Abstract
In order to reduce the influence of fuel sloshing on the interior noise of a passenger car, the fuel tank sloshing noise was first evaluated with a subjective evaluation method to determine the driving cycle of the car and the fuel filling percentage of the fuel tank in which the fuel tank sloshing noise is serious, and then two anti-wave boards with different structural characteristics were designed to reduce the fuel sloshing. On this basis, fuel sloshing in the fuel tank equipped with the newly designed and original anti-wave boards was simulated numerically; then, the anti-wave board with the best effect of inhibiting fuel from sloshing was selected based on the numerical results; finally, the anti-sloshing effect of the selected board was evaluated through the car road test. The test results show that the vibration acceleration magnitude at each monitoring point of the tank with the selected anti-wave board is significantly reduced compared with the original fuel tank, which indicates that the selected anti-wave board inhibits fuel from sloshing greatly.
Introduction
As there is increasingly improved attention to the automobile ride comfort and the application of low-noise engine and transmission system, the reduction of the fuel sloshing noise becomes one of the crucial noise, vibration, and harshness (NVH) issues.1,2 Liquid fuel in a partially filled automotive tank oscillates when subjected to sudden acceleration or deceleration. This is called liquid sloshing, which is a source of noise generation in an automobile.3–5 Due to sloshing, complex surface waves are generated and dynamic forces are exerted on the tank walls. This results in noise generation that is typically referred as sloshing noise. In order to control the liquid fuel sloshing, it is now a general method to try many different design schemes of the fuel tank structure and repeat the same tests so that a better scheme can be determined; this method is expensive and time-consuming. In recent years, numerical simulation method has been widely used in the control of the fuel sloshing in an automotive tank for it can make up for the shortcomings of test methods, such as long test cycle and high cost. Kamiya et al. 6 used fluid–structure coupling method to simulate fuel sloshing in the fuel tanks with or without baffles in different fuel filling ratios. The results from the analysis indicated that the wavelength and the amplitude of sloshing wave in the fuel tank without baffles are larger than those with baffles. Zhang and He 7 analyzed the influences of tank filling ratios, tank acceleration and the acceleration duration, and the pressure inside the tank on fuel sloshing based on the volume of fluid (VOF) model. Balthy et al. 8 used STAR-CCM+ software to simulate the fuel sloshing in the plastic and metal fuel tanks with or without baffle, respectively, and determined the position where the fuel sloshing has the largest impact on the fuel tank wall so that the monitoring points on the fuel tank surface can be arranged for radiated noise test. Hallez et al. 9 simulated the radiated noise of an automotive fuel tank with fluid–structure coupling method based on computational fluid dynamics (CFD), structural dynamics, and acoustic theory and concluded that it is feasible to use the numerical average kinetic energy from CFD simulation to evaluate fuel tank sloshing noise. Hoi Sum et al. 10 studied the fuel sloshing and the sloshing noise of fuel tanks with different structural characteristic baffles by combining numerical simulation with bench test, and the results show that baffles have good effects of inhibiting fuel from sloshing. In 2013, Vytla and Ando 11 used the SC/Tera and Abaqus software to simulate the fuel sloshing and the structural dynamics of a fuel tank with 40% and 53% fuel filling percentage under braked heavily and lightly conditions. The results show that the tank deformation will remain stable with the change of the filling level and has little influence on the fuel sloshing. Qi et al. 12 used STAR-CCM+ and ACTRAN, which calculated the fuel sloshing in the fuel tank under acceleration and deceleration conditions, and the results show that the fuel sloshing was sharp when the upper and lower structures of the fuel tank were gentle and no baffle was set. Zhang, 13 based on Flow-3D to simulate the fuel sloshing in the helicopter fuel tank, finally obtained the pressure distribution and the change of the fuel tank gravity during the sloshing process. Xu et al. 14 carried out a simulation analysis on the fuel sloshing in the fuel tank for starting, braking, bumping, and turning of a certain type of forklift. Marriott et al. 15 combined Lagrange and Euler solvers to simulate the fuel sloshing in the fuel tank. The results showed that the installation baffle had a good effect on inhibiting fuel sloshing and noise propagation.
