Study of the Thermal Insulation and Flow Field of Vehicle Front Exhaust Pipe

Abstract

Exhaust is generated by engine flows through a turbocharger, front exhaust pipe, and selective catalytic reduction (SCR) post-treatment device. The structure of the front exhaust pipe affects the temperature, velocity, and turbulent kinetic energy of exhaust and the Sauter Mean Diameter (SMD) of urea water solution (UWS). A high temperature and turbulent kinetic energy in the exhaust will promote the decomposition of UWS, and further accelerate the evaporation and atomization effect of the UWS droplets. Therefore, in order for the exhaust to reach a high temperature and turbulent kinetic energy, a double-layered pipe structure with air insulation was designed. The flow field and the atomization of UWS in the double-layered pipe based on hydroforming processing was investigated through numerical simulation. The thermal insulation simulation was verified by the temperature measurement system and the temperature drops between the double-layered pipe and the volcanic rock-wrapped pipe were also compared. The results indicate that the temperature at the outlet of the designed double-layered pipe was 3.5% higher than that of a single-layered pipe with the same structure, and the velocity at the outlet of the exhaust of the double-layered pipe was 16.1% higher than that of a single-layered pipe. The maximum turbulent kinetic energy in the double-layered pipe was 71 times that of the single-layered pipe. The design is not only conducive to the mixing of UWS and exhaust, but can also improve the atomization performance of UWS.

1. Introduction

The vehicle front exhaust pipe connects the turbocharger and selective catalytic reduction (SCR) post-processing device, and is used to discharge the exhaust generated by the engine and ensure the full combustion of fuel in the engine, so as to ensure the maximum output power of the vehicle [1]. This requires the vehicle front exhaust pipe to have good thermal insulation. In order to improve the thermal insulation of the exhaust pipe, Lu [2] designed an exhaust pipe with an interlayer structure with a thermal insulation effect of 30–50%. However, the long-term thermal insulation effect was poor because the vacuum environment between the interlayer structure cannot be maintained for a long time. Wang [3] wrapped the thermal insulation material around the pipe, but the wrapped thermal insulation material had poor wear resistance and was easy to peel off, and the outer wrapping layer was not easy to weld. Therefore, the development direction of the long-term application of the front exhaust pipe is not only to have good thermal insulation structure, but also to ensure that it is easy to process and assemble. Additionally, the structure of the exhaust pipe affects the temperature and velocity distribution of the exhaust [4,5,6].

The temperature of the exhaust depends on the thermal insulation effect of the front exhaust pipe, which is correlated with combustion quality, and pollutant emissions [7]. It plays an important role in the SCR and decomposition of UWS [8,9,10]. In addition, the decomposition of UWS is also related to turbulent kinetic energy. In the flow field, an increase in turbulent kinetic energy enhances the mixing of gas and makes the mixture more homogeneous [11]. The UWS is separated into a lot of droplets with many sizes under turbulent kinetic energy. The Sauter Mean Diameter (SMD) is the equivalent diameter of all UWS droplets. UWS droplets having a small SMD is conducive to droplet evaporation and the preparation of ammonia [12,13], prevents the formation of urea deposits [14], improves the conversion efficiency of nitrogen oxides [15], promotes heat and mass transfer between the droplets and exhaust [16,17], and allows droplet evaporation and atomization [18,19]. Therefore, developing a front exhaust pipe with a good thermal insulation effect and that is easy to process and look after in the long term is the direction of thermal insulation design.

For the processing of a front exhaust pipe, we adopt the hydroforming method. Compared with the parts formed by stamping welding, the weight of the pipe formed by hydroforming could be reduced by 20–40% [20], thea cost of exhaust pipes is also reduced [21]. In addition, an exhaust pipe with a more complex shape is possible by hydroforming [22]. In order to ensure a good thermal insulation and flow field performance, so as to ensure the decomposition of UWS, a double-layered exhaust pipe with air insulation based on hydroforming processing is designed in this article.

2. Materials and Method

2.1. The Structure of the Exhaust Pipe

In the vehicle layout, the front exhaust pipe is located behind the turbocharger and exhaust bellow, and before the post-processing device. After the exhaust flows out of the turbocharger, it passes through the exhaust bellow and the front exhaust pipe and is fully mixed with UWS at the nozzle before entering the post-processing device. The installation sequence of the turbocharger, exhaust bellow, front exhaust pipe, nozzle, and exhaust tail pipe is shown in Figure 1.

