Font Size: a A A

Experimental Study And Numerical Simulation Of Particle Agglomeration Characteristics In Fast Fluidized Bed

Posted on:2020-11-25Degree:MasterType:Thesis
Country:ChinaCandidate:Y H ChenFull Text:PDF
GTID:2431330578975004Subject:Thermal Engineering
Abstract/Summary:PDF Full Text Request
Fast fluidized bed(FFB)has been widely used in energy,chemical,metallurgical and other fields due to its advantages of good gas-solid contact,efficient heat and mass transfer and chemical reaction.Particle gathering to form clusters is a typical characteristic of gas-solid flow in fast fluidized beds leading to the uneven flow.At present,the understanding of particle clustering behavior and its characteristics is not enough.The mechanism of formation and development of clusters is still unclear.That makes the process of gas-solid flow and chemical reaction represented by the fast fluidized bed lack the necessary theoretical support.Therefore,in this paper,the high speed camera and optical fiber probe measurement method were used to carry out the experiment on the constructed fast fluidized bed.The cluster characteristics of Geldart A and B particles in the riser were studied respectively.At the same time,the particle clustering behavior in two-dimensional and three-dimensional fast fluidized bed was numerically simulated respectively based on MP-PIC method.And the simulation results were compared with the experimental results.The main contents and conclusions are as follows:The visualization system of gas-solid flow in quasi-two-dimensional fast fluidized bed was constructed.The clustering behavior of glass beads and FCC particles in the riser was studied within a wide operating range(Ug=3-7.5 m/s,Gs=5-392 kg/m2·s).The phenomenon of particle clustering in the riser was recorded by high-speed camera.Based on image processing technology,K-means clustering algorithm was used to realize the automatic identification of clusters.The results indicate that the algorithm has a good identification efficacy on the clusters and their internal structures in complex gas-solid two-phase flow.At the same time,based on image processing,it was found there is an obvious concentration gradient inside the clusters.In other words,a typical cluster consists of a structurally stable and dense cluster core and a loosely structured cluster cloud wrapped around it.The cluster characteristics such as cluster concentration,size(horizontal/axial dimension),structure(ratio of cluster width to length)and area fraction were obtained by using self-programmed procedure to analyze the image signal based on MATLAB.The effects of operating conditions,riser space position and material properties on cluster characteristics was studied.Results indicate the cluster core concentration is the highest and the cluster concentration is significantly higher than the mean solids concentration.Cluster and cluster core concentration increases with the solid circulation rate increases or the superficial gas velocity decreases,which is the same as the trend of the mean solids concentration.The relationship between the cluster concentration and the mean solids concentration was established which was almost unaffected by material properties.The mean ratio of cluster width to length in the riser is always less than 1,indicating that the long strip-shaped clusters in the riser are the majority with the length greater than the width.Moreover,with the decrease of superficial gas velocity and the increase of solid circulation rate,the irregularity of cluster shapes intensifies.The cluster size varies from a few millimeters to hundreds of millimeters.And particle mass contained in small-sized clusters accounts for more than 10%of the total solid mass.Therefore,the impact of small-sized clusters on gas-solid flow can not be ignored.The cluster area fraction decreases with the superficial gas velocity increasing and solid circulation rate decreasing.The cluster area fraction is zero under low solid circulation rate,indicating that the cluster core disappears at this time and the cluster only consists of loose cluster cloud.The local particle concentration pulsation signals at different horizontal positions of the riser at three heights were obtained by the transmission-type optical fiber measurement method.The clusters were identified from the pulsation signals based on the threshold method and the cluster concentration,frequency,existence time and other statistic characteristics were obtained.The effects of operating conditions and the space position in the riser on the cluster statistic characteristics were studied.And the results of optical fiber measurement were compared with that of high-speed camera.The results indicate that,the trend of mean solids concentration obtained by the two measurement methods is quantitatively consistent.When the solid concentration in the bed is lower,the result of optical fiber measurement is very close to that of high-speed camera while the result of optical fiber measurement is higher than that of high-speed camera under higher solid concentration.From the center region to the wall region along the horizontal direction,the cluster frequency gradually decreases while the mean cluster existence time increases.That is because the majority in the wall region is mostly clusters with high concentration and large size due to the wall effect.The gas velocity is high and the solid concentration is low in the center region,making large-sized clusters difficult to form.Based on the MP-PIC method,two-dimensional and three-dimensional fast fluidized bed mathematical models were established respectively.The EMMS drag model was used to simulate the flow of FCC particles in the riser.It is found that the gas-solid flow in the riser shows a typical ring-nuclear flow structure.The particles gather together in the riser to form clusters,indicating that the EMMS drag model can better simulate the uneven flow structure in the fast fluidized bed.At the same time,the numerical simulation results were compared with the experimental results,which indicating that the mean solid concentration in the two-dimensional riser is significantly higher than that in the three-dimensional riser due to the wall effect under the same operating condition.As the height of the riser increases,the mean cluster existence time gradually decreases.The simulation results of mean cluster existence time are lower than the experimental results,that is because the simulated inlet conditions are different from that in the experiment.On the contrary,the influence of the inlet and outlet structures is small in the middle of the riser,leading to the simulated results are close to the experimental results.
Keywords/Search Tags:fast fluidized bed, particle clustering, high-speed camera, transmission-type optical fiber probe measurement, numerical simulation
PDF Full Text Request
Related items