Analysis of Dust Reduction Characteristics of Multistage Tandem Dust Removal System

,


Introduction
Afected by material aging, severe weather, and highway loading, the asphalt layer of the asphalt pavement often develops various early damages such as ruts, cracks, and potholes within the design life [1]. Tese damaged pavements reduce vehicle safety and driving comfort. Te asphalt thermal regeneration pavement maintenance vehicle can heat the waste asphalt from a damaged pavement in place to swiftly repair it and extend its service life, while reducing maintenance costs and waste of raw materials [2]. During the heating and melting process, however, waste asphalt will produce a signifcant amount of organic waste gas, as well as a signifcant amount of dust. Te organic waste gas has a complex composition, mainly consisting of nitrogen oxides, sulfur oxides, polycyclic aromatic hydrocarbons, and other substances, and also contains strong carcinogens such as benzopyrene, carbazole, and anthracene derivatives, which pose a serious threat to the environment and the health of residents [3]. Terefore, special dust removal and exhaust gas purifcation systems must be designed for it.
Tere are few studies focused on vehicle-mounted systems for removing asphalt exhaust dust. Xu and Liu [4] conducted a study on particle agglomeration in lowtemperature dust collectors and found that the lower the inlet fue gas temperature of the dust collector, the more obvious the particle agglomeration phenomenon is. Alizadehyazdi et al. [5] used an electrostatic dusting device to remove dust particles from the adhesive surface, which is optimized to be more efective than the usual rinsing means. Mo et al. [6] proposed a dust removal technology for a coal cutting machine with the car to solve the problem of dust generated from the surface coal mining. A small dust collector is installed at the end of the coal cutter to prevent dust from spreading to the walkway side causing environmental pollution. Yi et al. [7] provided treating phenyl-containing organic waste gases with a surfactant absorbent solution. Yoshida et al. [8] investigated the rotational velocity distribution inside a dual inlet and single inlet dust collector. Te data analysis showed that the rotational velocity distribution inside the multiple inlet dust collector is more uniform and symmetrical, thus efectively reducing friction and decay of the internal cyclonic fow. Te experimental results also show that the former has a higher separation efciency than the latter. Utilizing the discrete conjugate approach, Elsayed [9] suggested a new fared exhaust pipe while optimizing the least pressure drop of the cyclone. When compared to the standard plain cyclone, the optimized cyclone saves 66 percent of the driving power. Hsiao et al. [10] investigated the cyclone separator's longitudinal and radial dimensional ratios and optimized the structure. It was discovered that raising the barrel's height might lengthen the period of particle rotation separation and boost the efectiveness of dust removal. Sgrott et al. [11] performed simulation optimization of a cyclone dust collector by using CFD software for 5 μm to 15 μm dust simulation. Wang et al. [12] designed a new two-stage on-line gas-liquid cyclone separator. It integrates the advantages of horizontal and vertical inline gas-liquid cyclone separator, and it can meet the separation requirement for both gas and liquid. Te tangential velocity, gas volume fraction, and pressure distribution inside the separator are studied by numerical simulation using computational fuid dynamics. In terms of numerical solution, Gopalakrishnan et al. [13] numerically investigated the particle separation through an axial vortex tube cyclone separator by solving the Reynolds-averaged Navier-Stokes equations using the RNG K-epsilon model as a discrete phase model (DPM) for turbulent closure and particles using the Eulerian-Lagrangian method. Yamasaki et al. [14] performed a multiobjective optimization of an axial cyclone separator to improve the overall performance at diferent velocities. Te separation efciency and pressure drop are taken as the objective functions.
Tis study ofers a structural design for a dust removal system with multistage combined treatment to address the issue that it is challenging to purify asphalt fumes by a single method. It includes a cyclone section, a spraying section, and an adsorption section and achieves efcient purifcation of complex composition exhaust gases by the multistage sectional treatment.

