基于神经网络和遗传算法的螺旋折流板换热器性能预测

摘要

将人工神经网络（ANN）应用于非连续螺旋折流板换热器的壳程换热和流阻分析。中试试验研究了具有3个螺旋角和2种管型的换热器。作为人工神经网络最常用的一种类型，将多层感知器神经网络（MLP）应用于本研究，使用一定的实验数据进行网络训练及预测。应用遗传算法（GA）对MLP的初始权值和阈值进行优化，预测结果精确。通过比较不同网络结构的预测误差来选择最适宜的网络结构为9-7-5-2。和关联结果比较可知MLP-GA网络对于换热器性能预测更加适合。此外，当使用MLP-GA方法在训练数据范围以外对壳程换热系数和压降进行预测时，网络预测结果和实验结果吻合程度也较高。因此，MLP-GA混合算法能够用来预测螺旋折流板管壳式换热器的传热和水力学性能。

	Service

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	Email Alert
	RSS
	作者相关文章
	孙永利
	王华金
	郝丽
	肖晓明

关键词：螺旋折流板换热器传热多层感知遗传算法

Abstract：

In this paper, an artificial neural network (ANN) was applied for shell-side heat transfer and flow resistance analysis of shell-and-tube heat exchangers with non-continuous helical baffles. Heat exchangers with three helical angles and two types of tubes were experimentally investigated. One of the commonly used types of neural networks, the multilayer perception (MLP), was trained with limited experimental data. The genetic algorithm (GA) was employed to optimize the initial weights and biases of the MLP network, which showed more accuracy in prediction. Different network configurations were also studied to search for a relatively better network for the prediction. Comparisons with the correlations demonstrate the superiority of the MLP-GA network. Furthermore, the network yielded agreeable results when it was used for predicting the shell-side heat transfer rate and pressure drop outside the range of the experiments. It is recommended that the MLP-GA combined method might be used to predict the thermal and hydraulic performance of shell-and-tube heat exchangers with helical baffles.

Keywords： shell-and-tube heat exchanger with helical baffles； heat transfer； multilayer perception； genetic algorithm

Received 2014-05-23;

Corresponding Authors: 郝丽,E-mail:haolitju@163.com。 Email: haolitju@163.com

About author: 孙永利(1967-),男,博士,副研究员,研究方向为化工传质与分离工程。

引用本文:

孙永利, 王华金, 郝丽, 肖晓明.基于神经网络和遗传算法的螺旋折流板换热器性能预测[J]. 化学工业与工程, 2016,33(4): 49-55

Sun Yongli, Wang Huajin, Hao Li, Xiao Xiaoming.Performance Prediction of Shell-and-Tube Heat Exchangers with Helical Baffles Using Multilayer Perception Neural Networks Optimized with Genetic Algorithm[J]. Chemcial Industry and Engineering, 2016,33(4): 49-55

[1] Master B, Chunangad K, Boxma A,et al. Most frequently used heat exchangers from pioneering research to worldwide applications[J]. Heat Transfer Engineering, 2006, 27(6): 4-11
[2] Stehl K P, Wadekar V V. Different strategies to improve industrial heat exchange[J]. Heat Transfer Engineering, 2002, 23(6): 36-48
[3] Chen G, Wang Q. Experimental and numerical studies of shell-and-tube heat exchangers with helical baffles[J]. ASME, 2009: 601-609
[4] Jafari N M R, Shafeghat A. Fluid flow analysis and extension of rapid design algorithm for helical baffle heat exchangers[J]. Applied Thermal Engineering, 2008, 28(11/12): 1 324-1 332
[5] Kral D, Stehlik P, Van Der Ploeg H, et al. Helical baffles in shell-and-tube heat exchangers, Part I: Experimental verification[J]. Heat Transfer Engineering, 1996, 17(1): 93-101
[6] Lei Y, He Y, Chu P, et al. Design and optimization of heat exchangers with helical baffles[J]. Chemical Engineering Science, 2008, 63(17): 4 386-4 395
[7] Lutcha J, Nemcansky J. Performance improvement of tubular heat exchangers by helical baffles[J]. Chemical Engineering Research & Design, 1990, 68: 263-270
[8] Peng B, Wang Q, Zhang C, et al. An experimental study of shell-and-tube heat exchangers with continuous helical baffles[J]. Journal of Heat Transfer, 2007, 129(10): 1 425-1 431
[9] Zhang J, Li B, Huang W, et al. Experimental performance comparison of shell-side heat transfer for shell-and-tube heat exchangers with middle-overlapped helical baffles and segmental baffles[J]. Chemical Engineering Science, 2009, 64(8): 1 643-1 653
[10] Diaz G, Sen M, Yang K, et al. Dynamic prediction and control of heat exchangers using artificial neural networks[J]. International Journal of Heat and Mass Transfer, 2001, 45: 1 671-1 679
[11] Pacheco-Vega A, Sen M, Yang K, et al. Neural network analysis of fin-tube refrigerating heat exchanger with limited experimental data[J]. International Journal of Heat and Mass Transfer, 2001, 44(4): 763-770
[12] Islamoglu Y, Kurt A, Parmaksizoglu C. Performance prediction for non-adiabatic capillary tube suction line heat exchanger an artificial neural network[J]. Energy Conversion and Management, 2005, 46: 223-232
[13] Hao P, Xiang L. Neural networks analysis of thermal characteristics on plate-fin heat exchangers with limited experimental data[J]. Applied Thermal Engineering, 2009, 29: 2 251-2 256
[14] Xie G, Wang Q, Zeng M, et al. Heat transfer analysis for shell-and-tube heat exchangers with experimental data by artificial neural networks approach[J]. Applied Thermal Engineering, 2007, 27(5/6): 1 096-1 104