A machine learning based approach to solve the aerosol dynamics coagulation model

IF 2.8 4区环境科学与生态学 Q2 ENGINEERING, CHEMICAL Aerosol Science and Technology Pub Date : 2023-08-23 DOI:10.1080/02786826.2023.2249074

Onochie Okonkwo, Rahul Patel, R. Gudi, P. Biswas

{"title":"A machine learning based approach to solve the aerosol dynamics coagulation model","authors":"Onochie Okonkwo, Rahul Patel, R. Gudi, P. Biswas","doi":"10.1080/02786826.2023.2249074","DOIUrl":null,"url":null,"abstract":"Abstract Solving aerosol dynamic models accurately to obtain the size distribution function is often computationally expensive. Conventional artificial neural network (ANN) models offer an alternative procedure to solve the aerosol dynamic equations. However, conventional ANN models can result in violation of aerosol mass conservation. To further enhance accuracy and reduce computational time, a hybrid ANN approach to solve the aerosol coagulation equation is developed, validated, and demonstrated. The methodology and assumptions for the development of the hybrid ANN model which provides an analytical closed form solution for aerosol coagulation is described. The ANN model is trained and validated using a dataset from an accurate sectional model. Following this, the hybrid ANN aerosol model is used to describe the evolution of aerosol in a furnace aerosol reactor. The hybrid ANN model results are compared to the accurate sectional and moment coagulation models. The hybrid ANN coagulation model prediction was found to accurately describe the evolution of the size distribution at a computational cost which is slightly more than the moment model but orders of magnitude less than the sectional model. Copyright © 2023 American Association for Aerosol Research Graphical abstract","PeriodicalId":7474,"journal":{"name":"Aerosol Science and Technology","volume":"57 1","pages":"1098 - 1116"},"PeriodicalIF":2.8000,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Aerosol Science and Technology","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1080/02786826.2023.2249074","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Abstract Solving aerosol dynamic models accurately to obtain the size distribution function is often computationally expensive. Conventional artificial neural network (ANN) models offer an alternative procedure to solve the aerosol dynamic equations. However, conventional ANN models can result in violation of aerosol mass conservation. To further enhance accuracy and reduce computational time, a hybrid ANN approach to solve the aerosol coagulation equation is developed, validated, and demonstrated. The methodology and assumptions for the development of the hybrid ANN model which provides an analytical closed form solution for aerosol coagulation is described. The ANN model is trained and validated using a dataset from an accurate sectional model. Following this, the hybrid ANN aerosol model is used to describe the evolution of aerosol in a furnace aerosol reactor. The hybrid ANN model results are compared to the accurate sectional and moment coagulation models. The hybrid ANN coagulation model prediction was found to accurately describe the evolution of the size distribution at a computational cost which is slightly more than the moment model but orders of magnitude less than the sectional model. Copyright © 2023 American Association for Aerosol Research Graphical abstract

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于机器学习的气溶胶动力学凝结模型求解方法

摘要准确求解气溶胶动力学模型以获得尺寸分布函数通常计算成本高昂。传统的人工神经网络（ANN）模型为求解气溶胶动力学方程提供了一种替代方法。然而，传统的人工神经网络模型可能会违反气溶胶质量守恒。为了进一步提高精度和减少计算时间，开发、验证并演示了一种求解气溶胶凝结方程的混合人工神经网络方法。介绍了开发混合人工神经网络模型的方法和假设，该模型为气溶胶凝结提供了一个分析闭合形式的解决方案。使用来自精确截面模型的数据集来训练和验证ANN模型。随后，使用混合人工神经网络气溶胶模型来描述炉式气溶胶反应器中气溶胶的演变。将混合人工神经网络模型的结果与精确的截面和力矩凝固模型进行了比较。发现混合人工神经网络凝固模型预测可以准确地描述尺寸分布的演变，计算成本略高于矩模型，但比截面模型低几个数量级。版权所有©2023美国气溶胶研究协会图形摘要

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Aerosol Science and Technology 环境科学-工程：化工

CiteScore

8.40

自引率

7.70%

发文量

审稿时长

3 months

期刊介绍： Aerosol Science and Technology publishes theoretical, numerical and experimental investigations papers that advance knowledge of aerosols and facilitate its application. Articles on either basic or applied work are suitable. Examples of topics include instrumentation for the measurement of aerosol physical, optical, chemical and biological properties; aerosol dynamics and transport phenomena; numerical modeling; charging; nucleation; nanoparticles and nanotechnology; lung deposition and health effects; filtration; and aerosol generation. Consistent with the criteria given above, papers that deal with the atmosphere, climate change, indoor and workplace environments, homeland security, pharmaceutical aerosols, combustion sources, aerosol synthesis reactors, and contamination control in semiconductor manufacturing will be considered. AST normally does not consider papers that describe routine measurements or models for aerosol air quality assessment.