CO2, O2 emissions, and engine VN values as the responses in the engine tests are given in Tables 4 and 5.

thumbnail Fig. 2

Flowchart of RSM.
Table 4

Test and predict values of engine responses for NOx, CO, CO2, and O2 emissions for canola biodiesel.

Table 5

Test and predicted values of responses for VN for canola biodiesel.

Table 6 provides the determined values of R-squared, adjusted (Adj.) R-squared, and prediction (Pred.) R squared.

Table 6

Model evaluation for canola biodiesel.

2.6 Desirability approach of RSM

RSM-based desirability approach was used to optimise process parameters for measuring engine responses. MINITAB statistical software was used for optimisation analysis. Each response was transformed to a dimensionless desirability value (d). It ranges between d = 0, which suggests that the response was completely unacceptable, and d = 1, which suggests that the response was more desirable. The criteria of this work are to minimise the values of exhaust emissions and engine VN.

3 Results and discussions

3.1 ANOVA method

The significance of the relationship between the input parameters and engine responses was determined by using the ANOVA method. Tables 7 and 8 show the F values and P values of input parameters and engine responses. A “P value” less than 0.05 is considered significant.

Table 7

ANOVA method for exhaust emissions.

Table 8

ANOVA method for engine VN.

3.2 Impacts of input parameters on NOx emissions

Combustion temperature, combustion duration, and local O2 concentration are the primary factors of NOx formation [31]. Figure 3 shows the effects of input parameters on NOx emissions. As can be seen, the NOx emissions were decreased with the addition of CNG. The reduction in NOx emissions was due to the cooling effect of the injection of cold CNG into the combustion chamber. This phenomenon reduced the peak combustion temperature and subsequently reduced NOx emissions [32].

thumbnail Fig. 3

Impacts of input parameters on NOx emissions.

The ANOVA method showed that NOx emissions were significantly influenced by CNG flow rate, biodiesel blend ratio, and engine speed. The coefficient of determination of R2 was 0.9911 and the difference between the adjusted R2 and predicted R2 was 0.122 and within the range of 0.2. The obtained quadratic equation for NOx emissions is given by equation (5).(5)

3.3 Impacts of input parameters on CO emissions

Figure 4 shows the effects of input parameters on CO exhaust emissions. The formation of CO emissions was due to inadequate O2 and insufficient combustion quality [33]. CO emissions value decreased with the increase in biodiesel ratio. This phenomenon was due to the fuel-bound O2 in biodiesel reducing the local fuel-rich zone inside the combustion chamber, therefore enhancing the complete oxidation of fresh carbon atoms or partially oxidised CO to CO2, and reducing CO emissions. On the other hand, CO emissions increased with the increment in the CNG addition since incoming air was replaced with CNG gas. The ANOVA model provided that the P values of CNG addition flow and engine speed were smaller than 0.0001. The goodness of fit for the CO model was 0.99. The regression formed of CO emissions is given by equation (6).(6)

thumbnail Fig. 4

Impacts of input parameters on CO exhaust emissions.

3.4 Impacts of input parameters on CO2 emissions

CO2 is the main source of global warming and ozone layer depletion. However, this molecule can be absorbed by plants and algae. Figure 5 gives the effects of input parameters on CO2 emissions. During the combustion of fuel, due to the insufficient amount of O2, all carbon molecules did not convert to CO2 molecules [34]. CO2 emissions values were increased by using biodiesel compared to conventional diesel fuel as explained in the section impacts of input parameters on CO emissions. The formation of CO2 is shown in equation (7).(7)

thumbnail Fig. 5

Impacts of input parameters on CO2 emissions.

3.5 Impacts of input parameters on O2 emissions

Extra O2 molecules may come from some alternative fuels, such as biodiesel. The amount of O2 in the exhaust gas is one of the important feedback parameters in adjusting the fuel injection duration [35]. In a diesel engine, the occurrence of O2 molecules in exhaust gases is generally related to an inefficient combustion process since the engine operates at lean air–fuel mixtures. Figure 6 demonstrates the various surface plots of O2 for different CNG additions, biodiesel ratios, and engine speeds. The formation of O2 is shown in equation (8).(8)

thumbnail Fig. 6

Impacts of input parameters on O2 emissions.