Aiming at the control of the sloshing noise of a fuel tank used in the passenger car, the subjective evaluation method is first adopted to evaluate the noise caused by fuel sloshing inside the crew compartment so that the driving cycle of the car and the fuel filling ratio of the fuel tank in which the fuel tank sloshing noise is serious are determined. Then, two new anti-wave boards are designed and the fuel sloshing in the tanks equipped with the new anti-wave boards and the original one is simulated numerically to determine the anti-wave board structure scheme with which the sloshing noise can be controlled effectively. Finally, the anti-sloshing effect of the selected board is evaluated through the car road test. The conclusions of this article have important reference values to the fuel sloshing noise control of similar vehicles.
Subjective evaluation of fuel sloshing noise of a passenger car
In order to determine the driving cycle of the car and the fuel filling percentage in which the fuel sloshing noise is serious, a subjective evaluation road test of the fuel sloshing noise is carried out. The test conditions are consistent with requirements by GB/T12534-90 “Motor vehicles—General rules of road test method.” The tested vehicle is under good condition without abnormal noise generation. The testers included vehicle driver and three professional evaluators. Subjective scoring is as follows: 10 points for full marks, representing no fuel sloshing; 1 is the lowest score, meaning sloshing noise is very serious and unacceptable. The total capacity of the fuel tank is 82 L, and the rated volume is 67 L. The fuel filling percentage of the fuel tank for experimental test is set to be four levels: 25%, 50%, 75%, and 100% of the rated volume. And the test cycle consists of tip in–out, rapid deceleration, reversing, and creeping driving of the car.
The subjective evaluation scores of the fuel sloshing noise are shown in Table 1. It can be seen from the table that the fuel sloshing noise is the most serious when the filling percentage of the fuel tank is 100% of the rated volume and car operates under creeping driving condition. The numerical simulation of fuel sloshing and experiments discussed in the following sections are carried out under this condition.
Subjective scoring results of the fuel sloshing noise.
Control of fuel sloshing
In order to reduce the fuel sloshing in the fuel tank, two anti-wave boards with different structural characteristics were designed, as shown in Figure 1. When the fuel in the fuel tank sways, part of the energy is used to overcome the resistance of the fuel itself, and the other part causes momentum loss and heat dissipation due to impact on the wall surface of the tank. The influence of the anti-wave board on the fuel sloshing in the fuel tank mainly includes the boundary layer damping to increase the energy loss, the internal damping to change the flow rate, and the flow damping to block. Therefore, installing an anti-wave board in the fuel tank can suppress the sloshing of the fuel. No. 1 is the original anti-wave board, and No. 2 and No. 3 are newly designed anti-wave boards. The thickness of both No. 1 and No. 2 is 3 mm, and the thickness of No. 3 is 2 mm. The length × width of the three kinds of anti-wave boards are 680 mm × 317 mm, 866 mm × 356 mm, and 852 mm × 366 mm; the diameter of the bleed holes are 20, 15.5, and 10 mm; and the number of drain holes are 23, 66, and 183, respectively. The main difference between the three types of anti-wave boards is the length of the anti-wave board, the diameter, and number of the bleed holes. Figure 1(d) shows the fuel tank geometry model. The length × width × height of the fuel tank is 1119 mm × 465 mm × 212 mm, and its thickness is 5.5 mm.
Geometric model anti-wave boards and fuel tank: (a) No. 1, (b) No. 2, (c) No. 3, and (d) the geometry of fuel tank.
Numerical simulation of fuel sloshing in the fuel tank
Simulation modeling
The coordinate system of the geometric and mesh models is defined as follows: the negative direction of X-axis is the direction of the car moving, and the positive direction of Y-axis points the right side of the car. The grids of the fuel tank and its new and original anti-wave boards generated by STAR-CCM+ software are shown in Figure 2. The element type is polyhedral, and the total number of elements is about 4.1 million.