New Multistage Combined Dust Removal
System. Te overall length, width, and height is 960 mm × 878 mm × 1180 mm, and the size of the arrangement space on the car is 1300 mm × 900 mm × 1600 mm. Te structure of the dust removal system is depicted in Figure 1, which includes a cyclone section, a spraying section, and an activated carbon adsorption section.
Due to the special cone-shaped structure of the cyclone section separator, the airfow changes from linear to highspeed rotational motion after the gas enters through the air inlet. Te dust particles are thrown against the cylinder wall by centrifugal force much greater than their own gravity, eventually settling in the dust collection box at the bottom of the cyclone dust collector. Figure 2 shows the air fow diagram of the dust removal system. Te exhaust gas enters the spray section after being pretreated in the cyclone section, which is connected to the spray section by a pipe. Te spraying tank, spraying pipeline, pumping machine, and mist elimination plate are all parts of the spraying portion. To absorb sulfde in the exhaust gas, the alkaline liquid stored in the spray tank is circulated and sprayed in the spray tank via the pump machine and spray pipeline. Te spraying portion is connected to the activated carbon adsorption section by a pipeline, and the activated  carbon adsorption tank is flled with honeycomb activated carbon for adsorbing powerful carcinogenic compounds. Te top of the activated carbon adsorption section has a centrifugal fan for suction.

Geometric Model of Cyclone Separator.
Te cyclone separator is an integral aspect of the asphalt waste gas pretreatment process for separating the gas-solid combination. Understanding the operating characteristics of the cyclone separator is crucial to improve the dust removal efciency of the dust removal system. However, the fow feld inside the cyclone section separator is complicated and variable, and the efects of various structures on separation efciency and pressure drop are frequently interactive, making the experimental analysis difcult [15,16]. As a result, this work uses numerical simulation to investigate the cyclone separator's internal fow feld characteristics as a foundation for enhancing the structure. Figure 3 is a schematic diagram of the structure of the cyclone separator, whose main structure includes inlet height a, inlet width b, barrel height Ho, total height H, exhaust pipe depth L, exhaust pipe diameter Da, barrel diameter D, and dust collection port diameter Db. Te annular space is the cylindrical space of the cyclone section separator, the height of the annular space is Ho, the separation space is the conical space of the cyclone section separator, and the height of the separation space is H-Ho.

Selection of Mathematical Model.
Te fow feld inside the cyclone separator is an extremely complex threedimensional, gas-solid two-phase cyclonic fow with signifcant anisotropy, so the Reynolds Stress Turbulence Model (RSM) is selected for the analysis. Te RSM model strictly considers the efects of fow line bending and vortices and has higher accuracy for complex turbulence simulations [17][18][19]. Te RSM set of basic control equations for turbulence consists of the continuity equation, the N-S equation, the Reynolds stress equation, the turbulent kinetic energy equation, and the dissipation rate equation.
Assuming that the fuid in the cyclone section separator is incompressible and isothermal, the corresponding control equation is as follows.
Continuity equation is given by Navier-Stokes equation is as follows: where μ g is the gas viscosity coefcient, ρ g is the gas density, and τ ij is the Reynolds stress term. Reynolds stress transport equation is as follows: where D ij is the difusive transport term, P ij is the stress production term. ij is the pressure-stress correlation term, and ε ij is the dissipation term. Turbulent energy equation is given by

