December 30, 2012
包谷是什么意思China Petroleum Processing and Petrochemical Technology
2012,V ol. 14, No. 4, pp 64-72
Simulation and Optimization
Corresponding Author: Dr. Li Chunyi, Telephone: +86-532-86981264; E-mail: chyli@upc.edu.
1 Introduction
Circulating fluidized beds (CFB) are widely ud in indus-trial process such as coal combustion and fluid catalytic cracking (FCC) process, as well as in other areas of the chemical, petrochemical, metallurgical, environmental, and energy industries [1-3]. However, particles form ag-gregates (clusters and particle strands) in industrial CFB rirs as a result of flow instability, energy minimization, and particle-wake interactions [4]. As the solids flux (G s , in kg/m 2s) increas, particle congregation with significant back-mixing is obrved, which is probably caud by the lift force that is reduced by shear near the wall in the low-velocity boundary layer. This back-mixing occurs due to the core-annulus mode of the particle flow [5-6]. The non-uniform solids distribution is obviously reflected in the gas velocity an
d in the radial gradient characteristic of the solids holdup [1]. This non-uniform solids distribution can affect the flow structure and further influence the reaction rates, the lectivity to desired intermediate products, the mass and heat transfer, the gas-solid contact efficiency, the coking reaction, and the erosion within the rir [2]. Zhu
defined the radial non-uniformity index (RNI) parameter to depict this non-uniform quality and provided a direct comparison between different operating conditions [7]. The current study us RNI as its main tool for analyzing the experimental data that shows the non-uniform features of solids’ radial profile.
Low gas back-mixing is necessary becau the expected product is gas. The catalysts require high gas velocity in the rir to reach the plug flow condition becau the con-tact time between gas and solid must be short [8]. Catalytic gas-pha reactions tend to require higher gas velocity, as well as higher solids flux and concentration, than gas-solid reactions such as fluid catalytic cracking [9]. In the former ca, the G s ranges commonly from 300 kg/m 2·s to 1 200 kg/m 2·s, with corresponding solids holdup values in the range of 0.1 to 0.25[10].
High-density circulating fluidized beds (HDCFB) with high solids flux at high gas velocity have received wide-spread rearch attention. Further work on HDCFB
Radial Non-uniformity Index Rearch on High-density,
High-flux CFB Rir with Stratified Injection
Geng Qiang 1; Wang Lu 1; Li Zhichao 1; Li Chunyi 1; Liu Yibin 1; You Xinghua 1, 2
(1. State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580;
2. Petrochemical Factory of the Yumen Oil-Field Company, PetroChina, Yumen 735200)
书法之美Abstract: A high-density, high-flux circulating fluidized bed (CFB) rir (100 mm in ID and 10.614 m in height) was ap-plied in a wide range of operating conditions (with solid fluxes up to 400 kg/m 2s and superficial gas velocities up to 12 m/s) to examine its radial non-uniformity dynamics. The solids holdup was determined through the u of a fiber-optic probe at 11 axial levels. The results indicated that under all operating conditions, the high superficial gas velocity and low solid flux-es maintained a low radial non-uniformity index (RNI). The high-density/flux CFB rir had veral unique characteristics, so that the peak of the radial solids holdup profile occurred at a position with r /R =0.8. The RNI and solids holdup at the cross-ctional position had a good logarithmic relationship at the low-density condition (with a mean solids holdup of <0.2), and the RNI decread when the m
ean solids holdup exceeded 0.2. Investigation of the dynamics of stratified injec-tion revealed that the feed ratio had an important effect on G s and on solids holdup distribution. A novel “<” shaped axial solids holdup profile was found. G s decread sharply when the up-flow feed ratio exceeded 0.5, and RNI was lowest when the up-flow feed ratio was 1.
Key words: fluidization; high density/flux CFB rir; radial non-uniformity index; stratified injection; feed ratio
systems is needed for better comprehension of their ad-vantages and limitations, which in turn will lead to the development of a more reliable reactor scale-up and more cost-effective units.
Bi and Zhu described operations with G s>200 kg/m2·s and solid holdup>0.10 throughout the entire rir. In a ries of studies on HDCFB rirs, there was an abnce of down-flow stream at the wall, a limited tendency to form clusters, and relatively flat axial density profiles without dilute zones[9,11]. Since the HDCFB concept was brought up, a ries of work had been done by many rearchers including HDCFB system design to achieve high solid holdup[12], effect of different solids feeder and exit rir configuration on flow behavior[13], lateral solids disperation study[14] and flow dynamics in the downer[15]. The stratified injection is an important and novel inge-nious approach designed to ac建筑施工员
hieve different reaction environment for the rir[16]. The Rearch Institute of Petroleum Processing (RIPP), which has developed the latest gas and diel manufacturing technology, divides the rir into four reaction ctions. The stratified injec-tion (with gasoline entering the bottom ction, and fresh feed and heavy oil entering the middle ction of the rir) can improve the liquefied petroleum gas (LPG) and diel yields when cracking gasoline and heavy oil feedstocks, respectively. The stratified injection method is also ud in FCC process to maximize the production of propylene, a vital component for the manufacture of ma-jor petrochemicals and other substances. Cracking heavy feedstock before LCG[17-18], or cracking naphtha before gas oil injection[19], can a lso improve propylene yield. Li invented a specific reactor type and method for propylene production that employs a combination of heavy and light olefin feeds[20]. Stratified injection style was also applied to SFI, MAXOFIN, MIP process[21-23].
