Design & CFD analysis of baffles of fuel tanker trucks for normal and grade highway conditions

Design & CFD analysis of baffles of fuel tanker trucks for normal and grade highway conditions

Abstract: Liquid sloshing is a kind of wave motion inside a partially filled tank. Fluid sloshing affects the stability of fuel tanker truck, because of the movement of the center of gravity during different dynamic conditions on regular as well as on the grade highways. Hence controlling sloshing is of prime importance for the prevention of accidents and casualties. For the present work, the tanker truck filled with kerosene fuel has been considered. The baffles are designed to control the phenomenon of sloshing to a maximum extent. The Computational Fluid Dynamics (CFD) analysis of the tank with and without baffles is performed in the commercial code of ANSYS Fluent® 16.0 with appropriate boundary conditions to study the amplitude of sloshing at different time instants. The baffles are designed in such a way that there would be no hindrance during loading and unloading of the tanker truck.

Keywords: Computational Fluid Dynamics (CFD); tanker truck; kerosene sloshing; rollover; baffles

Reference to this paper should be made as follows: Author. (xxxx) ‘Title’, Int. J. xxxxxxxxxxx xxxxxxxxxxx,

This paper is a revised and expanded version of a paper entitled [Designing Baffles of Fuel Tanker Trucks to Prevent Rollover] presented at [15TH International Conference on Humanizing Work and Work Environment, HWWE2017 at Aligarh Muslim University on December 8-10, 2017].

Table: 1. Nomenclature.

θ Roll angle

θ_p Pitch angle

a_z Lateral acceleration

a_x Longitudinal acceleration

g Acceleration due to gravity

tan∅ The slope of the free liquid surface

F Force

M Turning moment

m Mass

v ⃗ Velocity vector

ρ Density of kerosene

p Pressure

μ Dynamic viscosity

m ̇_pq Mass transfer from phase p to phase q

m ̇_qp Mass transfer from phase q to phase p

1 Introduction

India being the world’s third-largest importer of oil, distribute its oil products throughout the country mainly by tanker trucks According to 2016 open government data survey of India, about 202.9 million metric tons of crude oil was imported by India in 2015-2016. During transportation, various accidents have occurred resulting in loss of life and property. In those accidents, rollover instability was one of the major causes. A study by Department of Transportation, US (Pape et al., 2007) has reported that the average number of tanker rollover is about 1265 annually in the US, which accounts for 36.2% of the total number of heavy vehicle highway accidents. According to the data collected by Statistics Canada (Woodrooffe, 2000), it is shown that 83% of vehicles rollover accidents on the highway are caused by tanker trucks. There are many reasons responsible for tanker accidents such as over speeding, tire burst but the most common one is liquid sloshing (Kwon, 2011; Kang, 2001). According to the international guideline related with transportation of dangerous materials, for minimizing risks and to maintain internal pressure below safe limits in tanker trucks, the liquid to be transported is not filled completely. To prevent the free movement of the liquid and to minimize the effect of sloshing, internal barriers called baffles are used. As the computer technology has improved, various problems on liquid sloshing have been studied and a lot of investigations of the effect of baffles on sloshing have been done in past years (Akylidiz et al., 2006). (Eswaran et al., 2009) performed the Numerical analysis and verified the results by experimentally for the sloshing waves inside baffled and unbaffled cubic tank. (Salem et al. 2009) gave the literature review of fluid and container interactions, with the main focus on parameters that affect the stability of partly filled tanker trucks under various maneuvers. Since 1965, many studies have been performed to model the sloshing behavior inside the partially filled tanker trucks to optimize the shape and position of baffles (Liu and Lin, 2009). Numerical and CFD study has been performed during braking and lane change maneuvers of a tanker truck to evaluate the effect of fluid fill level, the position of baffles and shape of the tank (F. Cheli et al., 2013). For the quick prediction of rollover instability, a quasi-static roll plane model has been developed for a partially filled tank to calculate the forces and moments associated with the free liquid surface movement according to the longitudinal and lateral acceleration provided in the pitch plane and roll plane respectively (Ranganathan and Rakheja, 1993).