In order to ensure the temperature of the interior chemical reaction in the post-processing device and reduce the flow resistance of the exhaust, a double-layered exhaust pipe structure with air insulation was designed, as shown in Figure 2. The diameter of the front exhaust pipe depends on the outlet diameter of the turbocharger and the inlet diameter of the post-processing device. The shape of the front exhaust pipe depends on the size of the whole vehicle and installation position on the frame. The shape of the front exhaust pipe was achieved by using a pipe-bending machine. The interior side was compressed and the exterior side was stretched when the pipe is bent. The exterior side of the pipe became thinner and the strength of the exterior side decreased. Reinforcing ribs were arranged on the surface of the exhaust pipe in order to increase the strength of the exterior side. The plane expansion of the six curves constituting the reinforcing ribs is shown in Figure 2c. Among them, curve 1 and curve 2 are wavy, curve 3 and curve 4 are sinusoidal function curves, and curve 5 and curve 6 are U-shaped.

At both ends of the double-layered pipe and the adjacent area of the nozzle, the interior and exterior sleeves are bonded together. The other positions are separated from each other, forming the air gap area of the pipe body. The double-layered exhaust pipe with air insulation was manufactured by hydroforming. Using the Pascal principle, the pipe was closed and filled with fluid, which was pressurized until the fluid pressure forced the pipe to be formed. A high shape accuracy can be achieved by controlling the fluid pressure [23]. The blanks of the double-layered pipe were machined by a pipe-bending machine. The wall thickness of the inner and outer pipes of the exhaust double-layered pipe is 2.3 mm and the wall thickness of the outer pipe is 0.8 mm, as shown in Figure 2b. The external pipe was under pressure from the fluid and was gradually fitted with the mold. After this, the area of the air gap and reinforcing ribs were constructed. The structure of a single-layered pipe bent by a pipe-bending machine is shown in Figure 2d. The wall thickness of the single-layered pipe is 2 mm. By comparing the temperature distribution, velocity distribution, turbulent kinetic energy distribution, and droplet diameter distribution of UWS between the double-layered pipe and single-layered pipe, the thermal insulation performance, exhaust flow resistance, and atomization of UWS of the designed double-layered pipe were analyzed.

2.2. Temperature Measurement System

As shown in Figure 3, the temperature measurement system was composed of a burner, fan, mixer, frequency converter, control cabinet, computer, intake pipe and test exhaust pipe, as well as sensors. The burner was used to generate high-temperature exhaust and adjust exhaust temperature. The fan was used to provide flow and improve combustion. The mixer was used to ensure the successful ignition of the burner and the uniform mixing of high-temperature exhaust and air, so that the pressure of the temperature measurement system was close to the actual pressure of the engine. The frequency converter was used to adjust the frequency of the fan and the flow rate of the outlet. The control cabinet could protect the main electrical components and control the operation of the temperature measurement system. The computer realized the programming and display of the system parameters. Sensors were used to collect temperature data. The system used the hot gas generated by the combustion of the burner to simulate the exhaust of the diesel engine and adjusts the temperature and flow of the exhaust from the combustion chamber by changing the air and fuel flow in the combustion chamber.

2.3. Computational Fluid Dynamics

The flow field of the front exhaust pipe was analyzed through computational fluid dynamics (CFD). Using ANSYS Fluent (Academic Research Mechanical, Release 19.1, ANSYS, Inc., Pittsburgh, PA, USA), models of the exhaust pipe and the environment were established. In order to simplify the mathematical model, we made the following assumptions [24]:

(1)

Exhaust is regarded as an ideal gas.

(2)

There is no chemical reaction when the exhaust flows into the front exhaust pipe.

(3)

The flows in the channels are turbulence, and the Reynolds number is 57,250. In order to ensure the stability and accuracy of turbulence calculation, we chose the standard k − ε method.

(4)

The thermodynamics of the gases and the components materials are constants.

(5)

The effect of thermal resistance is negligible.

The specific simulation boundary conditions are shown in

Table 1. Boundary conditions.

Parameters	Value
Mass flow of exhaust Qe (kg/s)	0.156
Temperature of exhaust at the entrance Te (°C)	360
Density of exhaust ρ (kg/m3)	0.53
Mass flow of UWS QUWS (kg/s)	4.56 × 10−4
Temperature of UWS Tu (°C)	27
Temperature of the environment T (°C)	30
Wind speed of the environment V (m/s)	22.2

.

