基于人工神经网络的工质基础物性预测

石油炼制与化工 ›› 2024, Vol. 55 ›› Issue (1): 180-188.

基于人工神经网络的工质基础物性预测

林美金¹,董轩¹,洪小东^1,2,廖祖维¹,孙婧元¹,杨遥¹,王靖岱¹,阳永荣¹

1. 浙江大学化学工程与生物工程学院
2. 浙江大学杭州国际科创中心

收稿日期:2023-09-18 修回日期:2023-10-25 出版日期:2024-01-12 发布日期:2024-01-15
通讯作者: 洪小东 E-mail:hongxiaodong@zju.edu.cn
基金资助:
国家自然科学基金项目;浙江省尖兵领雁计划项目

PREDICTION OF PHYSICAL PROPERTIES OF WORKING FLUID BASED ON ARTIFICIAL NEURAL NETWORK

Received:2023-09-18 Revised:2023-10-25 Online:2024-01-12 Published:2024-01-15
Contact: Xiaodong Hong E-mail:hongxiaodong@zju.edu.cn

摘要/Abstract

摘要： 烃类及卤代烃是制冷及余热发电等热力学循环系统潜在的理想工质，但其数量繁多且多数物性参数未知，建立准确的物性预测模型对新型工质的开发至关重要。从多个公开数据库中收集了2 500多种烃类及卤代烃分子（含C,H,F,Cl）的基础物性参数，包括正常沸点（T_b）、临界温度（T_c）、临界压力（p_c）、偏心因子（ω），构建了一个工质物性数据库；进一步，通过改进基团贡献-人工神经网络（GC-ANN）的方法，模型的输入参数除基团外，还加入相对分子质量、T_b、约化维纳指数，建立了预测烃类及卤代烃分子 T_b,T_c,p_c, ω的神经网络模型，所开发模型的预测误差小于传统的GC-ANN的误差。

关键词: 新型工质, 物性预测, 基团贡献法, BP神经网络

Abstract: Traditional working fluids in thermodynamic cycles, such as refrigeration and waste heat power generation, have been associated with issues such as ozone layer depletion and global warming. The development of efficient and environmentally friendly novel working fluids has become a research focus. Hydrocarbons and halogenated hydrocarbons are ideal candidates, but their large number and many unknown thermophysical properties make it crucial to establish accurate models for predicting these properties in order to screen new working fluids effectively. In this study, the basic thermophysical parameters of more than 2 500 hydrocarbons and halogenated hydrocarbons containing C, H, F, and Cl atoms were collected from various public databases, including normal boiling point(T_b) , critical temperature(T_c) , critical pressure (p_c) and acentric factor (ω), and furtherly, by improving the method of group contribution-artificial neural network (GC-ANN), a neural network model for predicting T_b,T_c,p_c,and ω of hydrocarbons and halogenated hydrocarbons containing C, H, F, and Cl atoms was established by adding relative molecular mass, T_b and approximate wiener index to the input parameters of the model. The prediction errors of the models developed in this study were smaller than those of the traditional GC-ANN.

Key words: new working fluid, property prediction, group contribution method, BP neural network

林美金董轩洪小东廖祖维孙婧元杨遥王靖岱阳永荣. 基于人工神经网络的工质基础物性预测[J]. 石油炼制与化工, 2024, 55(1): 180-188.

