Reduction of fuel consumption using driving strategy in agricultural tractor

During the acquisition and use of an agricultural tractor, the consumption of fuel by the engine must be taken into account by the user, considering its nonrenewable origin and the marketing price. Appropriate strategies in driving tractors can help users to reduce production costs. In this context, the objective of this study was to evaluate the fuel consumption of an agricultural tractor using different driving strategies. Load levels were imposed, following the instructions from Organization for Economic Co-operation and Development Code 2, using an instrumented dynamometer car, in a concrete test track. Six load levels (30, 40, 50, 60, 70, and 80% of Q0) and three travel speeds (5.16, 7.29, and 10.48 km h-1) were applied for the strategies used (Full Throttle and Shifted Up Throttle Back), where Q0 is the traction force corresponding to the maximum power in the drawbar for each gear selected. Three repetitions were made, which totaled 108 experimental units (2 × 3 × 6 × 3), in a completely randomized experimental design. Shifted Up Throttle Back mode should be used as a strategy of tractor driving, because it can save up to 29.39% of fuel in relation to the Full Throttle mode, typically used by users.


Introduction
Alternative managing options for mechanized agricultural operations are being searched for, more specifically for the main source of power for them, the farm tractor, because of the costs of fuel, pollution generated by their combustion, and the necessity of diminishing energy consumption (Frantz et al., 2014).Tractor use in rural areas is very diverse, and the search for the optimized performance is justifiable (Vale et al., 2011).
While acquiring a tractor, the user usually considers technical and economic criteria; however, in addition, energy efficiency is another important factor to be analyzed (Silveira & Sierra, 2010).Considering that the cost of the energy used in machine operations has a significant impact on the total cost of mechanization, it can be reduced through improvement of the operational procedures (Stange et al., 1984); most of these procedures can be done by the owner, because they are about changing the functioning parameters of the machines (Márquez, 2012).
In this way, there is a necessity for knowledge about factors that may interfere with fuel consumption, so it can be as efficient as possible.Thus, according to Kim et al. (2013), it is important to analyze the effects of gear selection during mechanized agricultural operations.Different consumptions may be obtained for the same type of work, depending on the gear being used (Hansson et al., 2003).
Driving strategies are the answer to manage engine and transmission, aiming to reduce fuel consumption, and therefore, obtain higher efficiency in diesel oil use (Howard et al., 2013).This technique was described by Grisso et al. (2014a) as "Gear Up -Throttle Down" or "Shift Up -Throttle Back" (SUTB).According to the authors, this is a fuel-saving practice that can be used when the load in the drawbar power is lower (less than 75% of nominal power) and power take-off (PTO) speed is reduced.
Because of the necessity of obtaining more energy efficiency in agricultural use of tractors, the objective of this study was to evaluate the fuel consumption in a tractor using different driving strategies.

Research tractor
In the experiment conducted in a concrete test track at the Mechanic Agriculture Station, Madrid,Spain (40º 21' 34.77'' N;3º 43' 20.38'' W;613 m),Beauvais,France) was used, equipped with a four-stroke diesel cycle engine (66 AWI 695) from the same manufacturer, with six cylinders, a dislocated volume of 6.596 cm³, and a turbocharger with an intercooler.According to the test report, its maximum power and torque are 105.3kW at 1998 rpm and 615.8 Nm at 1100 rpm, respectively (IRSTEA, 2013).The power transmission system was equipped with a hydrostatic gearbox-type Powershift, with 24 gears forward and 24 reverse.The tractor had a total weight of 7,660 kg (75.14 kN), with a static mass distribution of 61% on the rear axle and 39% on the front axle.

Data acquisition
Fuel consumption was measured by the volumetric flow meter installed in the dynamometer car connected to the tractor, which applied level-controlled loads in the drawbar and housed the measuring device.The car braked the test tractor by means of a hydraulic circuit with a variable displacement pump.The car braking was given through a hydraulic circuit with a variable displacement pump.The dynamometer car had a maximum traction capacity of 15,000 kgf (147.15kN), and its mass for the experiment's performance was 17,85 kg (175.11kN).The experimental layout is shown in Figure 1.
The tractor's fuel consumption data were sent to the dynamometer car computer in real time every second, through a data transfer center positioned within the tractor cab.Before the experiment, the engine of the tractor had been run for 30 h using an electric dynamometer brake connected to the tractor by the PTO, where a load equivalent to 30% the nominal power provided by the manufacturer was applied.

