Font Size: a A A

Performance Modeling And Optimization Of Fixed-bed Gas-liquid Distributor

Posted on:2022-06-21Degree:MasterType:Thesis
Country:ChinaCandidate:L Y LiFull Text:PDF
GTID:2481306542985009Subject:Power Engineering and Engineering Thermophysics
Abstract/Summary:PDF Full Text Request
In recent years,the quality of crude oil in the world is declining year by year,and the trend of heavy and inferior quality is becoming more and more obvious.Hydrogenation technology is an important means to solve this problem.Although researchers have achieved great success in the research of hydrogenation catalysts.However,the reaction efficiency is also restricted by the development of the hydrogenation reactor and its internal components.At present,the most widely used hydrogenation process is the fixed-bed reactor,and the optimization design of its internal components has always been a research hotspot for researchers.Since the efficiency of the fixed bed reactor depends on the good distribution of the material on the catalyst bed,therefore,the gas-liquid distributor is one of its most important internal components.This paper uses the CFD simulation data of the Venturi-type gas-liquid distributor collected in the literature as the data support,modeling the performance of the Venturi-type gasliquid distributor using the support vector regression algorithm optimized by particle swarm optimization.Compare the predicted results of the built model with the actual data,there is a good consistency between the two,this proves that the established PSO-SVR model can be used as a CFD proxy model to make good predictions.And compared the built model with the standard SVR algorithm and several classic neural network algorithms.The results show that the model established by this method has faster modeling speed,higher accuracy,and better robustness and generalization ability.In the optimization part,this paper uses the PSO-SVR model as a CFD proxy model to provide response target data for response surface design schemes,thereby reducing the huge time cost of CFD simulation calculations under various design schemes.Then in the response surface optimization module obtain the optimal structural parameters corresponding to the minimum unevenness of liquid distribution,so as to realize the structural optimization of the Venturi gas-liquid distributor.Afterwards,in order to ensure the accurate value of the optimization results obtained using the PSO-SVR-RSM method,then CFD simulation is used to perform three-dimensional modeling,meshing and simulation calculations on the optimized structure.The CFD verification results show good consistency with the response surface optimization results,the coefficients of uneven liquid distribution are 0.159 and 0.162,and this result is much lower than the unevenness of liquid distribution corresponding to the optimal structure proposed in the literature.
Keywords/Search Tags:gas-liquid distributor, support vector regression, particle swarm optimization, response surface methodology, computational fluid dynamics
PDF Full Text Request
Related items