基于动态神经网络的水泥熟料堆积体建模方法

硅酸盐通报 ›› 2021, Vol. 40 ›› Issue (1): 55-63.

所属专题：水泥混凝土

基于动态神经网络的水泥熟料堆积体建模方法

闻岩¹, 黄泊霖¹, 袁林¹, 刘浩然², 鞠艳旭¹

1.燕山大学机械工程学院,秦皇岛 066004;
2.燕山大学信息科学与工程学院,秦皇岛 066004

收稿日期:2020-08-08 修回日期:2020-10-13 出版日期:2021-01-15 发布日期:2021-02-07
通讯作者: 袁林,高级实验师。E-mail:Info@ysu.edu.cn
作者简介:闻岩(1963—),男,博士,教授。主要从事机械设计理论及水泥机械研究。E-mail:wenyan@ysu.edu.cn
基金资助:
国家自然科学基金(51076135);河北省自然科学基金(E2018203398)

Modeling Method of Cement Clinker Accumulation Body Based onDynamic Neural Network

WEN Yan¹, HUANG Bolin¹, YUAN Lin¹, LIU Haoran², JU Yanxu¹

1. School of Mechanical Engineering,Yanshan University,Qinhuangdao 066004,China;
2. School of InformationScience and Engineering,Yanshan University,Qinhuangdao 066004,China

Received:2020-08-08 Revised:2020-10-13 Online:2021-01-15 Published:2021-02-07

摘要/Abstract

摘要： 工程中存在许多颗粒堆积问题。为了构建堆积体离散元数值模型,首先设计了堆积体结构参数采集试验,并以工程中篦冷机内的熟料堆积为背景,使用水泥熟料组建堆积体,采用切片法获取切片图像,随后进行一系列处理,获得了堆积体内颗粒的相关结构参数。同时,为了能够重建堆积体模型,使用神经网络训练了一个预测模型,并进行了相关验证。模型运行良好,表明在一定程度上能够满足堆积体建模的要求。

关键词: 颗粒, 水泥熟料, 堆积体, 神经网络, 建模, 离散元模型

Abstract: The particle packing is an important issue in engineering.In order to construct the discrete element numerical model of the particle accumulation body,an experiment for collecting its structural parameters such as diameter and coordinate was designed firstly.With the cement clinker packing of the grate cooler in the background,cement clinker was used to form the accumulation body.And a series of images were collected by cutting the particle accumulation body into slices.Then a battery of treatments was carried out to obtain its structural parameters.At the same time,a prediction model was trained based on neural network in order to rebuild accumulation body model and its statistical accuracy was verified.The good precision of the model indicates that it meets the modeling requirements to a certain extent.

Key words: particle, cement clinker, accumulation body, neural network, modeling, discrete element model

中图分类号:

TQ172.1

闻岩, 黄泊霖, 袁林, 刘浩然, 鞠艳旭. 基于动态神经网络的水泥熟料堆积体建模方法[J]. 硅酸盐通报, 2021, 40(1): 55-63.

WEN Yan, HUANG Bolin, YUAN Lin, LIU Haoran, JU Yanxu. Modeling Method of Cement Clinker Accumulation Body Based onDynamic Neural Network[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2021, 40(1): 55-63.