Test method and procedure
Because of the engine speed and the transmission relation of each gear, driving strategies were determined for the tractor.The gears 8 (2B), 10 (2D), and 14 (3B) (Figure 2) were used, which, for maximum engine speed without load (2,153 rpm), corresponded to travel speeds of 5.16, 7.29, and 10.48 km h -1 , respectively.The travel speed variation of 5 and 11 km h -1 was proposed to include a vast diversity of agricultural operations performed on the field.
To balance the same travel speeds proposed (5.16, 7.29, and 10.48 km h -1 ), a lower engine speed was selected (1,524 The load levels applied in the tractor were obtained from previous official tests, following the instructions from the Organization for Economic Co-operation and Development (OECD, 2014).For each gear selected, an initial engine speed of 2,153 rpm determined the traction force (Q0) corresponding the maximum power in the drawbar.Then, traction forces of 61.14 kN for gear 2B, 1,995 rpm from the engine; 46.43 kN for 2D, with 1,984 rpm; and 34.71 kN for 3B, with 1,978 rpm were obtained.
Afterward, for the two driving strategies evaluated, six load levels were applied to each gear (30, 40, 50, 60, 70, and 80% of Q0).Because of the electronic fuel injection system, even with a lower engine speed and longer gears, the tractor withstood the load levels applied.

Experimental and statistical procedures
The hourly (L h -1 ) and specific (g kW -1 h -1 ) fuel consumption of the engine were analyzed according to a design based on three factors, resulting in two driving strategies, three travel speeds, and six load levels applied to the tractor.Three repetitions were made, which totaled 108 experimental units (2 × 3 × 6 × 3), in a completely randomized experimental design, due to surface uniformity, using a concrete test track.
First, data were analyzed regarding its normality and homoscedasticity.Then, variables were submitted to an analysis of variance (ρ ≤ 0.05).When there was significance, the averages were analyzed by a Tukey test (ρ ≤ 0.05).For this purpose, Sisvar version 5.3 software was used (Ferreira, 2011).

Results and Discussion
After a variance analysis of the results of hourly and specific fuel consumption for the driving modes, travel speeds, and load levels evaluated, it was verified that the variables presented statistical difference (Table 1).Table 1.Summary of the analysis of variance for the hourly (L h -1 ) and specific (g kW -1 h -1 ) fuel consumption parameters

Hourly fuel consumption
The use of the driving strategy SUTB had the lowest consumption, in all load levels and travel speeds, when compared to FT (Figure 3).Depending on the project of the engine and other controlling factors, such as the fuel supply system, this driving strategy can provide fuel savings of 12 to 30%.Even with a low traction force, the fuel consumption for the same power was higher when engine speed was elevated.
For each travel speed, while the average load levels are incremented, the fuel consumption increases.According to Howard et al. (2013), the hourly fuel consumption is a linear function of the drawbar power (Figure 3).respectively.Transitory load levels applied on the tractor engine significantly affect fuel consumption, and for agricultural operations where the engine speed and torque variations are small, the effect on fuel efficiency is lower (Hansson et al., 2003).
The results presented in Table 2 show that, for all combinations, there was a difference between the two driving modes and six load levels, where hourly consumption varied from 10.86 L h -1 for SUTB, at 5.16 km h -1 , to 28.27 L h -1 for FT at 10.48 km h -1 .With the tractor being driven SUTB, engine fuel consumption was reduced, providing savings of up to 29.39% of fuel when compared with FT.
This percentage difference is lower as the applied load levels increase (Figure 3); however, it is statistically different between the modes.According to the American Society of Agricultural Engineers (ASAE, 2006), fuel consumption varies depending on its type, density, and viscosity and on engine applied load level percentage.
For a tractor operating with a load level of 50% of the maximum drawbar power, at a travel speed of 7.29 km h -1 , the potential fuel saving is 18.38% (20.24L h -1 with FT as opposed to 16.52 L h -1 with SUTB).The annual fuel saving from different driving strategies can be estimated by the multiplication of the total number of hours for which the tractor is used annually in agricultural operations by the difference of fuel consumption (Grisso et al., 2014b).
Considering the price paid for one liter of mineral diesel, the consumption reduction, and consequently, the cost of tractor work are important to justify the use of the SUTB strategy instead of FT.In addition, fuel combustion is an important source of pollutant gas emission (Estrada et al., 2016), resulting in considerable impact on the environment and human health (Lindgren et al., 2011).

