基于WOA-LSSVM模型的近临界区CO2物性预测

doi:10.3969/j.issn.1000-1158.2023.05.20

摘要
图/表
参考文献(32)
相关文章 (15)

全文: PDF (80452 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对CO2状态方程难以准确预测CO2在近临界区的物性参数的问题,采用以鲸鱼优化算法(WOA)优化最小二乘支持向量机(LSSVM)的组合模型(WOA-LSSVM),对近临界区CO2物性进行预测。预测结果表明:同REFPROP软件与PSO-LSSVM模型相比,WOA-LSSVM模型预测近临界区CO2物性具有更高的精度。相比REFPROP软件,WOA-LSSVM模型将密度与粘度预测结果的均方根误差由133.67、9.33降至35.61、1.58,平均相对误差由31.8%、30.25%降至6.88%、3.88%,决定系数由0.59、0.62提高至0.86、0.83。此外,相对误差在10%以下占比均由0%分别提高到69.23%、92.31%。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	贺三
	唐凯
	张茂超
	薛雅文
	薛世奇

关键词 ：计量学, CO2物性预测, 鲸鱼优化算法, 最小二乘支持向量机, 近临界区, 密度, 粘度

Abstract：Aiming at the problem that the existing CO2 state equation is difficult to accurately predict the physical parameters of CO2 in the near-critical region, the combination model (WOA-LSSVM) of the least squares support vector machine (LSSVM) optimized by the whale optimization algorithm (WOA) was adopted to predict the physical properties of CO2 in the critical region. The prediction results show that compared with the REFPROP software and the PSO-LSSVM model, the WOA-LSSVM model has higher accuracy in predicting the physical properties of CO2 in the near-critical region. Compared with REFPROP software, the WOA-LSSVM model reduces the root mean square error of density and viscosity prediction results from 133.67, 9.33 to 35.61, 1.58, and the average relative error from 31.8%, 30.25% to 6.88%, 3.88%, the coefficient of determination is 0.59, 0.62 to 0.86, 0.83. In addition, the proportion of relative errors below 10% increased from 0% to 69.23% and 92.31%.

Key words： metrology prediction of CO2 physical properties whale optimization algorithm least squares support vector machine near-critical region density viscosity

收稿日期: 2022-03-21 发布日期: 2023-05-18

PACS:	TB99
	TB973

基金资助:四川省科技厅应用基础研究项目(2020YJ0393);油气藏地质及开发工程国家重点实验室开放基金(PLN0448)

作者简介: 贺三(1975-), 男, 四川达州人, 西南石油大学教授, 硕士生导师, 主要从事油气流动保障技术、管道腐蚀与防护技术及油气田节能减排技术等方面的研究。Email: hesan@126.com

引用本文:

贺三,唐凯,张茂超,薛雅文,薛世奇. 基于WOA-LSSVM模型的近临界区CO2物性预测[J]. 计量学报, 2023, 44(5): 803-809.
HE San,TANG Kai,ZHANG Mao-chao,XUE Ya-wen,XUE Shi-qi. Prediction of CO2 Physical Properties in Near-Critical Region Based on WOA-LSSVM Model. Acta Metrologica Sinica, 2023, 44(5): 803-809.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2023.05.20 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2023/V44/I5/803