The above investigations demonstrate that although previ-ous studies have offered an insight into the application of stratified injection in the FCC process, the gas-solids flow behavior has received limited attention thus far.
2 Experimental
2.1 Apparatus
The schematic drawing of the experimental apparatus is shown in Figure 1. Experiments were conducted in the rir made of plexiglass, 0.1 m in diameter and 10.614 m in height. The solid material ud in experiments was spent FCC catalyst particles with a mean diameter of 80 μm and a particle density of 1 780 kg/m3. The size distri-bution of FCC particles is shown in Table 1. Compresd air was injected into the rir through a pressure stabiliza-tion valve (for maintaining a pressure of 0.17 MPa) and a rotameter. Four compresd air inlets were connected with the distributor symmetrically. Catalysts were transferred to the rir from the down-comer through a 30o angled pipe for mixing with the air including the pre-lift gas, the pre-fluidized gas, the lower feed gas and the upper feed gas. Catalysts lifted up to the top of the rir were routed to the cyclones, 500 mm in diameter, in which most of the catalysts were removed from the gas pha and returned to the hopper, 480 mm in diameter, with an inventory of 500—600 kg of catalysts. In order to simulate gas-solid flow of the stratified injection, the upper and lower feed
units were t up at a height of 0.8 m and 3.6 m, respec-Table 1㊀Size distribution of FCC particles
determined by screening
Mesh size, μm Mass faction, %
数学数列>1500.62
120—1507.64
96—12024.51
80—9619.46
75—8034.31
48—7513.47
<480.82
Geng Qiang, et al. Radial Non-uniformity Index Rearch on High-density, High-flux CFB Rir with Stratified Injection
China Petroleum Processing and Petrochemical Technology tively, above the bottom of the rir. This design could effectively improve the solids circulation flux by provid-ing twice the pressure head to achieve the high density/flux fluidized regime. The operating conditions of super-ficial gas velocity and the solids circulation rate ranged from 8 to 10 m/s and 200 to 400 kg/m 2·s, respectively.
2.2 Measurement
Solids circulating rate was calculated in the measur-ing system by measuring the difference of height in the
measuring tank for a given time period which could be achieved by stopwatch. The G s value can be expresd by the equation:
G
s 21 (1)in which r m is the measuring tank radius, 0.125 m, and r b is the bulk density in the meas
uring tank, 870.89 kg/m 3. The solids concentration was measured at 11 axial levels (at 0.53 m, 1.51 m, 2.19 m, 3.42 m, 4.32 m, 5.32 m, 6.42 m, 7.32 m, 8.16 m, 9.06 m, and 10.06 m, respectively,) and 11 dimensionless radial positions (with r /R =0.00, 0.16, 0.38, 0.50, 0.59, 0.67, 0.74, 0.81, 0.87, 0.92, and 0.97, respectively). The solids concentration was obtained by the PV-56 D optical fiber probes (manu-factured by the Institute of Process Engineering, Chine Academy of Sciences). The measurement mechanism of the optical fiber probe is shown in Figure 2. The probe
receives the light signal reflected from the particles that is simultaneously emitted by the light source in the probe. The optical signals will then be converted into voltage signals by a photo-multiplier and acquired by the PC-bad data sampling system. It is of great importance that the fiber probe should be calibrated to exactly transform voltage signals to the solids concentrations before being ud. The conversation from voltage signal to solids con-centration signal is shown in Figure 3. More details can be referenced in Li’s paper [24- 25].
Although the standard deviation could reflect the fluctuations of solids radial profile to some extent, it failed to depict the contrast between different ctions in the total rir. As a result a novel parameter RNI was defined by an equation hatched out by Zhu and Manyele [7] to achieve the contrasting function. Standard deviation ra-tio between the real radial distribution and the most non-uniform radi
al distribution of solids concentration under the same cross-ctional average solids concentration was calculated by the following equation:
RNI s ()ε (2)
where εsmf means solids concentration or solids holdup at minimum fluidization conditions. In this circulation fluidized bed, we obtained its value to be equal to 0.57 through experiments.
●—Fibre for light emitting; ○—fibre for light receiving
2012,14(4):64-72
Figure 3㊀Relationship between solids concentration signal
and voltage signal
3 Results and Discussion
3.1 Effect of operating conditions on RNI
Superficial gas velocity and solids flux are the primary factors affecting the radial profile of solids. Our study ap-plied the RNI parameter to reveal the effect of operating conditions on the uniformity of the solids radial profile. RNI changed from 0 to 1 as the concentration of the radial distribution of solids became increasingly non-uniform.The effect of superficial gas velocity and solids holdup on RNI is illustrated in Figure 4.
