基于深度置信网络的滴流床反应器持液量预测  

Prediction of liquid holdup in trickle bed reactor based on deep belief network

作　　者：翟剑 张玮[1] 李立毅 延会波 韩念琛[1] 黄伟[2] ZHAI Jian;ZHANG Wei;LI Li-yi;YAN Hui-bo;HAN Nian-chen;HUANG Wei(College of Chemistry and Chemical Engineering,Taiyuan University of Technology,Taiyuan 30024,China;Key Laboratory of Coal Science and Technology of Ministry of Education and Shanxi Province,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学化学化工学院,山西太原030024 [2]太原理工大学煤科学与技术教育部和山西省重点实验室,山西太原030024

出　　处：《天然气化工—C1化学与化工》2020年第6期93-98,共6页Natural Gas Chemical Industry

基　　金：国家重点研发项目(2018YFB0604603);山西省重点研发计划项目(201903D121027)。

摘　　要：针对滴流床反应器持液量预测难度大、精度低的问题,首次提出了一种基于深度置信网络(Deep Belief Network,DBN)的数据建模方法。首先采用受限玻尔兹曼机(Restricted Boltzmann Machine,RBM)提取实验数据的深层特征,然后使用一层反向传播(Back Propagation,BP)神经网络对深层特征进行训练,建立滴流床反应器持液量预测模型。模型测试结果表明,深度置信网络模型对滴流床反应器持液量具有良好的预测效果,测试样本的平均相对误差MRE为4.8%、决定系数R2为0.965,相比于传统的理论与半经验模型,具有精确性高、泛化性强的特点,且模型对滴流床反应器操作参数的变化具有显著性,符合持液量变化机理和反应器放大设计的要求。Aiming at the problem of high difficulty and low accuracy of the liquid holdup prediction in the trickle bed reactor,a data modeling method based on a deep belief network(DBN)was proposed for the first time.First,the Restricted Boltzmann Machine(RBM)was used to extract the deep features of the experimental data,and then the Back Propagation(BP)neural network was used to train the deep features to establish a prediction model for liquid holdup of trickle bed reactor.The test results of the model show that the deep belief network model has a good prediction effect on the liquid holdup of the trickle bed reactor.The Mean Relative Error(MRE)of the test sample is 4.8%and the determination coefficient R2 is 0.965.Compared with the traditional theory and semi-empirical model,the proposed model has the characteristics of high accuracy and strong generalization,and the model is significant to the change of the operating parameters of the trickle bed reactor,which meets the requirements of the liquid holdup change mechanism and the scale-up design of the reactor.

关 键 词：滴流床反应器 持液量 机器学习 深度置信网络 

分 类 号：TE62[石油与天然气工程—油气加工工程]