基于实验设计和反向传播神经网络的板式传热元件性能评估  

Performance Evaluation of Plate Heat Transfer Components Based on Design of Experiment Method and Back Propagation Neural Network Method

作　　者：王幼石 焦育 李旭昆 马金伟[4] 姚立影[4] 刘一凡[4] 吕庆欢 张楠楠 WANG You-shi;JIAO Yu;LI Xu-kun;MA Jin-wei;YAO Li-ying;LIU Yi-fan;LUQing-huan;ZHANG Nan-nan(Southwest Oil&Gasfield Materials Branch,CNPG Chengdu 610031,China;PetroChina Tarim Oilfield Branch,Korla 841000,China;Golmud Refinery of PetroChina Qinghai Oilfield Company,Golmud 850099,China;Lanpec Technologies Limited,Lanzhou 730070,China)

机构地区：[1]中国石油西南油气田物资分公司,四川成都610031 [2]中国石油塔里木油田公司,新疆库尔勒841000 [3]中国石油青海油田分公司格尔木炼油厂,青海格尔木850099 [4]甘肃蓝科石化高新装备股份有限公司,甘肃兰州730070

出　　处：《石油化工设备》2025年第2期15-20,共6页Petro-Chemical Equipment

摘　　要：针对新型鼓泡板式传热元件的传热性能参数,采用实验设计的方法通过实验测定不同工况下的总传热系数,通过响应曲面法拟合得到板型、工艺参数和总传热系数的二阶响应曲面模型。同时根据实验结果,采用计算机仿真方法建立总传热系数反向传播神经网络预测模型。验证实验结果表明,两种模型的预测值与实验测量值都有良好的一致性,在工业应用中,可为采用该类新型传热元件的板式热交换器提供设计和选型的理论支撑。With regard to the heat transfer performance parameters of a new type of dimple plate heat transfer components,Design of Experiment method was used to obtain the total heat transfer coefficient under different operating conditions through experiments.The second-order response surface model of the plate shape,process parameters,and total heat transfer coefficient wasftted using the response surface method.At the same time,based on the experimental results,a computer simulation method was used to establish a BP(Back Propagation)neural network prediction model for the total heat transfer coefficient.The verification experiment results show that the predicted values of the two models are in good agreement with the experimentally measured values,which can provide theoretical support for the design and selection of plate heat exchangers using this type of new heat transfer element in industrial applications.

关 键 词：传热元件 性能 评估 实验设计 反向传播神经网络 

分 类 号：TQ051.5[化学工程]