3.6 Impacts of input parameters on engine vibration

The engine vibration is influenced by the combustion quality, piston movement, fluid impact, and inertia of moving parts [36]. The results in Figure 7 enlighten that the engine vibration decreased with the increment of CNG addition and biodiesel blend ratio. The extra O2 content of biodiesel and the different combustion properties of CNG can affect the peak pressure rise rate, combustion duration, and ignition delay which results in less engine vibration severity [35, 37]. The engine vibration severity is shown in equation (9).(9)

thumbnail Fig. 7

Impacts of input parameters on engine vibration.

3.7 Impacts of input parameters on engine noise

Combustion, mechanical, intake, and exhaust noises are the main sources of diesel engine noise [38]. Figure 8 shows the decrease in engine noise due to the engine vibration decrement. Similar to the section impacts of input parameters on engine vibration, these results were the effect of increment in CNG addition and biodiesel blend ratio. The engine noise severity is given by equation (10).(10)

thumbnail Fig. 8

Impacts of input parameters on noise.

3.8 Optimisation and validation

Table 9 provides the optimisation criteria for dependent variables. Figure 9 demonstrates the obtained optimum input parameters and output values from canola biodiesel optimisation.

thumbnail Fig. 9

Optimisation of NOx, CO, CO2, O2 exhaust emissions and engine NV.
Table 9

Optimisation of test setup and desirability of engine responses.

The optimum engine responses were found at 93.77 ppm, 438.05 ppm, 1.47%, 18.59%, 37.17 m/s2 and 91.34 dB[A] for NOx, CO, CO2, O2, and engine VN respectively. These values were for optimum input parameters at 9.24 L/min of CNG addition, 40% biodiesel blend ratio, and 1524.24 rpm of engine speed. Engine tests were carried out to validate the optimal solution and the results are shown in Table 10.

Table 10

Validation of engine test results.

The engine test results were compared with RSM optimiser values. The errors were calculated using equation (11) [24].(11)

The errors for NOx, CO, CO2, O2, and engine VN were 4.2%, 3.8%, 4.9%, 0.25%, 4.12%, and 0.17% respectively. From previous literature [19], all error values were within the acceptable limit.

4 Conclusions

The impacts of input parameters such as CNG addition, canola biodiesel blend ratio, and engine speed on the engine responses (exhaust emissions, and engine VN) were investigated during engine tests. The RSM model was successfully utilised to predict and optimise the measured engine responses. The results from the study can be summarised into four segments. 1) The highest approachable desirability was 0.6710 in the optimal operating parameters at 9.24 L/min CNG addition, 40% biodiesel ratio, and 1524.24 rpm engine speed. 2) The optimum engine responses of NOx, CO, CO2, O2, and engine VN were obtained at 93.77 ppm, 438.05 ppm, 1.47%, 18.59%, 37.17 m/s2, and 91.34 dB[A], respectively. 3) The coefficient determination of R2 for NOx, CO, CO2, O2, and engine VN were 99.11%, 99.22%, 99.41%, 99.70%, 98.65%, and 98.60% respectively. 4) The errors between the optimised result and the engine test result for NOx, CO, CO2, O2, and engine VN were 4.2%, 3.8%, 4.9%, 0.25%, 4.12%, and 0.17% respectively. These errors were within the acceptable limit.

Conflicts of interest

No potential conflict of interest was reported by the authors.