Specific fuel consumption (SFC)
SFC was reduced while the applied load level was increased in the tractor (Figure 4), corroborating data obtained by *Averages followed by the same Greek letter in the column between driving strategies, capital letters in the column between travel speeds for the same driving strategy, and lowercase letters in the line do not differ between each other by a Tukey test, considering the nominal value of significance at 0.05  2. Hourly (L h -1 ) and specific (g kW -1 h -1 ) fuel consumption averages for Full Throttle (FT) and Shifted Up -Throttle Back (SUTB) driving strategies, in all three travel speeds and six load levels applied to the tractor From the regression equations presented in Figure 3A, an increase of 1.79 and 2.15 L h -1 is observed for each 10% increase in load levels applied on the tractor, for FT and SUTB, It can also be concluded that there is no difference in SFC for heavy load levels (70 and 80%), except in FT at 10.48 km h -1 (Table 2).According to Márquez (2012), SFC helps assess engine efficiency -in other words, the work that can be produced (kW h -1 ) from a gram of fuel, regardless of the available engine power.

Modes
A tendency is observed where, the higher the travel speed, the lower the specific fuel consumption of the engine (Figure 3), but there is a limit in relation to the dynamic traction coefficient, i.e., the traction force to the adherent mass ratio, for each selected work gear.This behavior was found by Lopes et al. (2003), where they report that the gear factor had significant influence on specific fuel consumption, with the variable decreasing while travel speed increased.Monteiro et al. (2011) obtained a saving of 9.5% in SFC when the travel speed was increased from 6.5 to 7.5 km h -1 , through a change of gear from B2 to B3, showing that the travel speed increase improved the tractor's energy yield.In a study performed by Jenane et al. (1996), seventh gear offers a greater opportunity to operate the tractor more efficiently than the fourth gear.Toledo et al. (2010) affirm that mechanized agricultural operations must be planned rationally for an increased profitability in the field.

Conclusions
1.The knowledge about and use of driving strategies in tractors are alternatives that can make, in general, the rate of fuel spent be reduced, making farm production more profitable and sustainable.
2. SUTB mode must be used as a driving strategy in tractors, because, under the conditions of this experiment, savings of up to 29.39% of fuel can be obtained when compared with FT, which is normally used by farmers.the ASAE (2006).According to this standard, the SFC is a nonlinear function from the drawbar power.In Figure 4, the high values of the coefficient of determination (R²) confirm the adjustment of the polynomial regression curves.
For all treatments evaluated, the SUTB strategy was statistically lower than FT regarding specific fuel consumption, determined by a Turkey test at 0.05 significance (Table 2).For drawbar power levels between 35 and 50% of maximum power, depending on the travel speed, the SUTB driving strategy of the tractor with Powershift transmission is more efficient than in a tractor with continuously variable transmission (Howard et al., 2013).
For the same travel speed, as load levels increase, the difference in SFC and driving strategies is reduced.This tendency was observed between the three travel speeds used, where the higher the travel speed, the lower the difference (Figure 4).
While analyzing driving strategies in each load level, it is possible to observe that, for all three travel speeds evaluated, the higher the load level applied on the tractor, the lower the specific fuel consumption.According to Acuña et al. (1995), the lowest values of specific fuel are caused by the increased power required of the tractor.Lower specific fuel values mean simultaneous optimization of engine performance, traction efficiency, and the adequacy of the implement for the tractor (Lyne et al., 1984).

Figure 1 .
Figure 1.Schematic representation of the experiment to obtain the parameters of fuel consumption efficiency

Figure 2 .
Figure 2. Engine speeds and work gears used that configure Full Throttle (FT) and Shifted Up -Throttle Back (SUTB) driving modes, as well as travel speeds, for the MF 7616 Dyna-6 tractor (Massey Ferguson, Beauvais, France) Travel speed (km h -1 ) Engine speed (rpm)

Figure 3 .
Figure 3. Hourly fuel consumption regarding the load levels applied to the tractor using two driving strategies for three travel speeds: (A) 5.16, (B) 7.29 and (C) 10.48 km h -1

Figure 4 .
Figure 4. Specific fuel consumption (SFC) regarding the load levels applied to the tractor, using two driving strategies for three travel speeds: (A) 5.16, (B) 7.29 and (C) 10.48 km h -1