原油总碳含量的粒子群优化集成神经网络预测模型
|
摘要
原油评价新技术的研究和应用成为目前世界石油炼制企业致力发展的方向,也是今后发展的必然趋势.本文采用核磁共振(nuclear magnetic resonance, NMR)光谱技术和粒子群优化集成神经网络(particle swarm optimiza tion-ensemble neural network, PSO-ERNN)建立了一种快速评价原油总碳物性指标预测模型.该模型以随机向量函数连接网络(random vector functional link network, RVFL)作为基本模型,采用正则化负相关学习策略集成基本模型,并采用粒子群优化算法优化各基本模型的最优隐含层节点数(L)以及集成规模的最佳集成个数(M),最后利用在线学习方法对模型进行更新.实例验证表明,所提出的模型显著提高了预报精度,避免了随机选择L和M对模型精度的影响,对提高原油评价精度与效率和及时满足加工炼制要求具有应用价值.
The research and application of the new technology of crude oil evaluation has become the trend of the world petroleum refining enterprises, and it is also the inevitable trend of development in the future. In this paper, a new method to rapidly evaluate total carbon of crude oil was established by using nuclear magnetic resonance(NMR) spectroscopy and ensemble neural network with random weights(ERNN) model which is optimized by particle swarm. The model uses the random vector functional link network(RVFL) as the basic model. A regularized negative correlation learning strategy is used to integrate the basic model. The particle swarm optimization(PSO) algorithm is used to optimize the hidden layer number(L) of each basic model and the number of integrated scale(M). Finally, the online learning method is used to update the model. Experiment results show that the proposed model significantly improves the prediction accuracy and avoids the influence of random choice of L and M on model precision. The model also improves the accuracy and efficiency of crude oil evaluation and it has a practical value to meet the requirements of oil refining in time.
The research and application of the new technology of crude oil evaluation has become the trend of the world petroleum refining enterprises, and it is also the inevitable trend of development in the future. In this paper, a new method to rapidly evaluate total carbon of crude oil was established by using nuclear magnetic resonance(NMR) spectroscopy and ensemble neural network with random weights(ERNN) model which is optimized by particle swarm. The model uses the random vector functional link network(RVFL) as the basic model. A regularized negative correlation learning strategy is used to integrate the basic model. The particle swarm optimization(PSO) algorithm is used to optimize the hidden layer number(L) of each basic model and the number of integrated scale(M). Finally, the online learning method is used to update the model. Experiment results show that the proposed model significantly improves the prediction accuracy and avoids the influence of random choice of L and M on model precision. The model also improves the accuracy and efficiency of crude oil evaluation and it has a practical value to meet the requirements of oil refining in time.
引文
[1] BARUNIK J, MANLINSKA B. Forecasting the term structure of crude oil futures prices with neural networks. Applied Energy, 2016,164(2):366–379.
[2] CHATRATH A, MIAO H, RAMCHANDER S. An examination of the flow characteristics of crude oil:evidence from risk-neutral moments. Energy Economics, 2016, 54(1):213–223.
[3] HE K X, QIAN F, CHENG H, et al. A novel adaptive algorithm with near-infrared spectroscopy and its application in online gasoline blending processes. Chemometrics and Intelligent Laboratory Systems, 2015, 140(1):117–125.
[4] QIU Teng, YUAN Hongfu, SUN Jia, et al. The rapid evaluation of crude petroleum by near infrared spectroscopy(NIR). Spectroscopy and Spectral Analysis, 2008, 28(10):131–132.(邱藤,袁洪福,孙甲,等.近红外光谱快速原油评价研究.光谱学与光谱分析, 2008, 28(10):131–132.)
[5] LI Danting, ZHANG Changjiang, WANG Jin. Infrared spectrum(IR)analysis based on wavelet transform. Spectroscopy and Spectral Analysis, 2006, 26(11):2024–2026.(李丹婷,张长江,汪劲,等.基于小波变换的红外光谱分析.光谱学与光谱分析, 2006, 26(11):2024–2026.)