References

  • Fangfang F., Alagumalai A., Mahian O. (2021) Sustainable biodiesel production from waste cooking oil: ANN modeling and environmental factor assessment, Sustain. Energy Technol. Assess 46, 101265. [Google Scholar]
  • Qadeer M.U., Ayoub M., Komiyama M., Daulatzai M.U.K., Mukhtar A., Saqib S., Ullah S., Qyyum M.A., Asif S., Bokhari A. (2021) Review of biodiesel synthesis technologies, current trends, yield influencing factors and economical analysis of supercritical process, J. Clean. Prod. 309, 127388. [CrossRef] [Google Scholar]
  • Helmiyati H., Budiman Y., Abbas G.H., Dini F.W., Khalil M. (2021) Highly efficient synthesis of biodiesel catalyzed by a cellulose@hematite-zirconia nanocomposite, Heliyon 7, 3, e06622. [CrossRef] [PubMed] [Google Scholar]
  • Silva G.S., Marques E.L.S., Dias J.C.T., Lobo I.P., Gross E., Brendel M., da Cruz R.S., Rezende R.P. (2012) Biodegradability of soy biodiesel in microcosm experiments using soil from the Atlantic Rain Forest, Appl. Soil Ecol. 55, 27–35. [CrossRef] [Google Scholar]
  • Çelebi K., Uludamar E., Tosun E., Yıldızhan Ş., Aydın K., Özcanlı M. (2017) Experimental and artificial neural network approach of noise and vibration characteristic of an unmodified diesel engine fuelled with conventional diesel, and biodiesel blends with natural gas addition, Fuel 197, 159–173. [CrossRef] [Google Scholar]
  • Shojae K., Mahdavian M., Khoshandam B., Karimi-Maleh H. (2021) Improving of CI engine performance using three different types of biodiesel, Process Saf. Environ. Prot. 149, 977–993. [CrossRef] [Google Scholar]
  • Felayati F.M., Semin, Cahyono B., Bakar R.A., Birouk M. (2021) Performance and emissions of natural gas/diesel dual-fuel engine at low load conditions: Effect of natural gas split injection strategy, Fuel 300, 121012. [CrossRef] [Google Scholar]
  • Shen Z., Wang X., Zhao H., Lin B., Shen Y., Yang J. (2021) Numerical investigation of natural gas-diesel dual-fuel engine with different piston geometries and radial clearances, Energy 220, 119706. [CrossRef] [Google Scholar]
  • Tosun E., Özcanlı M. (2021) Hydrogen enrichment effects on performance and emission characteristics of a diesel engine operated with diesel-soybean biodiesel blends with nanoparticle addition, JESTECH 24, 3, 648–654. [Google Scholar]
  • Keskin A., Gürü M., Altiparmak D. (2007) Biodiesel production from tall oil with synthesized Mn and Ni based additives: Effects of the additives on fuel consumption and emissions, Fuel 86, 7–8, 1139–1143. [CrossRef] [Google Scholar]
  • Yang W.W., Tang X.Y., Ma X., Li J.C., Xu C., He Y.L. (2023) Rapid prediction, optimization and design of solar membrane reactor by data-driven surrogate model, Energy 285, 129432. [CrossRef] [Google Scholar]
  • Petcharat N., Wiangkham A., Pichitkul A., Tantrairatn S., Aengchuan P., Bureerat S., Banpap S., Khunthongplatprasert P., Ariyarit A. (2023) The multi-objective optimization of material properties of 3D print onyx/carbon fiber composites via surrogate model, Mater. Today Commun. 37, 107362. [CrossRef] [Google Scholar]
  • Aubeelack H., Segonds S., Bes C., Druot T., Brezillon J., Bérard A., Duffau M., Gallant G. (2023) Surrogate model development for optimized blended-wing-body aerodynamics, J. Aircraft 60, 2, 437–448. [CrossRef] [Google Scholar]
  • Das S., Kashyap D., Bora B.J., Kalita P., Kulkarni V. (2021) Thermo-economic optimization of a biogas-diesel dual fuel engine as remote power generating unit using response surface methodology, TSEP 24, 100935. [Google Scholar]
  • Ye W., Wang X., Liu Y., Chen J. (2021) Analysis and prediction of the performance of free- piston Stirling engine using response surface methodology and artificial neural network, Appl. Therm. Eng. 188, 116557. [CrossRef] [Google Scholar]
  • Singh Y., Sharma A., Tiwari S., Singla A. (2019) Optimization of diesel engine performance and emission parameters employing cassia tora methyl esters-response surface methodology approach, Energy 168, 909–918. [CrossRef] [Google Scholar]
  • Adam I.K., Aziz A.A.R., Yusup S., Heikal M.R. (2016) Optimization of performance and emissions of a diesel engine fuelled with rubber seed- palm biodiesel blends using response surface method, Asian J. Appl. Sci. 4, 2, 401–421. [Google Scholar]
  • Said Z., Le D.T.N., Sharma P., Dang V.H., Le H.S., Nguyen D.T., Bui T.A.E., Nguyen V.G. (2022) Optimization of combustion, performance, and emission characteristics of a dual-fuel diesel engine powered with microalgae-based biodiesel/diesel blends and oxyhydrogen, Fuel 326, 124987. [CrossRef] [Google Scholar]
  • Shirneshan A., Almassi M., Ghobadian B., Borghei A.M., Najafi G. (2016) Response surface methodology (RSM) based optimization of biodiesel-diesel blends and investigation of their effects on diesel engine operating conditions and emission characteristics, Environ. Eng. Manag. J. 15, 12, 2771–2780. [CrossRef] [Google Scholar]
