汽油机瞬态工况油膜参数混沌时序LS-SVM预测研究
|
摘要
汽油机油膜参数具有多维非线性特性,当前使用的试验标定法及辨识法难以精确确定参数值,对此提出了混沌时序最小二乘支持向量机(LS-SVM)预测模型。已知汽油机油路系统在时间序列具有非线性混沌特性,对油膜参数试验标定数据进行相空间重构,采用支持向量机对重构后的数据进行训练及预测,得出预测结果,与BP神经网络模型及Elman神经网络模型的预测结果进行了对比分析。验证了LS-SVM模型具有更强的非线性预测能力,能够有效地提高油膜参数的预测精度。
Gasoline engine oil film parameters have multidimensional nonlinear properties. Currently, it is difficult to accurately determine the parameter values by the laboratory calibration method and identification method. This paper proposed a chaotic sequence of least squares support vector machine(LS-SVM) forecasting model. It was known that the gasoline engine oil circuit system had nonlinear chaotic time series characteristics. First of all, the phase space was reconstructed for the calibration data of oil film parameters. And then the support vector machine was used to train and forecast the reconstructed data and theprediction results were obtained. After that, the paper compared the prediction with the results of BP and Elman neural network models. Finally, a conclusion is drawn that the LS-SVM model has a stronger ability of nonlinear prediction and can efficiently improve the prediction accuracy of the film parameters.
Gasoline engine oil film parameters have multidimensional nonlinear properties. Currently, it is difficult to accurately determine the parameter values by the laboratory calibration method and identification method. This paper proposed a chaotic sequence of least squares support vector machine(LS-SVM) forecasting model. It was known that the gasoline engine oil circuit system had nonlinear chaotic time series characteristics. First of all, the phase space was reconstructed for the calibration data of oil film parameters. And then the support vector machine was used to train and forecast the reconstructed data and theprediction results were obtained. After that, the paper compared the prediction with the results of BP and Elman neural network models. Finally, a conclusion is drawn that the LS-SVM model has a stronger ability of nonlinear prediction and can efficiently improve the prediction accuracy of the film parameters.
引文
[1]朱航,王绍銧.电控汽油机进气道油膜特性参数的标定[J].汽车工程,2004,26(2):127-130.Zhu Hang,Wang Shaoguang.Calibration for Characteristic Parameters of Fuel Film in Intake Duct of EFI Gasoline Engine[J].Automotive Engineering,2004,26 (2):127-130.(in Chinese)
[2]李顶根,舒咏强.汽油机进气道油膜模型参数辨识算法的研究[J].内燃机学报,2009,4(27):363-369.Li Dinggen,Shu Yongqiang.Research for Identification Algorithm of Oil Film Model Parameter in Gasoline Engine Intake Duct[J].Transactions of CSICE,2009,4(27):363 -369.(in Chinese)
[3]钟祥麟.基于油膜模型的多点喷射汽油机瞬态工况控制研究[D].长春:吉林大学,2007.Zhong Xianglin.Study on Control of MPI Gasoline in Transient Condition Based on Fuel Film Model[D].Changchun:Jilin University,2007.(in Chinese)
[4]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-43.Zhang Xuegong.About the Statistical Learning Theory and Support Vector Machine(SVM)[J].Acta Automatica Sinica,2000,26(1):32-43.(in Chinese)
[5]TAKENS F.On the Numerical Determination of the Dimension Of An Attractor[C]//RAND D,YOUNG L S.Dynamical Systems and Turbulence,Warwick,1980,Lecture Notes in Mathematics,Spring-Verlag,1981(898):366-381.
[6]GRASSBERGER P,PROCACCIA I.Measuring the Strangeness of Strange Attractors[J].Theory of Chaotic Attractors,2004,9(1-2):189-208.
[7]吕金虎,占勇,陆君安.电力系统短期负荷预测的非线性混沌改进模型[J].中国电机工程学报,2000,20(12):80-83.LüJinhu,Zhan Yong,Lu Jun'an.The Nonlinear Chaotic Improved Model of Power System Short-Term Load Forecasting[J].Proceedings of the CSEE,2000,20(12):80 -83.(in Chinese)
[8]徐东辉,李岳林,解福泉,等.基于混沌RBF神经网络的汽油机瞬态工况油膜参数辨识研究[J].内燃机工程,2015,36(3):100-105.Xu Donghui,Li Yuelin,Xie Fuquan,et al.Study of Oil Film Parameters Identification in Gasoline Engine Transient Conditions Based on Chaos-RBF Neural Network[J].Chinese Internal Combustion Engine Engineering,2015,36(3):100-105.
[9]HENDRICKS E,VESTERHOLM T,KAIDANTZIS P,et al.Nonlinear Transient Fuel Film Compensation(NTFC)[C]//SAE Paper 930767,1993.
[10]HENDRICKS E,SORENSON S C.SI Engine Controls and Mean Value Engine Modeling[C]//SAE Paper910258,1991.