By using the law of conservation of momentum with a non-inertial frame of reference attached to the tank, the equation for the slope of the free liquid surface has been derived;

tan∅=(a_z/g-θ)/(1+a_z/g) , tan∅=(a_x/g-θ_p)/(1+a_x/g)

Figure 1. Schematic diagram indicating the shifting of the center of gravity and direction of force during cornering of a tanker truck.

For a given lateral acceleration, the resulting forces and turning moment can be calculated by using the following equations,

FZ = m*(az g) ; Fy = mg

Mx = Fy* Zcg + FZ* Ycg

After an experimental investigation it was found that during 40% to 70% liquid loading condition, reduction in overall height of center of gravity (C.G) of the tank takes place but during the cornering and braking, the resultant forces and rolls moment lead to a reduction in rolling stability and created unbalancing (Lloyd et al., 2002; Belakroum et al., 2010). The present work focuses on the design of lateral, longitudinal and bisectional baffles for a cylindrical shaped tanker truck, then to decide the best suitable baffle design for the safe and trouble-free transportation of liquid on normal as well as grade highways.

When tanker trucks go for the cornering on a normal highway the liquid bulk moves frequently and exerts the force on the side wall. During the uphill on grade highway or in hilly areas, the liquid mass used high forces on the rear wall of the tank while during downhill the sudden forces acting on front wall cause unbalancing and create the loss of directional control due to immediate impact and vibrations. Therefore, the CFD simulation is carried out for the design of baffles in such a way that tanker truck can moves smoothly on regular as well as grade highway. For the uphill and downhill, simulation is carried out considering a critical condition that is 35% grade areas like Baldwin Street, Dunedin, New Zealand, and Waipio Rd, Honokaa, Hawaii. The conversion chart from grade to the slope (angle) and symbol of grade highways are shown in Figure 2.

Figure 2. Grade highway symbol and conversion chart of grade percent to the angle of inclination.

2 Validation

For solving the actual problem, a validation case has been considered to examine the motion of the free surface of the liquid in a partially filled fuel rectangular tank (Singal et al., 2014). The same approach has been used to solve the problem. The result showed a close match as can be seen from Figure 4.

(a) Liquid interface at t=1.25 sec (b) Liquid interface at t=1.25sec

(c) Liquid interface at t=1.25 sec (d) Liquid interface at t=1.25 sec

Figure 4. (a), (c) represents the result obtained by Singal et al. and (b), (d) represent the results obtained during validation.

3 Methodology

The following cases have been considered for tanker truck,

1. Without baffles during acceleration.
2. Without baffles during cornering.
3. Lateral baffles during cornering.
4. Lateral baffles and one longitudinal baffle during acceleration.
5. Lateral baffles and one longitudinal baffle during cornering.
6. Lateral baffles and two longitudinal baffles during cornering.
7. Lateral and bi-sectional baffle during cornering (kerosene region 10m*2.5m*1.1m).

g’. Lateral and bi-sectional baffle during cornering (kerosene region 10m*2.5m*1.6m).

1. Without baffles during uphill on grade highways.
2. Lateral and bisectional baffles during uphill on grade highways.
3. Without baffles during downhill on grade highways.
4. Lateral & bisectional baffles during downhill on grade highways.
5. To check the design of bisectional baffle during unloading of a tanker truck.

The following assumptions have been considered,

The tank acceleration is assumed to be 9.8m/s2 during acceleration, braking and cornering on normal highways, while during uphill and downhill it is assumed to accelerate at 9.8 m/s2 and 5 m/s2 respectively.

Density and viscosity of kerosene are 750 kg/m3 and 0.0024 kg/m-s respectively.

4 Design

The geometry of the tank is generated by using commercial software Siemens NX-10. The geometry has the following specifications.

A cylindrical tank 10m long and circular cross-section of diameter 2.5m (see Figure 5).