In general, the solids’ radial concentration profile is more or less non-uniform under all operating conditions. At a constant G s (300 kg/m 2·s), an increasing U g can result in a decreasing RNI as a major trend, showing that higher U g can benefit the uniformity becau it reduces the aver-age solid holdup and weakens the core-annulus structure. With respect to the energy balance, the concentration in the core region was small and the gas-solid flow, which was basically in a disperd state of particles, was more fully developed, so the influence of operating conditions was relatively small. In the wall region, solids holdup was high, and in combination with the wall effect the particles
tended to gather more firmly. A greater energy is required to maintain the particle motion in the wall area [26].
Thus, changes in operating conditions affected the particle motion at the wall more directly, and higher gas velocity decread the wall effect and lowered the concentration of solids, making the radial profiles more uniform. At a constant U g (10 m/s), the value of RNI incread with a higher G s , due to the formation of significant amount of
clusters near the wall [1, 7, 26].
■—8 m/s; ●—10 m/s; ▲—12 m/s
●—200 kg/m 2s; ▲—300 kg/m 2s; ▼—400 kg/m 2s
童年故事梗概Figure 4㊀Influence of operating conditions on RNI
At the axial position, the rir was divided into three parts, namely: the den suspension upflow region, the flow regime transition region, and the fully developed region (as en in Figure 5). Under a high solids flux condition, the den suspension upflow (which was not a net downflow at the wall, with e s >0.2) was obtained in the bottom region [12], and the RNI was relatively small compared with the transition region, which had higher back-mixing that resulted in non-uniformity situations. The RNI decread as the height of the transition region incread becau the solids concentration remained con-stant in the core region and incread in the wall region. Figure 7 illustrates this regulation when the height is 4.32 m. Increasing the superficial gas velocity lengthens the fully developed region [27], so the RNI remains at a low level within a large range with higher U g . However, near the exit of the rir, RNI suddenly increas mostly becau of exit restraints, resulting in a certain degree of back-mixing.
Geng Qiang, et al. Radial Non-uniformity Index Rearch on High-density, High-flux CFB Rir with Stratified Injection
China Petroleum Processing and Petrochemical Technology
什么地听
Figure 6㊀The radial profiles of solids concentration at
different U g
■—8 m/s; ●—10 m/s; ▲—12 m/s
Figure 6 shows the effect of superficial gas velocity on radial profiles at different heights (i. e.: near the bottom of the rir, near the lower feeder, and near the upper feed-er). We divided the radial ction into two parts. The core region (at r/R =0—0.5), where solids holdup was less, and the wall region (at r/R =0.5—1.0), where solids holdup
was more, constituted the typical core annulus struc-ture, which was significant with the height. The solids holdup in the wall region was more nsitive to changes in operating conditions than that in the core region. An increasing superficial velocity lowered the solids holdup, but it was nearly constant in the wall region. This can also be elucidated in terms of the energy balance, as it has been previously mentioned.
The summit of the radial profiles curve is located near the position at r/R =0.8, but not clo to the position at r/R =1 as reported by other studies on low-density or low-flux con-ditions. In this study, we
found out that when the operat-ing conditions achieved high density or high flux, some gas flow occurred along the wall. The traditional explana-tion of the core-annulus structure is attributed to the wall effect, such as the interaction between particles and the wall, and the particle velocity decreas as the resistance of particle motion increas. The high density or flux of the CFB reactor may break this wall effect, shifting the resistance region from the wall to the position at r/R =0.8. However, this potential explanation requires further in-vestigation.
3.2 Effect of feed ratio on G s and RNI
Figure 7 plots the changes in solids flux for different up-flow feed rates. In general, the solids flux is constant under certain operating conditions for up-flow feed ratio. However, when the up-flow ratio incread to 0.45, the sol-ids flux decread rapidly from 470 kg/m 2·s to 400 kg/m 2·s. To find out the reason, we designed three different feed patterns, i. e.: a single up-flow feed, a single down-flow feed, and a feed with 0.43 up-flow ratio under the same op-erating conditions (with U g =10 m/s, and G s =300 kg/m 2·s). The axial profiles of the solids holdup at the bottom of rir depicted in Figure 8 show the values of the solids holdup in a decreasing order: single up-flow feed >strati-fied feed>single down-flow feed. In the ca of the single up-flow ratio, the pre-lift and pre-fluidized gas has little ability to carry solids at the bottom of the rir due to de-creasing velocity and dynamic hea
程文彬
d, resulting in a lower drag force, back-mixing of particles, and higher solids holdup. Thus, the regenerated catalysts flowing into the rir are faced with a greatest resistance.
Furthermore, the axial profile of the solids holdup in the
■—8 m/s; ●—10 m/s; ▲—12 m/s
■—200 kg/m 2s; ●—300 kg/m 2s; ▲—400 kg/m 2s红色精灵
Figure 5㊀Axial profiles of solids holdup under different
operating conditions
2012,14(4):64-72