参考文献

[1] HE M, WANG C, CHEN J, et al. Prediction of the critical properties of mixtures based on group contribution theory [J]. Journal of Molecular Liquids, 2018, 271.
[2] MONDEJAR M E, CIGNITTI S, ABILDSKOV J, et al. Prediction of properties of new halogenated olefins using two group contribution approaches [J]. Fluid Phase Equilibria, 2017, 433: 79-96.
[3] 任嘉辉, 刘豫, 刘朝, et al. 基于分子指纹和拓扑指数的工质临界温度理论预测 [J]. 化工学报, 2022, 73(04): 1493-1500.
[4] 王小艳, 司继林, 张达, et al. 纯物质临界参数估算方法的研究进展 [J]. 化工进展, 2012, 31(09): 1871-1877.
[5] 周传光，杨福胜，胡仰栋，于为人. 由化合物的沸点及比重推算临界参数 [J]. 计算机与应用化学, 1994, (02): 123-126.
[6] VEJAHATI F, NIKOO M B, MOKHATAB S, et al. Simple Correlation Estimates Critical Properties of Alkanes [J]. Petroleum Science and Technology, 2007, 25(9).
[7] JOBACK K G, REID R C. ESTIMATION OF PURE-COMPONENT PROPERTIES FROM GROUP-CONTRIBUTIONS [J]. Chemical Engineering Communications, 1987, 57(1-6).
[8] AMBROSE D. Correlation and Estimation of Vapor-liquid Critical Properties: I. Critical Temperatures of Organic Compounds [J]. Div Chem, 1978.
[9] CONSTANTINOU L, GANI R. New group contribution method for estimating properties of pure compounds [J]. AIChE Journal, 1994, 40(10).
[10] JORGE M-M, ELADIO P-F. Estimation of Pure Compound Properties Using Group-Interaction Contributions [J]. AIChE Journal, 1999, 45(3).
[11] 梁英华, 马沛生. A New Group-Contribution Method for Critical Properties [J]. Chinese Journal of Chemical Engineering, 2000, (01): 78-83.
[12] LYMPERIADIS A, ADJIMAN C S, JACKSON G, et al. A generalisation of the SAFT-γ group contribution method for groups comprising multiple spherical segments [J]. Fluid Phase Equilibria, 2008, 274(1-2): 85-104.
[13] LAZZúS J. ρ– T– P prediction for ionic liquids using neural networks [J]. Journal of The Taiwan Institute of Chemical Engineers - J TAIWAN INST CHEM ENG, 2009, 40: 213-232.
[14] GHARAGHEIZI F, ESLAMIMANESH A, MOHAMMADI A H. Determination of Critical Properties and Acentric Factors of Pure Compounds Using the Artificial Neural Network Group Contribution Algorithm [J]. Journal of Chemical and Engineering Data: the ACS Journal for Data, 2011, 56(5).
[15] LUKAWSKI M Z, DIPIPPO R, TESTER J W. Molecular property methods for assessing efficiency of organic Rankine cycles [J]. Energy, 2018, 142.
[16] WANG Q, MA P, JIA Q, et al. Position Group Contribution Method for the Prediction of Critical Temperatures of Organic Compounds [J]. Journal of Chemical and Engineering Data - J CHEM ENG DATA, 2008, 53.
[17] FRIEND D, HUBER M. Thermophysical Property Standard Reference Data from NIST [J]. International Journal of Thermophysics, 1994, 15: 1279-1288.
[18] WILDING W V, ROWLEY R L, OSCARSON J L. DIPPR? Project 801 evaluated process design data [J]. Fluid Phase Equilibria, 1998, 150-151: 413-420.
[19] JOBACK K G, REID R C. ESTIMATION OF PURE-COMPONENT PROPERTIES FROM GROUP-CONTRIBUTIONS [J]. Chemical Engineering Communications, 1987, 57(1-6): 233-243.
[20] MARRERO J, GANI R. Group-contribution based estimation of pure component properties [J]. Fluid Phase Equilibria, 2001, 183.
[21] SUKUMAR D, ASHA C. Unified artificial neural network-group contribution method for predictions of normal boiling point and critical temperature of refrigerants and related compounds [J]. International Journal of Refrigeration, 2022, 140.
[22] SOBATI M A, ABOOALI D. Molecular based models for estimation of critical properties of pure refrigerants: Quantitative structure property relationship (QSPR) approach [J]. Thermochimica Acta, 2015, 602.
[23] LAI N A. Prediction of the critical properties: A simple accurate strategy applied to environmentally friendly substances HFOs [J]. Fluid Phase Equilibria, 2020, 506(C).
[24] MONDEJAR M E, CIGNITTI S, ABILDSKOV J, et al. Prediction of properties of new halogenated olefins using two group contribution approaches [J]. Fluid Phase Equilibria, 2017, 433.
[25] MAURO B, LUIGI M. Comparison between Multi-Linear- and Radial-Basis-Function-Neural-Network-Based QSPR Models for The Prediction of The Critical Temperature, Critical Pressure and Acentric Factor of Organic Compounds [J]. Molecules (Basel, Switzerland), 2018, 23(6).