Lateral baffles with central and semi-circular circumferential holes of diameter 0.5m respectively whereas longitudinal baffles with a bottom hole of diameter 0.5m (see Figure 5).

The bisectional baffle is 10m long with circular and rectangular holes as shown in Figure 6.

(b)

Figure 5. Tanker with (a) Lateral and longitudinal baffles (b) Lateral and bisectional baffle.

Figure 6. Top view of the bisectional baffle.

Figure 7. Back view of the bisectional baffle.

The holes on the bisectional baffle are designed in such a way that during loading of the fuel tanker truck, the bottom part of the tank should be filled first to evenly distribute the bulk. Generally, in tanker trucks, there are individual top man-hole of diameter 0.25 m to each compartment for trouble-free loading. So, during the top loading, firstly the fuel will be collected on the bisectional baffle. Therefore, four holes with 0.25 m diameter are designed for each compartment to keep the volume flow rate constant.

The smallest angle needed to get the water drop to roll is 90 (Courbin, 2016). The shape of the bisectional baffle is decided to ensure that not a single drop of kerosene would accumulate over the bisectional baffle. It was found that when the shape of the bisectional baffle was flat, some of the fuel got accumulated at different corners of the tank. Hence, curved bisectional baffle making a slope of 90 (as shown in figure 7), is incorporated such that every single drop of kerosene liquid should come down to the base of the tank. The rectangular holes of dimension (0.5m * 0.1m) are also given to prevent the kerosene accumulation in between the circular holes, as shown in Figure 6.

5 Computational study and Modeling

ICEM-CFD was used to mesh the computational domain of the tank. Unstructured mesh with tetrahedral elements was used with a number of elements as 0.88 million and a number of nodes as 0.14 million after the grid independence test (see Figure 8).

Figure 8. Mesh of the computational domain.

ANSYS FLUENT® 16.0 commercial package was used to perform the CFD simulations by using VOF (volume of fluid) multi-phase model, k-epsilon viscous model. The energy equation is turned off because there is no exchange of heat energy or the process is assumed isothermal.

5.1 Governing Equations

Continuity Equation:

∇.v ⃗ = 0

Momentum Equation:

∂/∂t (ρv ⃗ )+∇.(ρv ⃗v ⃗ )=-∇p+∇.[μ(∇v ⃗+∇v ⃗^T )]+ρg ⃗+F ⃗

Volume Fraction Equation:

1/ρ_q [∂/∂t (α_q ρ_q )+∇.(α_q ρ_q v ⃗_q )=S_(α_q )+∑_(p=1)^n▒( m ̇_pq-m ̇_qp)]

5.2 Solver Details

Pressure-Velocity Coupling Scheme Fractional step (NITA)

Spatial discretization of Gradients Green-Gauss node based

Spatial discretization of Pressure Body force weighted

Spatial discretization of Momentum Second order upwind

Spatial discretization of Volume fraction Geo-Reconstruct

Transient Formulation First order implicit

6 Results and discussions

The liquid interface of the kerosene in the tanker truck during different dynamic conditions on normal highways with kerosene liquid region (10 m*2.5 m*1.1 m).

Tanker without baffles during acceleration.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

Tanker without baffles during cornering.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

(c) Tanker with lateral baffles during cornering.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

(d) Tanker with lateral and one longitudinal baffle during acceleration.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t =1.20 sec

(e) Tanker with lateral baffles and one longitudinal baffle during cornering.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

(f) Tanker with lateral and two longitudinal baffles during cornering.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

(g) Tanker truck with lateral & bisectional baffles during cornering.

t = 0.30 sec t = 0.60 sec t = 0.90 sec t = 1.20 sec

(g’) Tanker truck with lateral & bisectional baffle during cornering with kerosene liquid region (10m*2.5m*1.6m).

t = 0.25 sec t = 0.50 sec t = 0.75 sec t = 1.0 sec

Liquid interface of kerosene in the tanker truck on 35 percent grade (20 degree) highway with kerosene liquid region (10m*2.5m*1.6m).