参考文献

[1] 孙其诚,王光谦.颗粒物质力学导论[M].北京:科学出版社,2009:94+117.
SUN Q C,WANG G Q.Introduction to the mechanics of particulate matter[M].Beijing:China Science Publishing & Media Ltd.,2009:94+117 (in Chinese).
[2] NOLAN G T,KAVANAGH P E.Random packing of nonspherical particles[J].Powder Technology,1995,84(3):199-205.
[3] 刘彩玲,王亚丽,宋建农,等.基于三维激光扫描的水稻种子离散元建模及试验[J].农业工程学报,2016,32(15):294-300.
LIU C L,WANG Y L,SONG J N,et al.Experiment and discrete element model of rice seed based on 3D laser scanning[J].Transactions of the Chinese Society of Agricultural Engineering,2016,32(15):294-300 (in Chinese).
[4] 蒙建国,王春光,谢胜仕,等.基于3D扫描的马铃薯建模及运动仿真分析[J].农机化研究,2015,37(10):64-67.
MENG J G,WANG C G,XIE S S,et al.Based on 3D scanning modeling and motion simulation analysis of the potato[J].Journal of Agricultural Mechanization Research,2015,37(10):64-67 (in Chinese).
[5] 王美美,王万章,杨立权,等.基于EDEM的玉米子粒建模方法的研究[J].河南农业大学学报,2018,52(1):80-84+103.
WANG M M,WANG W Z,YANG L Q,et al.Research of discrete element modeling method of maize kernel based on EDEM[J].Journal of Henan Agricultural University,2018,52(1):80-84+103 (in Chinese).
[6] 余世科.基于离散元法的多尺度内聚颗粒模型构建及冲击破碎能耗分析[D].赣州:江西理工大学,2015.
YU S K.The establishment of multi-scale bonded particle model based on discrete element method and impact crushing energy consumption analysis[D].Ganzhou:Jiangxi University of Science and Technology,2015 (in Chinese).
[7] 李杨,佘成学,焦小亮.堆石料碾压试验的颗粒流模拟新方法[J].岩土力学,2017,38(10):3029-3038.
LI Y,SHE C X,JIAO X L.A new method for simulating rockfill roller compaction using particle flow code[J].Rock and Soil Mechanics,2017,38(10):3029-3038 (in Chinese).
[8] 李鹏鹏,周伟,熊美林,等.复杂形状颗粒DEM模拟及其对宏观力学响应影响研究[J].武汉大学学报(工学版),2018,51(6):478-486.
LI P P,ZHOU W,XIONG M L,et al.Study of DEM modeling of irregular shaped particle and its influence on macromechanical response[J].Engineering Journal of Wuhan University,2018,51(6):478-486 (in Chinese).
[9] 张文明.基于三维重建技术的篦冷机熟料冷却控制模型研究[D].秦皇岛:燕山大学,2010:35-37.
ZHANG W M.Research on cooling control model of grate cooler clinker based on three dimentional reconstruction[D].Qinhuangdao:Yanshan University,2010:35-37 (in Chinese).
[10] 张帅龙.基于LBM和FVM的水泥熟料换热研究[D].秦皇岛:燕山大学,2016:20.
ZHANG S L.Study of heat transfer in the cement clinker based on the LBM and FVM[D].Qinhuangdao:Yanshan University,2016:20 (in Chinese).
[11] 黄丰.多孔介质模型的三维重构研究[D].合肥:中国科学技术大学,2007:2-10.
HUANG F.Three-dimensional reconstruction and simulation of porous media[D].Hefei:University of Science and Technology of China,2007:2-10 (in Chinese).
[12] 马丁T.哈根,霍华德B.德姆斯,马克H.比勒,等.神经网络设计(原书第2版)[M].章毅,译.北京:机械工业出版社,2017:398-404.
Hagan M T,Demuth H B,Beale M H,el at.Neural network design,second edition[M].ZHANG Yi,translation.Beijing:China Machine Press,2017:398-404.
[13] WIDROW B,WINTER R.Neural nets for adaptive filtering and adaptive pattern recognition[J].Computer,1988,21(3):25-39.
[14] AMARI S,MURATA N,MULLER K R,et al.Asymptotic statistical theory of overtraining and cross-validation[J].IEEE Transactions on Neural Networks,1997,8(5):985-996.
[15] MÜLLER K R,FINKE M,MURATA N,et al.A numerical study on learning curves in stochastic multilayer feedforward networks[J].Neural Computation,1996,8(5):1085-1106.
[16] NGUYEN D,WIDROW B.Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights[C]//1990 IJCNN International Joint Conference on Neural Networks.June 17-21,1990,San Diego,CA,USA.IEEE,1990:21-26.
[17] MACKAY D J C.Bayesian interpolation[J].Neural Computation,1992,4(3):415-447.
[18] BAYES T.A letter from the late reverend mr.Thomas Bayes,F.R.S.to john canton,M.A.and F.R.S[J].Philosophical Transactions of the Royal Society of London,2006,53:269-271.
[19] GOUTTE C,LARSEN J.Adaptive regularization of neural networks using conjugate gradient[C]//Proceedings of the 1998 IEEE International Conference on Acoustics,Speech and Signal Processing,ICASSP '98 (Cat.No.98CH36181),May 15-15,1998,Seattle,1998:1201-1204.
[20] TIKHONOV A N.On the solution of ill-posed problems and the method of regularization[C].Dokl Akad Nauk Sssr,1963:501-504.
[21] SJÖBERG J,LJUNG L.Overtraining,regularization and searching for a minimum,with application to neural networks[J].International Journal of Control,1995,62(6):1391-1407.
[22] BOX G,JENKINS G M,REINSEL G C,et al.Time series analysis:forecasting and control[M].Fifth edition.Hoboken:Wiley,2016:21-31.