[11]杨斌,王绍銧.瞬态空燃比控制策略研究及智能控制的应用[J].内燃机工程,2002,23(5):10-14.Yang Bin,Wang Shaoguang.The Transient Air-Fuel Ratio Control Strategy Research and the Application of Intelligent Control[J].Chinese Internal Combustion Engine Engineering,2002,23(5):10-14.(in Chinese)
[12]吴义虎,宫焕春,欧林立,等.车用汽油机过渡工况进气管内油膜动态特性建模及仿真[J].长沙理工大学学报(自然科学版),2008,5(2):60-65.Wu Yihu,Gong Huanchun,Ou Linli,et al.Modeling and Simulating the Fuel Film Dynamic Characteristic in Intake Pipe of Automobile Gasoline Engine Under Transient Condition[J].Changsha University of Science and Technology(Natural Science),2008,5(2):60-65.(in Chinese)
[13]关新平,范正平,陈彩莲,等.混沌控制及其在保密通信中的应用[M].北京:国防工业出版社,2002:10-16.Guan Xinping,Fan Zhengping,Chen Cailian,et al.Chaos Control and Its Application in Secret Communication[M].Beijing:National Defence Industry Press,2002:10 -16.(in Chinese)
[14]何兴.基于Moto Tron平台的汽油机瞬态油膜补偿器研究[D].长沙:长沙理工大学,2013.He Xing.Study on Transient Film Compensator for Gasoline Engine Based on Moto Tron Development Platform[D].Changsha:Changsha University of Science and Technology,2013.(in Chinese)
[2]李顶根,舒咏强.汽油机进气道油膜模型参数辨识算法的研究[J].内燃机学报,2009,4(27):363-369.Li Dinggen,Shu Yongqiang.Research for Identification Algorithm of Oil Film Model Parameter in Gasoline Engine Intake Duct[J].Transactions of CSICE,2009,4(27):363 -369.(in Chinese)
[3]钟祥麟.基于油膜模型的多点喷射汽油机瞬态工况控制研究[D].长春:吉林大学,2007.Zhong Xianglin.Study on Control of MPI Gasoline in Transient Condition Based on Fuel Film Model[D].Changchun:Jilin University,2007.(in Chinese)
[4]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-43.Zhang Xuegong.About the Statistical Learning Theory and Support Vector Machine(SVM)[J].Acta Automatica Sinica,2000,26(1):32-43.(in Chinese)
[5]TAKENS F.On the Numerical Determination of the Dimension Of An Attractor[C]//RAND D,YOUNG L S.Dynamical Systems and Turbulence,Warwick,1980,Lecture Notes in Mathematics,Spring-Verlag,1981(898):366-381.
[6]GRASSBERGER P,PROCACCIA I.Measuring the Strangeness of Strange Attractors[J].Theory of Chaotic Attractors,2004,9(1-2):189-208.
[7]吕金虎,占勇,陆君安.电力系统短期负荷预测的非线性混沌改进模型[J].中国电机工程学报,2000,20(12):80-83.LüJinhu,Zhan Yong,Lu Jun'an.The Nonlinear Chaotic Improved Model of Power System Short-Term Load Forecasting[J].Proceedings of the CSEE,2000,20(12):80 -83.(in Chinese)
[8]徐东辉,李岳林,解福泉,等.基于混沌RBF神经网络的汽油机瞬态工况油膜参数辨识研究[J].内燃机工程,2015,36(3):100-105.Xu Donghui,Li Yuelin,Xie Fuquan,et al.Study of Oil Film Parameters Identification in Gasoline Engine Transient Conditions Based on Chaos-RBF Neural Network[J].Chinese Internal Combustion Engine Engineering,2015,36(3):100-105.
[9]HENDRICKS E,VESTERHOLM T,KAIDANTZIS P,et al.Nonlinear Transient Fuel Film Compensation(NTFC)[C]//SAE Paper 930767,1993.
[10]HENDRICKS E,SORENSON S C.SI Engine Controls and Mean Value Engine Modeling[C]//SAE Paper910258,1991.
[11]杨斌,王绍銧.瞬态空燃比控制策略研究及智能控制的应用[J].内燃机工程,2002,23(5):10-14.Yang Bin,Wang Shaoguang.The Transient Air-Fuel Ratio Control Strategy Research and the Application of Intelligent Control[J].Chinese Internal Combustion Engine Engineering,2002,23(5):10-14.(in Chinese)
[12]吴义虎,宫焕春,欧林立,等.车用汽油机过渡工况进气管内油膜动态特性建模及仿真[J].长沙理工大学学报(自然科学版),2008,5(2):60-65.Wu Yihu,Gong Huanchun,Ou Linli,et al.Modeling and Simulating the Fuel Film Dynamic Characteristic in Intake Pipe of Automobile Gasoline Engine Under Transient Condition[J].Changsha University of Science and Technology(Natural Science),2008,5(2):60-65.(in Chinese)
[13]关新平,范正平,陈彩莲,等.混沌控制及其在保密通信中的应用[M].北京:国防工业出版社,2002:10-16.Guan Xinping,Fan Zhengping,Chen Cailian,et al.Chaos Control and Its Application in Secret Communication[M].Beijing:National Defence Industry Press,2002:10 -16.(in Chinese)
[14]何兴.基于Moto Tron平台的汽油机瞬态油膜补偿器研究[D].长沙:长沙理工大学,2013.He Xing.Study on Transient Film Compensator for Gasoline Engine Based on Moto Tron Development Platform[D].Changsha:Changsha University of Science and Technology,2013.(in Chinese)