[6] DANIEL M, URIBE U, MURGICH J. Correlations between SARA fractions and physicochemical properties with 1HNMR spectra of vacuum residues from Colombian crude oils. Fuel, 2010, 89(1):185–192.
[7] CHEN Pu, CHU Xiaoli. Development of rapid analytical technologies in crude and heavy oil. Journal of Instrumental Analysis, 2012,31(9):1191–1198.(陈瀑,褚小立.原油及重油的快速分析技术进展.分析测试学报,2012, 31(9):1191–1198.)
[8] CHU Xiaoli, TIAN Songbai, XU Yupeng, et al. A study of crude oil rapid assay by near infrared spectroscopy. Petroleum Procession and Petrochemicals, 2012, 43(1):72–77.(褚小立,田松柏,许育鹏,等.近红外光谱用于原油快速评价的研究.石油炼制与化工, 2012, 43(1):72–77.)
[9] LUAN Guohong, HE Kaixun, CHENG Hui, et al. Octane model based on neural network by near-infrared spectroscopy and its application. Computers and Applied Chemistry, 2014, 31(1):63–68.(栾郭宏,贺凯迅,程辉,等.基于神经网络的近红外光谱辛烷值模型的研究及应用.计算机与应用化学, 2014, 31(1):63–68.)
[10] LIN Limin, CHEN Jian, JIN Jiajian. Nuclear magnetic resonance online analysis technology and its application in petrochemical unit. Automation in Petro-chemical Industry, 2004, 40(3):55–59.(林立敏,陈建,金加剑.核磁共振在线分析技术及其在炼油和化工装置中的应用.石油化工自动化, 2004, 40(3):55–59.)
[11] BREIMAN L. Bagging predictors. Maching Learning, 1996, 24(2):123–140.
[12] MOLINA V, URIBE U, MURGICH J. Partial least-squares(PLS)correlation between refined product yields and physicochemical properties with the 1H nuclear magnetic resonance(NMR)spectra of colombian crude oils. Energy and Fuels, 2007, 21(3):1674–1680.
[13] WANG Wei, CHAI Tianyou, ZHAO Lijie. Dynamic partial least squares modeling with recurrent neural networks of stable learning.Control Theory&Applications, 2012, 29(3):337–341.(王魏,柴天佑,赵立杰.带有稳定学习的递归神经网络动态偏最小二乘建模.控制理论与应用, 2012, 29(3):337–341.)
[14] CHENG W J, DING J L, KONG W J, et al. An adaptive chaotic PSO for parameter optimization and feature extraction of LS--SVM based modeling. 2011 American Control Conference. Rochester, NY, USA:IEEE, 2011:3263–3268.
[15] HANSEN L K, SALAMON P. Neural network ensembles. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990,12(10):993–1001.
[16] LIU Y, YAO X. Ensemble learning via negative correlation. Neural Networks, 1999, 12(10):1399–1404.
[17] CHEN H H, YAO X. Regularized negative correlation learning for neural network ensembles. IEEE Transactions on Neural Networks,2009, 20(12):1962–1979.
[18] DING J L, WANG H T, LI CH B, et al. An online learning neural network ensembles with random weights for regression of sequential data stream. Soft Computing, 2017, 21(20):5919–5937.
[19] MASILI A, PULIGHEDDU S, SASSU L, et al. Prediction of physical-chemical properties of crude oils by1HONMR analysis of neat samples and chemometrics. Magnetic Resonance in Chemistry,2012, 50(11):729–738.
[2] CHATRATH A, MIAO H, RAMCHANDER S. An examination of the flow characteristics of crude oil:evidence from risk-neutral moments. Energy Economics, 2016, 54(1):213–223.
[3] HE K X, QIAN F, CHENG H, et al. A novel adaptive algorithm with near-infrared spectroscopy and its application in online gasoline blending processes. Chemometrics and Intelligent Laboratory Systems, 2015, 140(1):117–125.