Mesh models of the fuel tank and its anti-wave boards: (a) mesh model of fuel tank, (b) mesh of No. 1, (c) mesh of No. 2, and (d) mesh of No. 3.
Fuel gravity is considered in the simulation, and the VOF model is adopted to simulate the distributions of the fuel phase and the air phase in the fuel tank. The properties of fuel and air are shown in Table 2.
Properties of fuel and air used in the model.
The boundary conditions, as shown in Figure 3, are imposed to the numerical models to represent the motion of the tank. These boundary conditions are obtained from the subjective evaluation test, and the initial car speed is 4.5 km/h. Based on the acceleration curve of the vehicle and the initial velocity, the velocity equation of the fuel tank is solved. The custom function in the STAR-CCM+ toolbar is used to define the movement of the fuel tank. The custom function written is shown in Figure 4.
Motion boundary conditions of the fuel tank.
Custom motion boundary.
Numerical results and analysis
Figure 5 shows the distribution of the air phase and the fuel phase in the tank installed with different anti-wave boards. Some conclusions drawn from the figures are as follows:
From 0 to 1 s, the fuel tank is decelerated from the initial speed 4.5 km/h until stopped; at the same time, the fuel flows from the right side to the left side of the tank and finally gathers in the left side and the middle of the fuel tank due to the fuel’s inertia and the anti-wave board’s obstruction. The No. 3 anti-wave board obstructs the fuel flow strongest among the three kinds of boards. This is because the bleed hole of the No. 3 anti-wave board had the smallest diameter and the largest number. When the fuel flowed through the holes, the flow damping occurs, and the fuel velocity changed to cause retardation. And because the bleed hole acted as internal damping, its momentum dissipation is greater.
From 1 to 3 s, the fuel tank is in a static state, but the fuel continues flowing to the right side of the tank after hitting the left wall of the fuel tank and then hits the right side wall of the fuel tank at about 1.25 s; finally, the oscillation of the liquid fuel surface tends to stop. The impact of the fuel oscillation in the fuel tank with No. 2 anti-wave board on the wall of the fuel tank is the largest, while that of the tank with No. 3 anti-wave board is the smallest among the three tanks.
In the case of 3 to 4 s, the fuel tank is accelerated to 4.5 km/h from stop; at the same time, the fuel flows to the right side of the tank and finally gathers in the right side and the middle of the fuel tank due to the fuel’s inertia and the anti-wave board’s obstruction. The No. 3 anti-wave board obstructs the fuel flow strongest among the three kinds of boards. The No. 3 anti-wave board interrupts the continuity of the fuel movement well and serves as a good motion block.
From 4 to 5 s, the fuel tank is decelerated until stopped, but the fuel continues flowing to the left side of the tank after hitting the right wall of the fuel tank, and then hits the left side wall of the fuel tank; finally, the oscillation of the liquid fuel surface tends to stop.
Fuel sloshing vs time.
In summary, the No. 3 anti-wave board has the best ability to inhibit fuel sloshing among the three kinds of anti-wave boards. The results indicate that the main factor affecting the anti-swaying effect of the anti-wave board is the diameter and number of the bleed hole, and the length of the anti-wave board has almost no effect.
Road test of fuel tank sloshing
In order to check whether the fuel sloshing in the tank with No. 3 anti-wave board is reduced compared with the original fuel tank with No. 1 anti-wave board, the vibration acceleration of the tank surface is tested through the car road test.
Monitoring point arrangement
Four points on the tank surface, named as A1, A2, A3, and A4, are chosen as monitoring points where four acceleration sensors are installed to monitor the vibration of the fuel tanks as a result of fuel sloshing. The four monitoring points, A1/A3, A2/A4, are located in the left rear/front and right rear/front of the upper tank surface, respectively, as shown in Figure 6.
Location of four monitoring points: (a) A1 and A3 and (b) A2 and A4.