(h) Tanker truck without baffles during uphill

t = 0.25 sec t = 0.50 sec t = 0.75 sec t = 1.0 sec

(i) Tanker truck with lateral & bisectional baffles during uphill

t = 0.25 sec t = 0.50 sec t = 0.75 sec t = 1.0 sec

(j) Tanker truck without baffles downhill

t = 0.25 sec t = 0.50 sec t = 0.75 sec t = 1.0 sec

(k) Tanker truck with lateral & bisectional baffle downhill

t = 0.25 sec t = 0.50 sec t = 0.75 sec t = 1.0 sec

(l) To check the design of bisectional baffle during unloading of fuel (kerosene region is 10m*2.5m*0.15m).

t = 0 sec t = 2.0 sec t = 5.0 sec t = 9.5 sec

The motion of fluid inside the tanker is found to be changed depending on the orientation of baffles. Initially, the fluid motion without baffles was observed during acceleration, braking, and cornering. It was found that during acceleration and braking with four lateral baffles inside tanker the fluid didn’t hit the front wall which reduces the impact and helps in comfortable braking without skidding. During turning while braking, the kerosene rotates inside the tanker which induces high cornering moment that may lead to the toppling of the tanker. When four lateral baffles are used there was a rotation of kerosene in each compartment. Although the cornering moment was found to be lower than the initial value, the magnitude was significantly higher. Using a baffle passing through the center longitudinally further reduced the turning moment, bringing it down to the safe range. The turning moment further reduced by using the 2 longitudinal baffles which caused the kerosene to rise from both the directions inside the tanker after striking the longitudinal baffle thereby canceling out turning moment.

After implementing the bisectional baffles, the movement of kerosene liquid is minimum during different dynamic conditions on normal highways as with small shift in CG there is a huge reduction of forces and turning moment on the side walls. When the volume of kerosene is just less than half of the tank, bisectional baffles reduce the free space for movement of the bulk, which results in a very small change in position of the center of gravity as shown in case (g). When volume of the kerosene is more than half of the tank then the bulk which is below the bisectional baffle is always at rest due to no free space for movement, while in the upper part kerosene liquid slosh but overall Centre of gravity is balanced by the bulk below bisectional baffle as showed in case (g’). During uphill and downhill, in case of the tank without baffle, the liquid bulk hits the front and rear wall respectively and exerts a sudden force which makes the tanker truck unbalance. Bisectional baffle resists the motion of kerosene and reduces free space which decreases the amount of sudden force on the wall. Improvised design of bisectional baffle prevents fuel accumulation and helps in trouble-free unloading of fuel from the tanker at rest. CFD simulation shown in case (l) shows that during unloading, whole kerosene bulk comes down to the bottom of the tank within 10 sec, which proves that the design of baffle is much effective.

Figure 9. Graph of force exerted on the side wall at different time instants during cornering.

Figure 10. Graph of turning moment acting at different time instants during cornering.

Figure 11. Graph of force exerted on the rear wall at different time instants during uphill.

Figure 12. Graph of force exerted on the front wall at different time instants during downhill.

7 Conclusion

During the comparison of the tank-without-baffles and tank with four lateral and a bisectional baffle, it has been observed that during cornering on normal highways there was 50% and 130% reduction in maximum force and turning moment acting on tank wall respectively (see Figure 9, 10). This is the maximum reduction among all the cases studied for different baffle orientations comparison. During uphill and downhill, bisectional baffle resists the kerosene liquid to move beyond it so change in the C.G of the whole system is considerably small in comparison to the tanker with longitudinal baffles. For uphill and downhill motion, the peak force for tanker without baffles is very high in between 0.7-0.8 seconds of fuel motion inside the tank which results in impact loading causing sudden instability, making the vehicle more prone to a sudden accident. The peak force is found to reduce to a greater extent for both up and downhill motion when we used bisectional baffles. The low-intensity impact is observed during uphill motion making it safer for truck motion (see Figure 11). Although the constant increase in force is observed for downhill, there is no sudden impact during downhill and graph is smoother which makes it easy for the driver to handle the truck (see Figure 12). Bisectional baffle helps in reducing the sloshing phenomenon to a great extent by constraining the motion of liquid in a fixed space. This type of baffle design will help in countering the problems in automotive as well as transportation industries.