Shock and Vibration 3
Dissipation rate equation is given by where C ε1 � 1.44 and C ε2 � 1.92. Te arithmetic-saving DPM model [20][21][22] was chosen for the simulation of dust particles instead of treating the particles as one phase alone (VOF multiphase fow model). Te cyclone separator has achieved 99% separation efciency for the particles with diameters above 10 μm, and the dust particles simulated in this paper are 1 μm to 10 μm in diameter.
Using a QUICK diferential format control discretization with second-order accuracy, the SIMPLE algorithm was employed to solve the coupled pressure-velocity equation [23]. Te separation efciency of the cyclone separator is calculated by the following equation:  When the number of grids reaches 425826, continuing to encrypt the grids has little efect on the results. At this time, the cloud map is smooth and continuous, and the results are highly credible. Te calculation is mainly judged by the residual values of the iterations to determine whether the calculation converges or not. In addition, the judgment can be assisted by monitoring the mass fow rates of the inlet and outlet, and the calculation can be regarded as converging when the mass fow rates of the inlet and outlet reach equilibrium. In this paper, the residuals of the continuity equation and momentum equation are less than 10 −4 , which can be regarded as the convergence of the computational model. Figures 5-7 show the cyclone section separator static pressure clouds, tangential velocity clouds, and turbulence intensity clouds at diferent exhaust pipe depths. Te simulation fndings reveal that the static pressure in the cyclone's negative pressure zone increases with exhaust pipe depth and always emerges at the exhaust pipe's entrance. Te negative pressure creates a pumping force that converts the cyclone's downward cyclonic fow into an upward cyclonic fow. Te static pressure on both sides of the cyclone wall increases with the depth of the exhaust pipe, and the total pressure drop also increases.

Te Development Law of Internal Flow Field.
Te closer to the central column core, the lower the static pressure, and as the depth of the exhaust pipe increases, the core's static pressure distribution becomes more uniform and symmetrical. Te tangential velocity cloud shows the same trend. Te tangential velocity of the column core becomes more stable as the depth of the exhaust pipe increases, while velocity fuctuation decreases and the velocity of the updraft slows, making it more difcult for tiny particles to be carried away from the cyclone separator by the updraft, increasing the cyclone separator's dust and gas separation efciency.

Device Improvement and
Comparison. An antiventuri cyclone dust collector with a scrambling column is designed in Figure 8. Te venturi has a narrow central portion and two wide sections, which causes the airfow to change from coarse to thin, accelerates the airfow velocity, and creates a low pressure area at the venturi outlet's outer circle [24]. Te middle of the antiventilator, on the other hand, is wide and the two sides are narrow, which has a retarding impact on the airfow and efectively reduces the rising speed of the airfow in the center of the cyclone dust collector under the same shape and size premise. Te spoiler column at the bottom plays an important role in destroying the reverse cyclone there. Te spoiler column forces the downward airfow to reverse ahead of time, avoiding pulling up fne dust particles that have accumulated at the bottom of the dust collector when it reverses at the bottom. Tis measure improves the efciency of dust removal. Figure 9 shows the static pressure distribution cloud of the original and improved structures. Figure 9(a)     Shock and Vibration Figure 9(b) shows the static pressure distribution of the improved structure. Te simulation results show that compared with the ordinary cyclone with the same exhaust pipe depth, two negative pressure zones are formed inside the exhaust pipe of the antiventuri cyclone separator. Te two negative pressure zones appear in the narrow section of the exhaust pipe, and the static pressure distribution is more symmetrical. Figure 10 shows the tangential velocity distribution cloud of the original and improved structures, with Figure 10(a) being the original structure and Figure 10(b) being the improved structure. Te local vortex at the bottom of the improved device is signifcantly reduced, and the maximum tangential velocity is 4.55 m/s, which is less than 9.74 m/s of the original device, and the updraft in the center of the device is more stable, which is conducive to improving       the separation efciency. Figure 11 shows the axial velocity distribution cloud of the original and improved structures, with Figure 11(a) being the original structure and Figure 11(b) being the improved structure. Te axial velocity distribution can more specifcally refect changes in the upward velocity of the central airfow. In the original device, the reverse airfow from the bottom of the dust collector can easily develop into a local turbulent airfow, and in the improved device, this phenomenon is efectively suppressed. Figure 12 shows the velocity vector diagram of the air core of the improved device and the original device, the air core in the center of the improved device is smoother, while multiple turbulent    cyclones appear inside the original device. Te bottom column has a clear infuence on suppressing turbulent cyclones by causing airfow to move upward earlier, which should be reversed at the bottom. Figure 13 Figure 12(a) and the modifed device in Figure 12(b).