Test results and analysis
The waterfall plots of the vibration acceleration magnitude at the four monitoring points of the tank with No. 3 and No. 1 anti-wave boards are shown in Figures 7–10. The horizontal and longitudinal coordinates of the plot represent frequency and time, respectively. The frequency ranges from 0 to 1000 Hz. Each creeping condition takes 4 s, and the cycle is three times for a total of 12 s. Different colors in the figures represent the magnitude of vibration acceleration, and the color is set to be red once the magnitude of vibration acceleration is more than 0.05 m/s2 for convenient comparison. The following results can be seen from the figures:
During the test, for three cycles, the fuel tanks in which the anti-wave board 1 is installed are in decelerated processes at 0–1, 4–5, and 8–9 s; in stationary state at 1–3, 5–7, and 9–11 s; and in acceleration processes at 3–4, 7–8, 11–12 s; the fuel tanks installed with the anti-wave board 3 are in decelerated processes at 3–4, 8–9, and 13–14 s; in stationary state at 4–6, 9–11, 14–16 s; and in acceleration processes at 6–7, 11–12, and 16–17 s.
Three acceleration peaks appear at all monitoring points within the test cycle, which is caused by the impact of the fuel sloshing on the fuel tank wall when the vehicle was decelerating.
The vibration intensity at each monitoring point is different in X, Y, and Z directions. The vibration intensity at the point A3 is almost uniform in X, Y, and Z directions; the vibration intensities at the points A1 and A4 are stronger in Z and X directions, respectively, and weak in the other direction; and vibration intensity at the point A2 is weak in all three directions. Therefore, the vibration intensity at the point A3 is the strongest, followed by the point A4, and the point A2 is the weakest among the four measuring points.
Meanwhile, it is also visible that the vibration acceleration magnitude at the monitoring points of the fuel tank with the No. 3 anti-wave board is much smaller than those of the tank with the No. 1 anti-wave board. Especially, at A3, measurement point where vibration is the largest, the fuel tank installed with the No. 3 anti-wave board effectively reduces the acceleration in the frequency band of 400–600 Hz at the X direction, and the acceleration in the frequency band of 600–700 Hz at the Y and Z directions. Therefore, it can be inferred that the No. 3 anti-wave board inhibits fuel from sloshing greatly.
Vibration acceleration at the point A3: (a) X direction, (b) Y direction, and (c) Z direction.
Vibration acceleration at the point A1: (a) X direction, (b) Y direction, and (c) Z direction.
Vibration acceleration at the point A4: (a) X direction, (b) Y direction, and (c) Z direction.
Vibration acceleration at the point A2: (a) X direction, (b) Y direction, and (c) Z direction.
Conclusion
Subjective evaluation on fuel sloshing noise of the fuel tank is carried out through the car road test and the results show that the fuel sloshing noise is strongest when the fuel tank is filled with fuel of 100% of the rated volume and the car is running under creeping driving condition.
In order to reduce the fuel sloshing noise under creeping driving condition, two anti-wave boards with different structural characteristics are designed, and then the fuel sloshing in the three different anti-wave boards including the original anti-wave board is simulated numerically. The results show that No. 3 anti-wave board (one of the newly designed) has the best effect of inhibiting fuel from sloshing compared with No. 2 anti-wave board and original anti-wave boards. Because its bleed hole has the smallest diameter and the largest quantity, it has the strongest effect on the change of the fuel flow velocity and the largest energy dissipation effect, reducing the impact of fuel sloshing on the fuel tank wall. And it indicates that the main effect of the anti-sway effect is the diameter and number of the bleed holes, and the length of the anti-wave board is hardly affected.
The car road test is carried out separately for the fuel tanks with No. 3 anti-wave board and the original tank. The results show that three acceleration peaks appear at all monitoring points within the test cycle, which is caused by the impact of the fuel sloshing on the fuel tank wall when the vehicle is at deceleration condition; the vibration acceleration magnitude at the monitoring points of the fuel tank with the No. 3 anti-wave board is much smaller than those of the tank with the No. 1 anti-wave board, especially in the 400–1000 Hz band of the left front and right front measurement points where the vibration is large. It can be inferred that the No. 3 anti-wave board inhibits fuel from sloshing greatly.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