  • Simsek S., Uslu S., Simsek H., Uslu G. (2021) Improving the combustion process by determining the optimum percentage of liquefied petroleum gas (LPG) via response surface methodology (RSM) in a spark ignition (SI) engine running on gasoline-LPG blends, Fuel Process Technol. 221, 106947. [CrossRef] [Google Scholar]
  • Singh T.S., Rajak U., Samuel O.D., Chaurasiya P.K., Natarajan K., Verma T.N., Nashine P. (2021) Optimization of performance and emission parameters of direct injection diesel engine fuelled with microalgae Spirulina (L.) – Response surface methodology and full factorial method approach, Fuel 285, 119103. [CrossRef] [Google Scholar]
  • Krishnamoorthy V., Dhanasekaran R., Rana D., Saravanan S., Kumar B.R. (2018) A comparative assessment of ternary blends of three bio-alcohols with waste cooking oil and diesel for optimum emissions and performance in a CI engine using response surface methodology, Energy Convers. Manag. 156, September 2017 337–357. [CrossRef] [Google Scholar]
  • Subbaiah M.V., Reddy S.S.K., Prasad B.D. (2022) Optimization of performance and emission characteristics of common direct injection diesel engine using response surface methodology, Mater. Today Proc. 68, 1294–1304. [CrossRef] [Google Scholar]
  • Kumar A.N., Kishore P.S., Brahma Raju K., Ashok B., Vignesh R., Jeevanantham A.K., Nanthagopal K., Tamilvanan A. (2020) Decanol proportional effect prediction model as additive in palm biodiesel using ANN and RSM technique for diesel engine, Energy 213, 119072. [CrossRef] [Google Scholar]
  • Sharma A., Singh Y., Singh N.K., Singla A. (2019) Sustainability of jojoba biodiesel/diesel blends for DI diesel engine applications- taguchi and response surface methodology concept, Ind. Crops. Prod. 139, 111587. [CrossRef] [Google Scholar]
  • Patel P.D., Lakdawala A., Patel R.N. (2016) Box–Behnken response surface methodology for optimization of operational parameters of compression ignition engine fuelled with a blend of diesel, biodiesel and diethyl ether, Biofuels 7, 2, 87–95. [CrossRef] [Google Scholar]
  • Ramachander J., Gugulothu S.K., Sastry G.R.K., Panda J.K., Siva Surya M. (2021) Performance and emission predictions of a CRDI engine powered with diesel fuel: A combined study of injection parameters variation and Box–Behnken response surface methodology based optimization, Fuel 290, 120069. [CrossRef] [Google Scholar]
  • How H.G., Masjuki H.H., Kalam M.A., Teoh Y.H. (2018) Influence of injection timing and split injection strategies on performance, emissions, and combustion characteristics of diesel engine fueled with biodiesel blended fuels, Fuel 213, 106e14. [Google Scholar]
  • Uslu S. (2020) Optimization of diesel engine operating parameters fueled with palm oil-diesel blend: Comparative evaluation between response surface methodology (RSM) and artificial neural network (ANN), fuel 276, 117990. [CrossRef] [Google Scholar]
  • Şimşek S., Uslu S. (2020) Investigation of the effects of biodiesel/2-ethylhexyl nitrate (EHN) fuel blends on diesel engine performance and emissions by response surface methodology (RSM), Fuel 275, 118005. [CrossRef] [Google Scholar]
  • Polat S. (2016) An experimental study on combustion, engine performance and exhaustemissions in a HCCI engine fuelled with diethyl ether–ethanol fuel blends, Fuel Process Technol. 143, 140–150. [CrossRef] [Google Scholar]
  • Paul A., Panua R.S., Debroy D., Bose P.K. (2014) Effect of compressed natural gas dual fuel operation with diesel and pongamia pinnata methyl ester (PPME) as pilot fuels on performance and emission characteristics of a CI (compression ignition) engine, Energy 68, 495–509. [CrossRef] [Google Scholar]
  • Pulkrabek W.W. (1997) Engineering Fundamentals of the Internal Combustion Engine, Pearson, New Jersey, USA, pp. 229–261. [Google Scholar]
  • Hoang A.T. (2021) Combustion behavior, performance and emission characteristics of diesel engine fuelled with biodiesel containing cerium oxide nanoparticles: A review, Fuel Process Technol. 218, 106840. [CrossRef] [Google Scholar]
  • Cheikh K., Sary A., Khaled L., Abdelkrim L., Mohand T. (2016) Experimental assessment of performance and emissions maps for biodiesel fueled compression ignition engine, Appl. Energy 161, 320–329. [CrossRef] [Google Scholar]
  • Yaşar A., Keskin A., Yıldızhan Ş., Uludamar E. (2019) Emission and vibration analysis of diesel engine fuelled diesel fuel containing metallic based nanoparticles, Fuel 239, 1224–1230. [CrossRef] [Google Scholar]
  • Yusaf T.F., Buttsworth D.R., Saleh K.H., Yousif B.F. (2010) CNG-diesel engine performance and exhaust emission analysis with the aid of artificial neural network, Appl. Energy 87, 5, 1661–1669. [CrossRef] [Google Scholar]
  • Hazar H., Tekdogan R., Sevinc H. (2021) Investigating the effects of oxygen enrichment with modified zeolites on the performance and emissions of a diesel engine through experimental and ANN approach, Fuel 303, 121318. [CrossRef] [Google Scholar]

