融合机理与数据的篦冷机温度软测量模型
|
摘要
针对篦冷机内部温度难以在线测量的问题,提出一种融合渗流换热机理与神经网络数据辨识的篦冷机温度软测量模型。首先考虑局部非热平衡效应,并结合多孔介质渗流换热理论建立床层内气固换热的机理模型;其次以床层边界条件与机理模型的温度预测值作为网络输入量,机理模型的温度预测偏差作为输出量,建立双并联前馈神经网络,并采用快速学习网算法训练网络参数;将辨识后的人工神经网络作为篦冷机气固渗流换热机理的补偿器,建立混合温度软测量模型。利用篦冷机生产运行数据进行了二次风温的软测量实验,实验结果表明,与机理模型相比混合软测量模型测量精度有较大提高,并且与实验数据的误差较小,混合软测量模型对于篦冷机温度具有较好的测量精度。
The temperature inside the grate cooler is difficult to be measured on line. To solve this problem,one kind of soft-sensing model of integration seepage heat transfer mechanism and neural network data identification is proposed. Firstly,the non-equilibrium of local thermal is considered. The mechanism model of gas-solid heat exchange based on the porous medium seepage heat transfer theory is formulated. Then,the boundary conditions and temperature prediction of the mechanism model is utilized as the network input. The temperature prediction deviation of the mechanism model as output a double-parallel feedforward neural network is established. The fast learning network algorithm is adopted to obtain the parameters of the double-parallel feedforward neural network. Finally,the neural network is used as the error compensator for the seepage heat transfer mechanism model. In this way,a mixed temperature soft measurement model is established. The soft measurement experiment of secondary wind temperature is carried out by using the production and operation data of grate cooler. Compared with the mechanism model,experimental results show that the measurement precision of the mixed soft measurement model is improved obviously. The error of the experimental data is small. The mixed soft-sensing model has the advantage of high accuracy for the temperature measurement inside the grate cooler.
The temperature inside the grate cooler is difficult to be measured on line. To solve this problem,one kind of soft-sensing model of integration seepage heat transfer mechanism and neural network data identification is proposed. Firstly,the non-equilibrium of local thermal is considered. The mechanism model of gas-solid heat exchange based on the porous medium seepage heat transfer theory is formulated. Then,the boundary conditions and temperature prediction of the mechanism model is utilized as the network input. The temperature prediction deviation of the mechanism model as output a double-parallel feedforward neural network is established. The fast learning network algorithm is adopted to obtain the parameters of the double-parallel feedforward neural network. Finally,the neural network is used as the error compensator for the seepage heat transfer mechanism model. In this way,a mixed temperature soft measurement model is established. The soft measurement experiment of secondary wind temperature is carried out by using the production and operation data of grate cooler. Compared with the mechanism model,experimental results show that the measurement precision of the mixed soft measurement model is improved obviously. The error of the experimental data is small. The mixed soft-sensing model has the advantage of high accuracy for the temperature measurement inside the grate cooler.
引文
[1]MUJUMDAR K S,GANESH K,KULKARNI S B,et al.Rotary cement kiln simulator(Ro CKS):Integrated modeling of pre-heater,calciner,kiln and clinker cooler[J].Chemical Engineering Science,2007,62(9):2590-2607.
[2]GARDEIK H,ROSEMANN H,STEINBACH V.Thermal assessment of clinker coolers-recommendations of the VDZ working group on clinker coolers[J].Zement-Kalk-Gips,Edition A,1987,40(5):230-237.
[3]TOUIL D,BELABED H,FRANCES C,et al.Heat exchange modeling of a grate clinker cooler and entropy production analysis[J].International Journal of Heat and Technology,2005,23(1):61-68.
[4]TRUBAEV P A.Exergy analysis of thermal processes in the building materials industry[J].Theoretical Foundations of Chemical Engineering,2006,40(2):175-182.
[5]LIU Z,WANG Z,YUAN M,et al.Thermal efficiency modelling of the cement clinker manufacturing process[J].Journal of the Energy Institute,2015,88(1):76-86.
[6]王美琪,刘彬,闻岩,等.篦冷机内水泥熟料温度的软测量[J].机械工程学报,2016,52(6):159-165.WANG M Q,LIU B,WEN Y,et al.Temperature softsensing of cement clinker in grate cooler[J].Journal of Mechanical Engineering,2016,52(6):159-165.