Based on the assumptions, the calculating process of Reynolds number is as follows:

We assume the exhaust includes carbon dioxide, water vapor, nitrogen and oxygen, when the fuel burns completely. The mass ratio, volume, density and dynamic viscosity of each component of the exhaust are shown in

Table 2. The mass ratio, volume, density and dynamic viscosity.

Component of the Exhaust Gases	Mass Ratio	Volume/m3	Density (360 °C)/ kg·m−3	Dynamic Viscosity (360 °C)/× 10−5 Pa·s
carbon dioxide	0.14	0.0258	0.8472	2.91181
water vapor	0.11	0.0494	0.3473	2.27979
nitrogen	0.73	0.2113	0.5390	3.06817
oxygen	0.02	0.0051	0.6157	3.60596

.

The density of exhaust $\rho$ is 0.5350 kg·m⁻³. Therefore, the volume flow of exhaust Q_V is as follows:

$$QV=\frac{Q_e}{\rho }={0.2916 \mathrm{m}}^3/\mathrm{s}$$

(1)

The inner diameter d of exhaust pipe is 0.116 m. According to the continuity equation, the velocity of inlet v is as follows:

$$v=\frac{4QV}{\pi d^2}=27.5919 \mathrm{m}/\mathrm{s}$$

(2)

According to the empirical formula, the calculation formula of dynamic viscosity of gas mixture is as follows:

$$\mu =\frac{\sum y_i{\mu }_i{M}_i^{0.5}}{\sum y_i{M}_i^{0.5}}$$

(3)

$\mu$ is the dynamic viscosity of gas mixture, $y_i$ is the molar percent of gas (i), ${\mu }_i$ is the dynamic viscosity of gas (i), and Mi is the molar mass of gas (i). According to Dalton’s partial-pressure laws, the molar percent of gas is equal to the volume ratio of gas. So, the dynamic viscosity of four gases is 2.9910 × 10⁻⁵ Pa·s. The Reynolds number is as follows:

$$\mathrm{Re}=\frac{\rho vd}{\mu }=57250$$

(4)

The mathematical model consists of a series of partial differential equations. The governing equations of the model include a continuity equation, momentum equations, an energy equation, a species equation and turbulence equations, which are solved by the simple scheme. The equations are as follows:

(1)

Continuity equation

$$\frac{\partial \rho }{\partial t}+\nabla \left({\rho u}_i\right)=0$$

(5)

The continuity equation is an expression of the application of the law of conservation of mass in hydrodynamics.

(2)

Momentum equations

$$\rho \frac{\partial u_i}{\partial t}{ + \rho (u}_i ·\nabla u_i)=\nabla \left\{-P+\left({\mu +\mu }_t\right)\left[\left(\nabla u_i+\nabla u_i^T\right)-\frac{2}{3}\nabla u_{i}I\right]\right\}$$

(6)

The momentum equation is an expression of Newton’s second law of hydrodynamics.

(3)

Energy equation

$$\frac{\partial {(\rho u}_{i}T)}{\partial x_i}=\frac{\partial }{\partial x_j}\left[\left(\Gamma +{\Gamma }_t\right)\frac{\partial T}{\partial x_j}\right]$$

(7)

This is the energy conservation equation.

Based on the assumptions, the geometric model was preprocessed through ANSYS SCDM to ensure cell zones. In the cell zones of the single-layered pipe, defective geometries are repaired and nozzle is ignored. In the cell zones of the double-layered pipe, the exhaust in the exhaust pipe is separated into four parts (as shown in Figure 4), besides repairing defective geometries and ignoring the nozzle, so that a high-quality grid could be obtained. Due to the fact that only the wall of the nozzle would affect the exhaust flow field, the nozzle was ignored [25]. The exhaust that flows through the bellow and front exhaust pipe was analyzed. The model of cell zones was extended in the direction of the characteristic length in order to reduce reverse flow.

After establishing the calculation area model, Ansys Meshing was used for mesh generation, and the tetrahedral mesh was used. The section and enlarged area of the two models are shown Figure 5. The geometric dimensions of the exhaust bellow, front exhaust pipe and nozzle were small, and the grid was densified. The number of grids of the single-layered pipe was 14,201,339, and that of the double-layered pipe was 18,998,615.