All Tables

Table 1

Fuel properties of diesel, canola biodiesel, and methane.

In the text
Table 2

Engine technical specifications.

In the text
Table 3

Input factors and their levels.

In the text
Table 4

Test and predict values of engine responses for NOx, CO, CO2, and O2 emissions for canola biodiesel.

In the text
Table 5

Test and predicted values of responses for VN for canola biodiesel.

In the text
Table 6

Model evaluation for canola biodiesel.

In the text
Table 7

ANOVA method for exhaust emissions.

In the text
Table 8

ANOVA method for engine VN.

In the text
Table 9

Optimisation of test setup and desirability of engine responses.

In the text
Table 10

Validation of engine test results.

In the text

All Figures

thumbnail Fig. 1

Schematic view of engine test setup.
In the text
thumbnail Fig. 2

Flowchart of RSM.
In the text
thumbnail Fig. 3

Impacts of input parameters on NOx emissions.
In the text
thumbnail Fig. 4

Impacts of input parameters on CO exhaust emissions.
In the text
thumbnail Fig. 5

Impacts of input parameters on CO2 emissions.
In the text
thumbnail Fig. 6

Impacts of input parameters on O2 emissions.
In the text
thumbnail Fig. 7

Impacts of input parameters on engine vibration.
In the text
thumbnail Fig. 8

Impacts of input parameters on noise.
In the text
thumbnail Fig. 9

Optimisation of NOx, CO, CO2, O2 exhaust emissions and engine NV.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.