[7]SHAO W,CUI Z,CHENG L.Multi-objective optimization of cooling air distributions of grate cooler with different clinker particles diameters and air chambers by genetic algorithm[J].Applied Thermal Engineering,2017,111(1):77-86.
[8]SHAO W,CUI Z,CHENG L.Multi-objective optimization of cooling air distribution of grate cooler withdifferent inlet temperatures by using genetic algorithm[J].Science China Technological Sciences,2017,60(3):345-354.
[9]李海滨,郝晓辰,刘彬,等.基于模糊神经网络的篦式冷却机熟料流动建模[J].仪器仪表学报,2005,26(S2):171-173.LI H B,HAO X C,LIU B,et al.Modelling for clinker flow of the grate cooler based on fuzzy neural networks[J].Chinese Journal of Scientific Instrument,2005,26(S2):171-173.
[10]PANI A K,VADLAMUDI V K,MOHANTA H K.Development and comparison of neural network based soft sensors for online estimation of cement clinker quality[J].Isa Transactions,2013,52(1):19-29.
[11]赵朋程,刘彬,孙超,等.基于IQPSO优化ELM的熟料质量指标软测量研究[J].仪器仪表学报,2016,37(10):2243-2250.ZHAO P CH,LIU B,SUN CH,et al.Soft sensor for cement clinker quality indicator based on IQPSO optimize ELM[J].Chinese Journal of Scientific Instrument,2016,37(10):2243-2250.
[12]赵彦涛,单泽宇,常跃进,等.基于MI-LSSVM的水泥生料细度软测量建模[J].仪器仪表学报,2017,38(2):487-496.ZHAO Y T,SHAN Z T,CHANG Y J,et al.Soft sensor modeling for cement fineness based on least squares support vector machine and mutual information[J].Chinese Journal of Scientific Instrument,2017,38(2):487-496.
[13]DEHGHAN M,VALIPOUR M S,KESHMIRI A,et al.On the thermally developing forced convection through a porous material under the local thermal non-equilibrium condition:An analytical study[J].International Journal of Heat and Mass Transfer,2016,92(1):815-823.
[14]冯绍航,徐德龙,李辉,等.篦冷机中气固两相换热过程的模拟研究[J].西安建筑科技大学学报(自然科学版),2007,39(2):224-229.FENG SH H,XU D L,LI H,et al.Simulation study on heat transfer between clinker and gas of the grate cooler[J].Journal of Xi'an University of Architecture and Technology(Natural&Science),2007,39(2):224-234.
[15]胡道和,徐德龙,蔡玉良.气固过程工程学及其在水泥工业中的应用[M].武汉:武汉理工大学出版社,2003:132-133.HU D H,XU D L,CAI Y L.Gas Solid Process Engineering and Its Application in Cement Industry[M].Wuhan:Wuhan University of Technology Press,2003:132-133.
[16]姜培学,司广树,任泽霈.粘性耗散及变物性对多孔介质中对流换热的影响研究[J].工程热物理学报,2000,21(5):590-594.JIANG P X,SI G SH,REN Z P.Numerical investigation on the effects of viscous dissipation and variable thermophysical properties on forced convection heat transfer in porous media[J].Journal of Engineering Thermophysics,2000,21(5):590-594.
[17]李国强.新型人工智能技术研究及其在锅炉燃烧优化中的应用[D].秦皇岛:燕山大学,2013.LI G Q.Research of a novel artificial intelligent technology and its application to boiler combustion optimization[D].Qinhuangdao:Yanshan University,2013.
[18]HUANG G B,ZHU Q Y,SIEW C K.Extreme learning machine:Theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[19]LI G,NIU P,WANG H,LIU Y.Least square fast learning network for modeling the combustion efficiency of a 300 WM coal-fired boiler[J].Neural Networks,2014,51(3):57-66.
[2]GARDEIK H,ROSEMANN H,STEINBACH V.Thermal assessment of clinker coolers-recommendations of the VDZ working group on clinker coolers[J].Zement-Kalk-Gips,Edition A,1987,40(5):230-237.
[3]TOUIL D,BELABED H,FRANCES C,et al.Heat exchange modeling of a grate clinker cooler and entropy production analysis[J].International Journal of Heat and Technology,2005,23(1):61-68.
[4]TRUBAEV P A.Exergy analysis of thermal processes in the building materials industry[J].Theoretical Foundations of Chemical Engineering,2006,40(2):175-182.
