多向梯度孔径反应载体的结构主动设计  

Active structural design of reaction support with multi-directional gradient pore-sizes

作　　者：郑天清 邱泓桑 詹友基 徐继璇 章蒙蒙 王江江 李棒棒 ZHENG Tianqing;QIU Hongsang;ZHAN Youji;XU Jixuan;ZHANG Mengmeng;WANG Jiangjiang;LI Bangbang(Fujian Key Laboratory of Intelligent Machining Technology and Equipment,School of Mechanical&Automotive Engineering Fujian University of Technology,Fuzhou 350118 China)

机构地区：[1]福建理工大学机械与汽车工程学院福建省智能加工技术及装备重点实验室,福建福州350118

出　　处：《厦门大学学报(自然科学版)》2025年第2期327-335,共9页Journal of Xiamen University:Natural Science

基　　金：福建省自然科学基金(2024J01825);福建建省中青年教师教育科研项目(JAT220219);福建理工大学科研启动金项目(GY-Z220200);福建省财政厅2023年科技专项资金(GY-Z23274)。

摘　　要：[目的]探究隐式三角函数参数与多孔反应载体孔径大小之间的内在关系,实现具有多向梯度孔径的反应载体结构主动设计,进而为应用于传质传热和制氢反应的多孔反应载体的多向梯度孔径设计奠定基础.[方法]在经典隐式三角函数的基础上,解析其中与孔径大小相关的参数,并建立该参数与坐标位置之间的关联,开展梯度孔径反应载体的结构设计;同时通过变参数的方法,分析新型隐式三角函数参数的改变对多孔反应载体孔径梯度的影响;利用多孔反应载体孔径梯度与其对应的隐式三角函数参数训练BP神经网络,建立多孔反应载体孔径梯度与隐式三角函数参数之间的映射关系,进而实现孔径梯度的主动设计.[结果]I-WP型隐式三角函数可获得具有较高比表面积的多孔反应载体.在此基础上,通过建立隐式三角函数参数C与坐标位置之间的关联,实现了多向梯度孔径反应载体的结构设计.基于BP神经网络,实现了多孔反应载体的孔径梯度主动调控,其中,隐式三角函数参数A(用于调控载体的孔径大小及孔径梯度)的预测误差率在2%~14%,参数T(用于辅助调控载体的孔径大小)的预测误差率在0.1%~2%.[结论]隐式三角函数可实现多向梯度孔径反应载体的结构设计,BP神经网络可实现多孔反应载体的孔径梯度主动设计.[Objective]The relationship between parameters of implicit trigonometric function and the pore-sizes size of porous reaction support is explored,and the active design of the structure of reaction support with multi-directional gradient pore-sizes is accomplished.Consequently,we have laid a design foundation for the study of the influence of the gradient pore-sizes of porous reaction support on the mass and heat transfer capacity and the pore-sizes gradient of porous reaction support with high hydrogen production performance.[Methods]Four classical implicit trigonometric functions are used to build the porous reaction support.The implicit trigonometric function,which can help build the reaction support with high specific surface area,is selected with the multi-directional gradient pore-sizes.Based on the function,its parameters related to the pore-size are analyzed,and the correlation between these parameters and the coordinate position is established to carry out the structural design of the reaction support with gradient pore-sizes.Meanwhile,the influence,related to the change of parameters imbedded in the new implicit trigonometric function,on the pore-sizes gradient of the porous reaction support is analyzed by the method of variable parameters.Furthermore,BP neural network is used to build a mathematical model of the influence of pore-sizes gradient of porous reaction support on the selection of parameters of the implicit trigonometric function.As a result,we can predict these parameters of the implicit trigonometric function required by the porous reaction support with a specific pore-sizes gradient.Then,based on these predicted parameters of implicit trigonometric function,the structure of reaction support with gradient pore-sizes is constructed,and the active design of pore-sizes gradient of the support is attained.[Results]Specific surface areas of porous reaction supports,which are constructed using four classic implicit trigonometric functions,are 18.640,20.185,15.564,and 22.407 mm 2/mm 3,respectively.D

关 键 词：反应载体 主动设计 梯度孔径 隐式三角函数 神经网络 

分 类 号：TH12[机械工程—机械设计及理论]