[4] QIU Teng, YUAN Hongfu, SUN Jia, et al. The rapid evaluation of crude petroleum by near infrared spectroscopy(NIR). Spectroscopy and Spectral Analysis, 2008, 28(10):131–132.(邱藤,袁洪福,孙甲,等.近红外光谱快速原油评价研究.光谱学与光谱分析, 2008, 28(10):131–132.)
[5] LI Danting, ZHANG Changjiang, WANG Jin. Infrared spectrum(IR)analysis based on wavelet transform. Spectroscopy and Spectral Analysis, 2006, 26(11):2024–2026.(李丹婷,张长江,汪劲,等.基于小波变换的红外光谱分析.光谱学与光谱分析, 2006, 26(11):2024–2026.)
[6] DANIEL M, URIBE U, MURGICH J. Correlations between SARA fractions and physicochemical properties with 1HNMR spectra of vacuum residues from Colombian crude oils. Fuel, 2010, 89(1):185–192.
[7] CHEN Pu, CHU Xiaoli. Development of rapid analytical technologies in crude and heavy oil. Journal of Instrumental Analysis, 2012,31(9):1191–1198.(陈瀑,褚小立.原油及重油的快速分析技术进展.分析测试学报,2012, 31(9):1191–1198.)
[8] CHU Xiaoli, TIAN Songbai, XU Yupeng, et al. A study of crude oil rapid assay by near infrared spectroscopy. Petroleum Procession and Petrochemicals, 2012, 43(1):72–77.(褚小立,田松柏,许育鹏,等.近红外光谱用于原油快速评价的研究.石油炼制与化工, 2012, 43(1):72–77.)
[9] LUAN Guohong, HE Kaixun, CHENG Hui, et al. Octane model based on neural network by near-infrared spectroscopy and its application. Computers and Applied Chemistry, 2014, 31(1):63–68.(栾郭宏,贺凯迅,程辉,等.基于神经网络的近红外光谱辛烷值模型的研究及应用.计算机与应用化学, 2014, 31(1):63–68.)
[10] LIN Limin, CHEN Jian, JIN Jiajian. Nuclear magnetic resonance online analysis technology and its application in petrochemical unit. Automation in Petro-chemical Industry, 2004, 40(3):55–59.(林立敏,陈建,金加剑.核磁共振在线分析技术及其在炼油和化工装置中的应用.石油化工自动化, 2004, 40(3):55–59.)
[11] BREIMAN L. Bagging predictors. Maching Learning, 1996, 24(2):123–140.
[12] MOLINA V, URIBE U, MURGICH J. Partial least-squares(PLS)correlation between refined product yields and physicochemical properties with the 1H nuclear magnetic resonance(NMR)spectra of colombian crude oils. Energy and Fuels, 2007, 21(3):1674–1680.
[13] WANG Wei, CHAI Tianyou, ZHAO Lijie. Dynamic partial least squares modeling with recurrent neural networks of stable learning.Control Theory&Applications, 2012, 29(3):337–341.(王魏,柴天佑,赵立杰.带有稳定学习的递归神经网络动态偏最小二乘建模.控制理论与应用, 2012, 29(3):337–341.)
[14] CHENG W J, DING J L, KONG W J, et al. An adaptive chaotic PSO for parameter optimization and feature extraction of LS--SVM based modeling. 2011 American Control Conference. Rochester, NY, USA:IEEE, 2011:3263–3268.
[15] HANSEN L K, SALAMON P. Neural network ensembles. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990,12(10):993–1001.
[16] LIU Y, YAO X. Ensemble learning via negative correlation. Neural Networks, 1999, 12(10):1399–1404.
[17] CHEN H H, YAO X. Regularized negative correlation learning for neural network ensembles. IEEE Transactions on Neural Networks,2009, 20(12):1962–1979.
[18] DING J L, WANG H T, LI CH B, et al. An online learning neural network ensembles with random weights for regression of sequential data stream. Soft Computing, 2017, 21(20):5919–5937.
[19] MASILI A, PULIGHEDDU S, SASSU L, et al. Prediction of physical-chemical properties of crude oils by1HONMR analysis of neat samples and chemometrics. Magnetic Resonance in Chemistry,2012, 50(11):729–738.