Mohammad Amin Moradkhani, Ali Reza Miroliaei, Nasim Ghasemi, Seyyed Hossein Hosseini, Mikel Tellabide, Martin Olazar
{"title":"Minimum spouting velocity of fine particles in fountain confined conical spouted beds using machine learning and least square fitting approaches","authors":"Mohammad Amin Moradkhani, Ali Reza Miroliaei, Nasim Ghasemi, Seyyed Hossein Hosseini, Mikel Tellabide, Martin Olazar","doi":"10.1002/cjce.25429","DOIUrl":null,"url":null,"abstract":"<p>The present study concerns the development of new models to estimate the minimum spouting velocity (<i>U</i><sub>ms</sub>) in various configurations of fountain-confined conical spouted beds (FC-CSBs) with fine particles. Existing literature correlations were found to be inaccurate for FC-CSBs. Therefore, smart modelling techniques were employed to design more accurate predictive tools. The radial basis function (RBF) approach provided the best predictions for systems without draft tubes as well as those with open-sided draft tubes. Additionally, the Gaussian process regression (GPR) approach yielded the best predictions for systems with nonporous draft tubes. The mean absolute percentage error (MAPE) values for the testing phase were 5.80%, 5.67%, and 5.59%, respectively. These models consider how bed shape and particle properties affect <i>U</i><sub>ms</sub>. The sensitivity analysis was conducted to determine the factors with more importance in controlling <i>U</i><sub>ms</sub>. Finally, simpler correlations were derived for <i>U</i><sub>ms</sub> prediction in different FC-CSB configurations, with accuracy around 12% error.</p>","PeriodicalId":9400,"journal":{"name":"Canadian Journal of Chemical Engineering","volume":"103 2","pages":"880-898"},"PeriodicalIF":1.6000,"publicationDate":"2024-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Canadian Journal of Chemical Engineering","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/cjce.25429","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
引用次数: 0
Abstract
The present study concerns the development of new models to estimate the minimum spouting velocity (Ums) in various configurations of fountain-confined conical spouted beds (FC-CSBs) with fine particles. Existing literature correlations were found to be inaccurate for FC-CSBs. Therefore, smart modelling techniques were employed to design more accurate predictive tools. The radial basis function (RBF) approach provided the best predictions for systems without draft tubes as well as those with open-sided draft tubes. Additionally, the Gaussian process regression (GPR) approach yielded the best predictions for systems with nonporous draft tubes. The mean absolute percentage error (MAPE) values for the testing phase were 5.80%, 5.67%, and 5.59%, respectively. These models consider how bed shape and particle properties affect Ums. The sensitivity analysis was conducted to determine the factors with more importance in controlling Ums. Finally, simpler correlations were derived for Ums prediction in different FC-CSB configurations, with accuracy around 12% error.
期刊介绍:
The Canadian Journal of Chemical Engineering (CJChE) publishes original research articles, new theoretical interpretation or experimental findings and critical reviews in the science or industrial practice of chemical and biochemical processes. Preference is given to papers having a clearly indicated scope and applicability in any of the following areas: Fluid mechanics, heat and mass transfer, multiphase flows, separations processes, thermodynamics, process systems engineering, reactors and reaction kinetics, catalysis, interfacial phenomena, electrochemical phenomena, bioengineering, minerals processing and natural products and environmental and energy engineering. Papers that merely describe or present a conventional or routine analysis of existing processes will not be considered.