Key Performance Comparisons.
Z � 550 mm height section is located in the middle of the row pipe, and this position is the most representative of the full development of cyclones, stable fow velocity. Compared with the original device, the central cyclone of the improved device has a smaller rise velocity, and the maximum rise velocity of the central airfow of the original device is 17.3 m/s, while the maximum rise velocity of the central airfow of the improved device is 15.83 m/s. Tis is basically consistent with the data in Figure 11, which indicates that the central cyclone is signifcantly weakened in the improved device. Te center cyclone will roll up the dust settled at the bottom during the rising process, causing the secondary entrainment phenomenon, and the smaller rising speed means that the dust at the bottom is more difcult to escape with the updraft.
Te axial velocity of the two devices rises at the same rate at frst, but when the airfow reaches the exhaust pipe, the upgraded device's maximum axial rise speed is substantially slower than the original device. In the exhaust pipe of the modifed device, the airfow decelerates when passing through the fared section and accelerates when passing through the narrow section, and the total average velocity is less than that of the original device. At various inlet speeds, Figures 15 and 16 show the total dust removal efciency and total pressure drop for the modifed and original units.
Te total dust removal efciency of both the improved unit and the original unit increased with the increase of the inlet air speed, and the total dust removal efciency of the improved unit was consistently greater than that of the original unit, which proved the advantage of the improved unit. When the air inlet speed is 6 m/s, the dust removal efciency of the original device is 91.9%, and the dust removal efciency of the improved device is 97.0%, which is 5.1% higher than the original device. When the air inlet speed is 8 m/s, the dust removal efciency of the original device is 93.14%, and the dust removal efciency of the improved device is 97.3%, which is 4.16% higher than the original device, and the diference between them keeps decreasing as the air inlet speed increases.
In addition, the diference between the two dust removal efciencies decreases as the inlet air velocity increases. When the inlet speed is low, under 10 m/s, the total pressure drop of the improved device is slightly larger than that of the original device, but as the inlet speed rises, the total pressure drop of the improved device rises more quickly. For this reason, the inlet speed of the improved device should not be set too high. 2.

Conclusions
(1) Tis paper proposes a structural design method for the dust removal system of an asphalt hot regeneration pavement maintenance vehicle, using the concept of multistage combined treatment to complete the design of cyclone section, spraying section,  and adsorption section. Te cyclone section uses a cyclone separator to remove dust particles from the exhaust gas; the spray section absorbs sulfdes through circular spraying of lye; and the adsorption section adsorbs major carcinogens through honeycomb activated carbon.
(2) Te CFD numerical simulation was used to analyze the structure of the cyclone section of the dust removal system, and the static pressure and tangential velocity inside the cyclone section were basically symmetrically distributed. Te static pressure distribution and tangential velocity feld inside the cyclone become increasingly symmetrical and stable as the depth of the exhaust pipe rises; the dust removal efciency is positively related to the symmetry and stability of the fow feld. (3) A antiventilator cyclone with a scrambling column was designed. Te venturi tube is narrow in the middle and wide at the ends, while the antiventuri tube is wide in the center and narrow at the ends. Tis design reduces the rising cyclone's speed and prevents the dust that has settled at the bottom of the dust collection tube from being raised once more, thus improving the dust removal efciency. Te presence of the scrambling column has a key role in suppressing the turbulent fow at the bottom of the cyclone, and the vortex eccentricity is signifcantly reduced or even eliminated, which makes the fow feld inside the cyclone more stable. (4) Further study on the experimental validation of the device will be carried out, including the construction of the prototype and experimental platform.

Data Availability
Te data used to support the fndings of this study are included within the article.

Conflicts of Interest
Te authors declare that they have no conficts of interest.