The calculation model of the numerical simulation defines the inlet urea injection mass flow, inlet temperature of exhaust, mass flow of the UWS, temperature of the UWS, temperature of the environment, and wind speed of the environment. The pressure of the outlet was standard atmospheric pressure. A transient model was applied when setting the parameters of UWS. The time step was 0.001s, the maximum number of iterations in the time step was 20, the number of time steps was 1000, and the duration was 1s. There were four holes in the nozzle. The residuals were set by default (the energy was 10⁻⁶ and others were 10⁻³). In the calculation, water was used instead of UWS to simulate the formation and distribution of spray. Nishad found that water and UWS had similar kinetic spray trends [26], thus simplifying the calculation. In addition, considering the heat exchange between the external environment and the exhaust pipe, the model of the exhaust pipe included three heat transfer modes: (1) the exhaust flowing at high speed and high temperature in the exhaust pipe belongs to forced convection heat transfer; (2) natural convection heat transfer occurs between the exhaust pipe and the environment; (3) the heat transfer between solids in the exhaust pipe belongs to heat conduction [27].

3. Results and Discussion

3.1. Temperature Measurement System

The exhaust temperature of the temperature measurement system was set at 360 °C and the flow was set at 560 Kg/h. The temperature sensors at the inlet and outlet of the double-layered pipe were measured 15 times, and the temperature difference between the inlet and outlet of the double-layered pipe was obtained by subtracting the temperature, as shown in Figure 6. The average temperature drop of the double-layered pipe was 33.4 °C. The average temperature drop of the volcanic rock-wrapped pipe was 53.4 °C. The results show that the thermal insulation of the double-layered pipe was better than that of volcanic rock-wrapped single-layered pipe.

3.2. Computational Fluid Dynamics

The temperature difference between the inlet and outlet is an important index to determine the thermal insulation performance of the front exhaust pipe. According to the numerical simulation calculation, the temperature difference between the inlet and outlet of the single-layered pipe was 76.07 °C, and the temperature difference of the double-layered pipe was 73.38 °C. Before injecting the UWS, the temperature difference of the double-layered pipe was 28.7 °C, determined by numerical simulation. The experiment result is 33.4 °C. These are a little different, which verifies the effectiveness of our work. This indicates that the thermal insulation performance of the double-layered pipe was improved by 3.5%. By comparing the temperature fields of the double-layered pipe and the single-layered pipe, as shown in Figure 7a,b, it can be seen that the temperature distributions in both are almost the same in the exhaust bellow part. In the exhaust bellow, the temperature near the corrugation is lower than that of in the middle. Near the nozzle, the UWS flows into the injection pipe, which reduces the temperature around the injection pipe, resulting in a significant decrease in the temperature of the double-layered and the single-layered pipes in this area. The temperature is relatively low at the bottom of the single-layered pipe, mainly because the UWS in the single-layered pipe cannot be fully reacted and moves to the bottom of the pipe body under the influence of gravity, resulting in a relatively low temperature at the bottom of the pipe. In general, the temperature distribution of the double-layered pipe is more uniform than the single-layered pipe. The thermal insulation performance was improved, so as to ensure the full reaction of UWS.

The thermal insulation of the front exhaust pipe will affect the chemical reaction in the post-processing device, and the flow velocity in the front exhaust pipe reflects the resistance and energy loss of the exhaust. After passing through the nozzle, the velocity distribution begins to be asymmetrical. This result is same as the study of Qiu [28], which verifies the effectiveness of our work. According to the study of Qiu, the asymmetrical velocity distribution might be because the UWS broke into droplets. The droplets are subjected to random motion in the coupling space by the effect of the swirling air. This random motion breaks the symmetry of the velocity distribution. By analyzing the velocity field distribution of the single-layered pipe and the double-layered pipe, it can be seen that the velocity difference between the inlet and outlet of the double-layered pipe was 3.7 m/s, and the velocity difference between the inlet and outlet of the single-layered pipe was 4.41 m/s. The velocity difference between the inlet and outlet of the double-layered pipe was 16.1% lower than that of the single-layered pipe. This indicates that the flow resistance in the double-layered pipe was further reduced. As shown in Figure 8, in the exhaust bellow part, the velocity distribution of the two was almost the same. In the exhaust bellow, the velocity near the corrugation was lower than that of the middle. The velocity changes greatly near the nozzle. The velocity distribution in the double-layered pipe was more uniform than that of single-layered pipe, which improved the velocity distribution of the front exhaust pipe. The flow resistance and energy loss of exhaust was reduced in the double-layered pipe.