［8］	刘贵军, 陈德奇, 胡练, 等. 基于CFD方法的PCHE窄小流道内超临界CO2流动传热特性数值研究［J］. 核动力工程, 2019, 40(3): 12-16.
［16］	Iwasaki H, Takahashi M. Viscosity of carbon dioxide and ethane［J］. The Journal of Chemical Physics, 1981, 74(3):1930-1943.
［6］	程健. 超临界CO2流动传热的数值分析［D］. 北京:华北电力大学, 2020.
［9］	王科技. 竖直管内超临界CO2流动与传热特性的数值模拟和实验研究［D］. 济南:山东大学, 2021.
［11］	孙博文. 高含硫气藏硫溶解度理论模型研究［D］. 成都:西南石油大学, 2018.
［28］	曹杰, 王维庆, 王海云, 等. 基于改进PSO-LSSVM的电采暖短期负荷预测研究［J］. 计算机仿真. 2022, 39(2):34-39.
［29］	金秀章, 李京. 基于互信息PSO-LSSVM的SO2浓度预测［J］. 计量学报. 2021, 42(5): 675-680.
［31］	左智科, 陈国彬, 刘超, 等. 反馈精英鲸鱼优化算法优化LSSVM的热耗率软测量建模［J］. 计量学报, 2019, 40(2):259-265.
	Teng K. Improvement of the calculation method for the pressure of the pipeline node of the carbon dioxide fire extinguishing system［J］. Industrial Water & Wastewater, 2004(2):68-71.
［3］	Hendriks E, Kontogeorgis G Industrial Requirements for Thermodynamics and Transport Properties［J］. Industrial & Engineering Chemistry Research, 2010, 49(22): 11131-11141.
［4］	Mazzoccoli M, Bosio B, Arato E. Analysis and Comparison of Equations-of-State with p-ρ-T Experimental Data for CO2 and CO2-Mixture Pipeline Transport［J］. Energy Procedia, 2012, 23: 274-283.
［7］	朱兵国, 吴新明, 张良, 等. 垂直上升管内超临界CO2流动传热特性研究［J］. 化工学报, 2019, 70(4):1282-1290.
［12］	张倩. 多环芳香烃在超临界CO2中的溶解度模型研究［D］. 成都:西南石油大学, 2017.
［15］	何欣欣,裴东升,陈会勇,等. 二氧化碳热物性方程研究进展及应用［J］. 热力发电, 2021, 50(5):27-33.
［24］	Iwasaki H, Takahashi M. Viscosity of carbon dioxide and ethane［J］. The Journal of Chemical Physics, 1981, 74(3):1930-1943.
［26］	Vapnik V N. Statistical learning theory［M］. New York:Wiley, 1998.
［2］	Li K, Zhou X, Tu R, et al. The f characteristics of supercritical CO2 leakage from a pipeline［J］. Energy, 2014, 71:665-672.
［10］	陈华胜. 高含硫天然气中硫溶解度的预测模型研究［D］. 重庆:重庆大学, 2019.
［14］	Shams R, Esmaili S, Rashid S, et al. An intelligent modeling approach for prediction of thermal conductivity of CO2［J］. Journal of Natural Gas Science And Engineering, 2015, 27(1):138-150.
［18］	Zhang X, Zhang X, Han B, et al. determination of constant volume heat capacity of mixed supercritical fluids and study on the intermolecular interaction［J］. The Journal of Supercritical Fluids, 2002, 24(3):193-201.
［19］	Langenfeld J J, Hawthorne S B, Miller D J, et al. Method for determining the density of pure and modified supercritical fluids［J］. Analytical Chemistry (Washington), 1992, 64(19):2263-2266.
［20］	Torres A, Romero J, Macan A, et al. Near critical and supercritical impregnation and kinetic release of thymol in LLDPE films used for food packaging［J］. The Journal of Supercritical Fluids, 2014, 85:41-48.
［21］	Hojjati M, Yamini Y, Khajeh M, et al. Solubility of some statin drugs in supercritical carbon dioxide and representing the solute solubility data with several density-based correlations［J］. The Journal of Supercritical Fluids, 2007, 41(2):187-194.
［23］	Sih R, Dehghani F, Foster N R. Viscosity measurements on gas expanded liquid systems-Methanol and carbon dioxide［J］. The Journal of Supercritical Fluids, 2007, 41(1):148-157.
	Cao J, Wang W Q, Wang H Y, et al. Research on Short-term Load Forecasting of Electric Heating Based on Improved LSSVM［J］. Computer Simulation, 2022, 39(2):34-39.
	Jing X Z, Li J. Prediction of SO2 Concentration Based on Mutual Information PSO-LSSVM［J］. Acta Metrologica Sinica, 2021, 42(5):675-680.
	Zuo Z K, Chen G B, Liu C, et al. Soft Sensor Model of Heat Rate Based on Optimized LSSVM by FEWOA［J］. Acta Metrologica Sinica, 2019, 40(2):259-265.
［1］	滕凯. 二氧化碳灭火系统管法的改进［J］. 工业用水与废水, 2004(2):68-71.
	Liu G J, Chen D Q, Hu L, et al. Numerical Investigation on Flow and Heat Transfer Characteristics of S-CO2 in Narrow Space Channel of PCHE Based on CFD Simulation［J］. Nuclear Power Engineering, 2019, 40(3):12-16.
［17］	Araus K A, Casado V, Del Valle J M, et al. Cosolvent effect of ethanol on the solubility of lutein in supercritical carbon dioxide［J］. The Journal of Supercritical Fluids, 2019, 143:205-210.
［27］	Suykens J A, Vandewalle J. Least squares support vector machine classifiers［J］. Neural Processing Letters, 1999, 9(3):293-300.
	Wu Z Q, Wang G Y, Xie Z K, et al. Joint Estimation of the Capacity and SOC of Lithium Battery Based on WOA-UKF Algorithm［J］. Acta Metrologica Sinica, 2022, 43(5):649-656.
［32］	肖荣鸽, 靳帅帅, 庄琦, 等. 基于WOA-BP算法的持液率预测模型研究［J］. 化学工程, 2022, 50(1):67-73.
［5］	Clarke D D, Vasquez V R, Whiting W B, et al. Sensitivity and uncertainty analysis of heat-exchanger designs to physical properties estimation［J］. Applied Thermal Engineering, 2001, 21(10):993-1017.
	Zhu B G, Wu X M, Zhang L, et al. Flow and heat transfer characteristics of supercritical CO2 in vertical tube［J］. CIESC Journal, 2019, 70(4):1282-1290.
［13］	Afshin T, Ali B, Adel N, et al. Predictive model based on ANFIS for estimation of thermal conductivity of carbon dioxide［J］. Journal of Molecular Liquids, 2016, 224:1266-1274.
	He X X, Pei D S, Chen H Y, et al. Research progress and applications of thermophysical equations of carbon dioxide［J］. Thermal Power Generation, 2021, 50(5):27-33.
［22］	Vesovic V, Wakeham W A, Olchowy G A, et al. The transport properties of carbon dioxide［J］. Journal of Physical and Chemical Reference Data, 1990, 19(3):763-808.
［25］	van der Gulik P S. Viscosity of carbon dioxide in the liquid phase［J］. Physica A, 1997, 238(1):81-112.
［30］	Mirjalili S, Lewis A. The Whale Optimization Algorithm［J］. Advances in Engineering Software, 2016, 95:51-67.
［31］	吴忠强,王国勇,谢宗奎, 等. 基于WOA-UKF算法的锂电池容量与SOC联合估计［J］. 计量学报, 2022, 43(5):649-656.
	Xiao R G, Jin S S, Zhuang Q, et al. Prediction model of liquid holdup based on WOA-BP network［J］. Chemical Engineering, 2022, 50(1):67-73.