基于动态神经网络的水泥熟料堆积体建模方法

Modeling Method of Cement Clinker Accumulation Body Based onDynamic Neural Network

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王上, 刘松辉, 周蓉, 张跃宏, 张海波. 超轻硫铝酸盐水泥基发泡材料的制备及硬化性能研究[J]. 硅酸盐通报, 2021, 40(3): 723-730.
[2]	张福龙, 梁潇文. 半固态搅拌制 SiC 颗粒增强镁基复合材料过程的数值模拟[J]. 硅酸盐通报, 2021, 40(3): 984-989.
[3]	曹金成;张翔. 基于随机场和数学形态学的水泥基材料三维孔隙结构建模研究[J]. 硅酸盐通报, 2020, 39(9): 2769-2776.
[4]	李朝元;刘赞群;陈娟;朱嘉慧;周蕴婵. 水泥-乳化沥青-橡胶颗粒砂浆拉伸性能研究[J]. 硅酸盐通报, 2020, 39(8): 2549-2556.
[5]	闻岩;鞠艳旭;黄泊霖;袁林. 熟料颗粒角速度实验及预测研究[J]. 硅酸盐通报, 2020, 39(7): 2099-2105.
[6]	谢祥兵;马鹏;梁林园;童申家;李广慧. 纳米TiO2协同纳米ZnO/SBS复合改性沥青耐紫外光老化性能评价[J]. 硅酸盐通报, 2020, 39(7): 2354-2361.
[7]	陈伟;孟皞;颜岩;袁波. 废橡胶颗粒联合改性用于制备水泥砂浆机理研究[J]. 硅酸盐通报, 2020, 39(6): 1715-1721.
[8]	郭锐;刘丹;杜钰;余龙舟. 神经网络在矿物加工中预测应用的研究现状及展望[J]. 硅酸盐通报, 2020, 39(4): 1178-1185.
[9]	赵美程;饶美娟;邓青山;陶永征. 铜离子掺杂对高铁低钙水泥熟料性能的影响[J]. 硅酸盐通报, 2020, 39(2): 396-401.
[10]	王洒;杨谋存;朱跃钊. 蓝钨纳米颗粒的化学改性及其纳米流体热冲击稳定性实验研究[J]. 硅酸盐通报, 2020, 39(2): 631-638.
[11]	李延伟. 聚羧酸超塑化剂与纳米SiO2颗粒协同作用对固井水泥浆性能的影响[J]. 硅酸盐通报, 2020, 39(10): 3121-3125.
[12]	冯庆革;方灿东;杨义;吴飚;赵政术;柏秀奎;朱伟伟;王东波;刘峥. 不同重金属和氯对水泥熟料及其超早期水化特性的影响[J]. 硅酸盐通报, 2020, 39(10): 3230-3236.
[13]	袁琦;郑玉;李志辉;孟祥宇;周圆;刘文杰. 颗粒稳定泡沫法制备钢渣泡沫陶瓷[J]. 硅酸盐通报, 2020, 39(10): 3334-3339.
[14]	姚艳飞;丁松雄;陈延信. 水泥熟料全窑系统热效率模型构建[J]. 硅酸盐通报, 2020, 39(1): 124-131.
[15]	黄诗轶;夏令;宋少先. 有机物对针铁矿团聚氧化石墨烯(GO)的影响[J]. 硅酸盐通报, 2019, 38(7): 2329-233.