It can be inferred that the use of baffles can significantly reduce the force and cornering moment acting on the tank wall which is the major cause of toppling and thereby will minimize accidents and casualties caused by tanker truck rollovers.

8 References

Akyildız, H., & Ünal, N. E. (2006). Sloshing in a three-dimensional rectangular tank: numerical simulation and experimental validation. Ocean Engineering, 33(16), 2135-2149.

Belakroum, R., Kadja, M., Mai, T. H., & Maalouf, C. (2010). An efficient passive technique for reducing sloshing in rectangular tanks partially filled with liquid. Mechanics Research Communications, 37(3), 341-346.

Cheli, F., D’Alessandro, V., Premoli, A., & Sabbioni, E. (2013). Simulation of sloshing in tank trucks. International Journal of Heavy Vehicle Systems, 20(1), 1-18.

Courbin, L. (2016). Rethinking Superhydrophobicity. Physics, 9, 23.

Dai, L., & Xu, L. (2006). A numerical scheme for dynamic liquid sloshing in Lateral cylindrical containers. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 220(7), 901-918.

D’ALESSANDRO, V. I. N. C. E. N. Z. O. (2012). Modeling of tank vehicle dynamics by fluid sloshing coupled simulation (Doctoral dissertation, Italy).

Eswaran, M., Saha, U. K., & Maity, D. (2009). Effect of baffles on a partially filled cubic tank: Numerical simulation and experimental validation. Computers & Structures, 87(3), 198-205.

Hasheminejad, S. M., & Aghabeigi, M. (2009). Liquid sloshing in half-full Lateral elliptical tanks. Journal of sound and vibration, 324(1), 332-349.

Kang, X. (2001). Optimal tank design and directional dynamic analysis of liquid cargo vehicles under steering and braking (Doctoral dissertation, Concordia University).

Kwon, C. (2011, December). Conditional value-at-risk model for hazardous materials transportation. In Simulation Conference (WSC), Proceedings of the 2011 Winter (pp. 1703-1709). IEEE.

Liu, D., & Lin, P. (2009). Three-dimensional liquid sloshing in a tank with baffles. Ocean engineering, 36(2), 202-212.

Lloyd, N., Vaiciurgis, E., & Langrish, T. A. G. (2002). The effect of baffle design on longitudinal liquid movement in road tankers: an experimental investigation. Process Safety and Environmental Protection, 80(4), 181-185.

Mucino V. H. Saunders E. Salem, M. I. and M. Gautam. Lateral sloshing in partially filled elliptical tanker trucks using a trammel pendulum. In. J. Heavy Vehicle Systems, Vol. 16, Nos. 1/2, pp. 207-224, 2009).

Pape, D. B., Harback, K., McMillan, N., Greenberg, A., Mayfield, H., Chitwood, J. C., … & Brock, J. (2007). Cargo tank roll stability study.

Rakheja, S., & Ranganathan, R. (1993). Estimation of the rollover threshold of heavy vehicles carrying liquid cargo: a simplified approach. International Journal of Heavy Vehicle Systems, 1(1), 79-98.

Singal, V., Bajaj, J., Awalgaonkar, N., & Tibdewal, S. (2014). CFD analysis of a kerosene fuel tank to reduce liquid sloshing. Procedia Engineering, 69, 1365-1371.

Wang, W., Zhang, W., Guo, H., Bubb, H., & Ikeuchi, K. (2011). A safety-based approaching behavioural model with various driving characteristics. Transportation research part C: emerging technologies, 19(6), 1202-1214.

Woodrooffe, J. (2000). Evaluation of dangerous goods vehicle safety performance. Report TP, (13678-E).