The turbulent kinetic energy is an important indicator in evaluating the turbulent mixing. An increase in turbulent kinetic energy can promote the mixing of the UWS and exhaust, and then accelerate the decomposition of urea. The turbulent kinetic energy distribution of exhaust is shown in Figure 9, along the central axis of the exhaust bellow and front exhaust pipe. The turbulent kinetic energy significantly changed at the position of the urea nozzle in both the double-layered pipe and the single-layered pipe. Before reaching the urea nozzle, the turbulent kinetic energy of the fluid in the double-layered pipe is higher than that in the single-layered pipe. After the fluid flows through the urea nozzle, the turbulent kinetic energy distributions in the double-layered pipe and the single-layered pipe are closer. In general, the turbulent kinetic energy in the double-layered pipe is higher than that in the single-layered pipe. The reason for the surge of turbulent kinetic energy near the urea nozzle is that the velocity of the exhaust at the inlet of the mainstream is 40 times that of UWS. So, the UWS obtains kinetic energy from the exhaust, and this causes a drastic perturbation at the gas and liquid interface when the UWS is injected into the exhaust pipe [29]. The maximum turbulent kinetic energy was 67,715 m²/s² in the double-layered pipe, while it was 957 m²/s² in the single-layered pipe.

The atomization effect is an important index in evaluating the decomposition of UWS [30]. Based on turbulent kinetic energy distribution, the atomization of UWS was studied in the front exhaust pipe. The cumulative volume percentage is shown in Figure 10, which was obtained by the Rosin–Rammer function. d_0.5 is the corresponding diameter of UWS droplets when the cumulative volume percentage is 50%. d_0.9 is the corresponding diameter of UWS droplets when the cumulative volume percentage is 90%. By comparing the cumulative volume percentages of UWS droplets in the double-layered pipe and the single-layered pipe, it was found that when the diameter of the UWS is less than 85.6 μm, the volume fraction of small droplets in the double-layered pipe is higher. When the diameter of UWS droplets is greater than 85.6 μm, the droplet volume fraction in the single-layered pipe is higher. In order to further compare the distribution of UWS droplets in the two kinds of front exhaust pipes, the SMD of UWS was calculated. SMD is defined as the equivalent diameter of the droplet group with the total surface area and the total volume of the droplet group when the ideal uniform spray field is equal to the actual uniform spray field. This reflects the ratio of the volume of all droplets to the expected surface area. The smaller the SMD of UWS spray, the larger the total surface area of the droplets and the exhaust, and the better the atomization effect [31]. The formula of SMD is as follows:

$$D_{32}=\frac{{{\displaystyle \int }}_{D_{min}}^{D_{max}}{D}^{3}dN}{{{\displaystyle \int }}_{D_{min}}^{D_{max}}{D}^{2}dN}$$

(8)

D₃₂ is SMD, D_min is the minimum diameter of droplet, D_max is the maximum diameter of a droplet, and N is the number of droplets with diameter D. The SMD of droplets was 232 μm in the single-layered pipe. The SMD of droplets was 214 μm in the double-layered pipe. It was easier to obtain spray in the double-layered pipe. According to the study of Kun [32], the SMD of droplets is 83–86 μm. Therefore, the SMD of droplets still needs to be verified by experiments in this double-layered exhaust pipe with an air insulation design.

Figure 10. Cumulative volume percentage of UWS droplets.

Figure 10. Cumulative volume percentage of UWS droplets.

4. Conclusions

(1)

A double-layered exhaust pipe with air insulation based on hydroforming processing was designed. The thermal insulation performance was enhanced through the utilization air as the insulation. Reinforcing ribs were added to the front exhaust pipe in order to increase the mechanical strength.

(2)

The temperature difference between the inlet and outlet of the double-layered pipe was 2.69 °C lower than that of the single-layered pipe. The thermal insulation performance was improved by 3.5%. The temperature was increased in the SCR system, which is conducive to a full reaction and reduces the emission of nitrogen oxide.

(3)

The velocity difference between the inlet and outlet of the double-layered pipe with air insulation was 0.71 m/s lower than that of the single-layered pipe. The velocity distribution was improved in the double-layered pipe. The flow resistance of the exhaust was reduced in double-layered pipe.

(4)

There was greater turbulent kinetic energy near the nozzle in the double-layered pipe than the single-layered pipe, which promoted the mixing of the UWS and exhaust. Furthermore, it increased the distribution range of the UWS in the front exhaust pipe. UWS droplets in the double-layered pipe had a smaller SMD of 214 μm. This is more conducive to the atomization of UWS.