[5]LIU Z,WANG Z,YUAN M,et al.Thermal efficiency modelling of the cement clinker manufacturing process[J].Journal of the Energy Institute,2015,88(1):76-86.
[6]王美琪,刘彬,闻岩,等.篦冷机内水泥熟料温度的软测量[J].机械工程学报,2016,52(6):159-165.WANG M Q,LIU B,WEN Y,et al.Temperature softsensing of cement clinker in grate cooler[J].Journal of Mechanical Engineering,2016,52(6):159-165.
[7]SHAO W,CUI Z,CHENG L.Multi-objective optimization of cooling air distributions of grate cooler with different clinker particles diameters and air chambers by genetic algorithm[J].Applied Thermal Engineering,2017,111(1):77-86.
[8]SHAO W,CUI Z,CHENG L.Multi-objective optimization of cooling air distribution of grate cooler withdifferent inlet temperatures by using genetic algorithm[J].Science China Technological Sciences,2017,60(3):345-354.
[9]李海滨,郝晓辰,刘彬,等.基于模糊神经网络的篦式冷却机熟料流动建模[J].仪器仪表学报,2005,26(S2):171-173.LI H B,HAO X C,LIU B,et al.Modelling for clinker flow of the grate cooler based on fuzzy neural networks[J].Chinese Journal of Scientific Instrument,2005,26(S2):171-173.
[10]PANI A K,VADLAMUDI V K,MOHANTA H K.Development and comparison of neural network based soft sensors for online estimation of cement clinker quality[J].Isa Transactions,2013,52(1):19-29.
[11]赵朋程,刘彬,孙超,等.基于IQPSO优化ELM的熟料质量指标软测量研究[J].仪器仪表学报,2016,37(10):2243-2250.ZHAO P CH,LIU B,SUN CH,et al.Soft sensor for cement clinker quality indicator based on IQPSO optimize ELM[J].Chinese Journal of Scientific Instrument,2016,37(10):2243-2250.
[12]赵彦涛,单泽宇,常跃进,等.基于MI-LSSVM的水泥生料细度软测量建模[J].仪器仪表学报,2017,38(2):487-496.ZHAO Y T,SHAN Z T,CHANG Y J,et al.Soft sensor modeling for cement fineness based on least squares support vector machine and mutual information[J].Chinese Journal of Scientific Instrument,2017,38(2):487-496.
[13]DEHGHAN M,VALIPOUR M S,KESHMIRI A,et al.On the thermally developing forced convection through a porous material under the local thermal non-equilibrium condition:An analytical study[J].International Journal of Heat and Mass Transfer,2016,92(1):815-823.
[14]冯绍航,徐德龙,李辉,等.篦冷机中气固两相换热过程的模拟研究[J].西安建筑科技大学学报(自然科学版),2007,39(2):224-229.FENG SH H,XU D L,LI H,et al.Simulation study on heat transfer between clinker and gas of the grate cooler[J].Journal of Xi'an University of Architecture and Technology(Natural&Science),2007,39(2):224-234.
[15]胡道和,徐德龙,蔡玉良.气固过程工程学及其在水泥工业中的应用[M].武汉:武汉理工大学出版社,2003:132-133.HU D H,XU D L,CAI Y L.Gas Solid Process Engineering and Its Application in Cement Industry[M].Wuhan:Wuhan University of Technology Press,2003:132-133.
[16]姜培学,司广树,任泽霈.粘性耗散及变物性对多孔介质中对流换热的影响研究[J].工程热物理学报,2000,21(5):590-594.JIANG P X,SI G SH,REN Z P.Numerical investigation on the effects of viscous dissipation and variable thermophysical properties on forced convection heat transfer in porous media[J].Journal of Engineering Thermophysics,2000,21(5):590-594.
[17]李国强.新型人工智能技术研究及其在锅炉燃烧优化中的应用[D].秦皇岛:燕山大学,2013.LI G Q.Research of a novel artificial intelligent technology and its application to boiler combustion optimization[D].Qinhuangdao:Yanshan University,2013.
[18]HUANG G B,ZHU Q Y,SIEW C K.Extreme learning machine:Theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[19]LI G,NIU P,WANG H,LIU Y.Least square fast learning network for modeling the combustion efficiency of a 300 WM coal-fired boiler[J].Neural Networks,2014,51(3):57-66.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |