混沌时间序列分析方法研究及其应用
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
对于油田生产来说,要保证一个好的经济效益,就必须有一个高的、稳定的产油量。这是油田生产开发的中心任务。油田产量预测是科学管理油田和制定经济计划的依据。目前,各油田已拥有一套完善的包括单井数据在内的油田产量数据管理系统,急需将这些数据利用起来具体地指导油田的实际生产开发。
     油田开发是一个复杂的非线性动力学系统,油田产量变化受多种因素控制,导致其表现形式既有确定性又有随机性。总的来说,油田产量预测是一个多因素非线性预测问题。目前所采用的油田产量预测方法的预报相对误差都在10%左右,并且这些方法对油田开发过程的时变性和各种随机干扰因素具有不适应性。因此,有必要借助于其它的分析工具对油田产量进行预测。我们可以得到油田产量的历史数据,可将这些数据看成是时间序列并利用时间序列分析的方法对其进行建模及预测。
     本论文以油井产量时间序列为对象,对RLS算法、混沌时间序列性质鉴别方法、混沌时间序列预测方法等进行了深入细致的研究。
     论文首先分析RLS算法的性能,在此基础上为了提高收敛速度,提出一种改进的RLS算法,利用改进的RLS算法对某油田油井产量进行了建模及预测。通过对预测结果的分析发现RLS预测方法不能对油井产量进行精确地多步预测。
     其次,为了更好的了解油井产量时间序列地性质,利用相空间重构方法重构该序列的吸引子,并计算其维数和最大Lyapunov指数,指出油井产量时间序列具有混沌特征。同时,针对伪近邻算法效率低的缺点,提出一种快速伪近邻法选择嵌入维数的新方法,该方法可将原算法时间复杂度由从O(M(M-1))降低到O(3M)。
     再次,对训练支持向量机的序列最小优化方法进行了深入研究,针对原算法选取工作集过于随机的缺点,提出一种基于遗传算法选择工作集的新方法,该方法能够保证每次所选取的工作集使得目标函数变化最大,从而使目标函数尽快向极值收敛。仿真结果表明,该方法可大大加快训练速度;同时,针对以往支持向量机参数选取过于主观的缺点,提出了基于遗传算法方法选择支持向量机参数的方法,仿真结果表明,这种新的参数选择方法可在不明显增加支持向量个数的基础上减小泛化误差。
The most import task for oilfield is to ensure a high and stable output. The accuracy of prediction is the basis of oilfield management. Now, most oilfields have held a set of output data including single well data. And they urgently want to use these data for the development of oilfield.
    In fact, the process of oilfield development is a complicated nonlinear dynamics. In one word, the prediction for oilfield output is nonlinear. At present, the relative error of the existing prediction methods for oilfield output is about 10%, and these methods can not fit to stochastic noises. We have the oilfield output data, so we can use time series analysis tool for modeling and prediction.
    The research object of this thesis is the time series of some oil wells outputs. The research fields of the thesis include RLS algorithm, the methods for detection chaos in time series, and the methods for time series prediction.
    First, the characteristics of RLS algorithm are analyzed deeply, and an improved RLS algorithm is proposed. With the proposed algorithm, oil well outputs are predicted. Simulation results show that, RLS algorithm can not make long time prediction.
    Second, in order to understand the characteristics of time series of oilfield outputs more clearly, phase space reconstruction is used to reconstruct the attractor of the time series. The maximum Lyapunov exponent and the dimension of the attractor show that the time series of oil well outputs are chaotic.
    Studies show that the method of false neighbors is not effective, so a quick false
    neighbor method that can improve the time complexity from O(M(M-1)) to O(3M).
    To avoid selecting work set of sequential minimal optimization algorithm
    randomly, a new method based on GA is proposed. The work set selected by the
    proposed method can make the change of target function as large as possible.
    Simulation results show that the proposed method can speed the training process.
    And to select suitable parameters with support vector machine, a GA-based
    method for parameter selection is proposed, too. Simulation results show that this
    new method can reduce generalization error with little increase in support vectors.
    Finally, based on the identification of characteristics of oil well outputs time
引文
[1] 刘合著.油田联合站集输系统控制技术.石油工业出版社,北京,2003.5
    [2] 陈胜利,郭冬琼,张(王月).一种油田产量和可采储量的预测模型及应用.承德石油高等专科学校学报.2004,6(4):32-35页
    [3] 胡泽.注水开发油田产油量产水量动态预报的新方法.西南石油学院学报。1997,19(2):37-42页
    [4] 陈程,孙义梅.用高伯茨——马尔可夫链模型预测油田年产油量.河南石油.1995,9(3):30-35页
    [5] 胡泽,贾永禄.油田产量预报的多维时间序列神经网络模型.西南石油学院学报.2003,25(3):33-36页
    [6] 杨立钦,顾岚编著.时间序列分析与动态数据建模.北京理工大学出版社,北京:1988.12
    [7] E. R Carson, H. J. Leicester, A. V. Roudsari. Linear and non-linear time series analysis in non-invasive physiological measurement. 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Amsterdam 1996: 970-971P
    [8] 何书元 编著.应用时间序列分析.北京大学出版社, 北京:2003.9
    [9] 谢红梅,俞卞章.非线性时间序列分析的关键技术及其应用.计算机工程与应用.2002,14:79-81页
    [10] 王凤兰,闻邦椿.混沌经济时间序列的建模与系统识别。经济经纬.2004,1:15-17页
    [11] R. Pintelon, J. Schoukens. Time Series Analysis in the Frequency Domain. IEEE Trans. on Signal Processing. 1999, 47(1): 206-210页
    [12] 田铮译.时间序列的理论与方法.高等教育出版社,北京:2001.7
    [13] H. Akaike. A new look at the statistical model identification.IEEE Trans. On Automatic Control. 1974, 19(6): 176-723页
    [14] 朱洪涛,严国萍.基于自适应滤波的信号处理系统建模及其系统实现.仪器仪表学报.2003,24(4):242-246页
    [15] 王新龙,申功勋,丁杨斌.利用GPS进行车辆动态定位的自适应建模研究.控制与决策.2005,20(1):103-107页
    [16] L. D. Parramann, M. D. Najar. Adaptive online load forecasting via time series modeling. Electric Power Systems Research. 1995, 32:219-225P
    [17] R. K. Srivastava, D. S. McRae, M. T. ODman. An adaptive grid algorithm for air-quality modeling. Journal of Computational Physics. 2000,165:437-472P
    [18] H. Yang, J. Ni. Adaptive model estimation of machine-tool thermal error based on recursive dynamic modeling strategy. International Journal of Machine??Tools & Manufacture.2005,45:1-11P
    [19] 刘飞,张鹏林.基于卡尔曼滤波的时间序列影像背景估计.地理空间信息.2005,3:12-13页
    [20] S. C. Felter. An overview of decentralized Kalman filter techniques. Proceedings of the 1990 IEEEsouther Tier Technical Conference. 1990, 4: 79-87页
    [21] 陈玉,和卫星.卡尔曼滤波在心电信号QRS波检测中的尝试.计算机仿真.2004,21(12):98-100页
    [22] 张家树,消先赐.混沌时间序列的volterra自适应预测.物理学报.2000,49(3):403-407页
    [23] 张家树,消先赐.混沌时间序列的volterra自适应预测.物理学报.2000,49(3):403-407页
    [24] 吴启晖,王金龙,沈良,任国春.用于自适应天线的稳健的RLS算法.电子学报.2002,30(6):893-895页
    [25] S. Haykin. Adaptive tracking of linear time-variant systems by extended RLS algorithms. IEEE Trans on Signal Processing. 1997, 45(5)1118-1128P
    [26] 李靖,王树勋,汪飞.基于多项式变幻的自适应Chirp信号参数估计.吉林大学学报(工学版).2004,34(4):617-621页
    [27] 孙锦华,金力军.几种改进型RLS算法在自适应滤波系统中的应用.重庆邮电学院学报.2003,15(3):14-17页
    [28] 华琳.梯度可变遗忘因子RLS算法在智能天线中的应用.信息与电子工程.2004,2(1):33-36页
    [29] 张维松,常华,刘珩.一种基于RLS算法主动噪声控制系统的研究.中国农业大学学报.2004,9(3):62~64页
    [30] C. Papaodysseus, G. Roussopoulos, A. Panagopoulos. Using a fast RLS adaptive algorithm for efficient speech processing. Mathematics and Computers in Simulation. 2005, 68:105-113P
    [31] K. Eom. D. Park. Analysis of overshoot phenomena in initialization stage of RLS algorithm. Signal Processing. 1995, 44: 329-339P
    [32] N. E. Hubing, S. T. Alexander. Statistical analysis of initialization methods for RLS adaptive filters. IEEE Trans. on Signal Process. 1991, 39(8): 1793-1804P
    [33] D. J. Park, K. S. Eom. Analysis of initial convergence behaviors of the recursive least-squares algorithm via the stochastic region of attraction.Proceedings of 32nd SICE. Kanazawa, Japan. 1993, August: 1249-1252P
    [34] D. J. Park and B. E. Jun. Selfperturbing recursive least squares algorithm with fast tracking capability. Electronics Letters. 1992, 28(6): 558-559P[35] J. Jiang and R. Cook. Fast parameter tracking RLS algorithm with high noise immunity. Electronics Letters. 1996, 28(22): 2043-2045P
    [36] K. Eom and D. Park.. Fast tracking and noise-immunised RLS algorithm based on kalman filter. Electronics Letters. 1996, 32(25):2311-2312P
    [37] S. Qiao. Fast adaptive RLS algorithms: a generalized inverse approach and analysis. IEEE Trans. on Signal Processing. 1991, 39(6):1455-1459P
    [38] J. M. Cioffi, T. Kailath. Fast recursive-least-squares transversal filters for adaptive filtering. IEEE Trans.on Acoustics,Speech, and Signal Processing. 1984, 32(2):304-337P
    [39] C. F. So, S. C. Ng and S. H. Leung. Gradient based variable forgetting factor RLS algorithm. Signal Processing. 2003, 83: 1163-1175P
    [40] D. J. Park, et al. Falst tracking RLS algorithm using novel variable forgetting factor with unity zone. Electron. Lett. 1991, 27:2150-2151P
    [41] S. Song, et al. Gauss newton variable forgetting factor recursive least squares for time varying parameter tracking. Electron. Lett. 2000, 36: 988-990P
    [42] B. Chun, B. Kim, Y. H. Lee. Generalization of exponentially weighted RLS algrorithm based on a state-space model. Proceedings of the 1998 IEEE International Symposium on Circuits and Systems. Monterey, CA. 1998:198-201P
    [43] M. Djendi, A. Guessoum, A. Benallal, M. Bouchard. New combinatorial methods for the improvement of the convergence speed and the tracking ability of the fast stable RLS adaptive algorithm. Proceedings of IEEE-EURASIP First International Symposium on Control, Communications and Signal Processing. Hammamet, Tunisia. 2004:681-684P
    [44] 邓拥军,杨知行,郭兴波.一种基于RLS算法的改进自适应算法DRLS.学术论坛.2004,34(6):10-12页
    [45] G. E. Bottomley, S. T. Alexander. A theoretical basis for the divergence of conventional recursive least squares filters. Proceedings of IEEE Conference on Speech, Signal Processing. Glasgow, Scotland. 1989:908-911P
    [46] P. Fabre, C. Gueguen. Improvement of the fast recursive least squares algorithms via normalization: a comparative study. IEEE Trans. on Acoust, Speech, Signal Processing. 1986, 34: 296-308P
    [47] 熊鹰,梁树雄,尹俊勋.Improved performance analysis of recursive least squares algorithm. 华机理工大学学报.2001,29(11): 32-36页
    [48] G. E. Bottomley, S. T. Alexander. A novel approach for stabilizing recursive least squares filters. IEEE Trans. on Signal Process. 1991, 39(8): 1770-1779P
    [49] 邓自立,郭一新.油田产油量、产水量动态预报.油气开发系统专辑.石油工业出版社,1991[50] 刘永红,范跃祖.激光陀螺漂移的非平稳随机信号建模.战术导弹控制技术.2004,3:99-102页
    [51] 裴连君,王仲林.Arps递减曲线与甲型水驱曲线的相关性及参数计算.石油勘探与开发.1999,26(3):62-67页
    [52] 毕生.灰色预测方法在油田增产措施评价中的应用.大庆石油地质与开发.2003,22(4):37-40页
    [53] 范海军,陈月明.用组合模型预测油田产油量.石油大学学报(自然科学版).1998,22(3):54-55页
    [54] 赵锋,陈祥光.预测水驱油田含水率的一种联解模型.计算机仿真.2004,21(10):48-51页
    [55] 刘玉忠,尚明忠.自然递减率与综合递减率关系研究.油气地质与采收率.2002,9(4):51-52页
    [56] 洪时中.非线性时间序列分析的最新进展及其在地球科学中的应用前景.地球科学进展.1999,14(6):559-565页
    [57] 陈国华,盛昭瀚.基于Lyapunov指数的混沌时间序列识别.系统工厂理论方法应用.2003,12(4):317-320页
    [58] J. Theiler, S. Eubank. Testing for nonlinearity in time series: the method of surrogate data. Physica D. 1992, 58: 77-94P
    [59] D. T. Kaplan, L. Glass. Direct test for determinism in a time series. Phys Rev Lett, 1992, 68(4):427-430P
    [60] R. Wayland, D. Bromley, D. Pickett. Recognizing determinism in a time series. Phys Rev Lett, 1993, 70(5):580-582P
    [61] A. Potapov, J. Kurths. Correlation integral as a tool for distinguishing between dynamics and statistics in time series data. Physica D. 1998,120:369-385
    [62] 温权,张勇传,程时杰.辨识混沌时间序列中的确定性.水电以源科学.2001,19(3):72-75页
    [63] H. S. Kim, R. Eykholt, J. D. Salas. Nonlinear dynamics, delaytimes, and embedding windows. Physica D. 1999, 127: 48-60P
    [64] Y. C. Lai, D. Lerner. Effective scaling regime for computing the correlation dimension from chaotic time series. Physica D: Nonlinear Phenomena. 1995, 115(1-2): 1-18P
    [65] R. M. Kil, S. H. Park, S. Kim. Optimum window size for time series prediction. Proceedings of IEEE 19th International Conference. Chicago, USA. 1997, Oct. 30: 1421-1424P
    [66] Y. C. Lai, D. Lerner, R. Hayden. An upper bound for the proper delay time in chaotic time-series analysis. Physics Lett A. 1996, 218:30-34P
    [67] I. Matsuba, H. Suyari, S. Weon. Practical chaos time series analysis with financial applications. Proceeding of ICSP2000. 265-271P[68] 朱义东,李玉林.基于最大Lyapunov指数方法预测油田产量.西南石油学院学报.2005,27(3):32-33页
    [69] 田玉楚.混沌时间序列的时滞判定.物理学报.1997,46(3):442-447页
    [70] 肖方红,阎桂荣,韩宇航.混沌时序相空间重构参数确定的信息论方法.物理学报.2005,54(2):550-556页
    [71] Y. C. Tian, F. Gao. Extraction of delay information from chaotic time series based on information entropy. Physica D. 1997, 108:113-118P
    [72] Y. C.Tian. Delay estimation in chaotic time series. 1791-1795P
    [73] M. Otani, A. Jones. Automated embedding and creep phenomenon in chaotic time series
    [74] 马红光,李夕海.相空间重构中嵌入维和时间延迟的选择.西安交通大学学报.2004,38(4):335-338页
    [75] 王海燕,盛昭瀚.混沌时间序列相空间重构参数的选取方法.东南大学学报(自然科学版).2000,30(5):113-117页
    [76] 蒋传文,候志俭,李承军.求取混沌时间序列嵌入维数的一种神经元网络方法.水电能源科学.2000,19(3):12-14页
    [77] S. D. Broomhead, P. G. King. Extracting qualitative dynamics form experimental data. Physica D. 1986. 9: 198-208P
    [78] 袁坚,肖先赐.非线性时间序列的高阶奇异谱分解.物理学报.1998,47(6):897-905页
    [79] 孟庆芳,张强,潘金凤.四阶累计量用于最小嵌入维数估计的新方法.系统工程理论与实践.2005,9:83-88P
    [80] S. Heidari, L. C. Nikias. characterzing chaotic attractors using fourth-order off-diagonal cumulant slices. Conference Record of the 27th Asilomar conference on Signals systems and computers. Pacific Grove, USA, 1993: 466-470P
    [81] L. Cao. Practical method for determining the minimum embedding dimension of a scalar time series. Physica D. 1997, 110:43-50P
    [82] L. Cao. Practical method for determining the minimum embedding dimension of a scalar time series. Physica D. 1997, 110: 43-50P
    [83] H. N. Packard, RJ. Crutchfield, D. J. Farmer. Geometry from a time series. Phys Rev Lett. 1980, 45(3):712-716P
    [84] F. Takens. Detecting strange attractor in turbulence. Lecture Notes in Mathematics. 1981, 898(2):361-381P
    [85] W. Liebert, H. G. Schuster. Proper choice of the time delay for the analysis of chaotic time series. Phys Lett A. 1989,142:107-112P
    [86] A. M. Fraser, L. H. Swinney. Independent coordinates from mutual information. Phy Rev A. 1986, 33:1134-1140P[87] 杨光正,无岷,张晓莉编著.模式识别.合肥:中国科学技术大学出版社,2001
    [88] 胡静,杨宗凯.基于局部投影方法的混沌信号去噪.华中科技大学学报(自然科学版).2002,30(4):66-68页
    [89] 宫立新,程国平,岳毅宏.基于组合策略的Lyapunov指数谱的计算.系统工程理论与实践.2005,7:93-97页
    [90] A. M. Fraser. Information and entropy in strange attractors. IEEE Trans. Inform Theory. 1989, 35(2): 245-262P
    [91] M. J. Greene, S. J. Kim. The calculation of Lyapunov spectra. Physica D, 1987, 24: 213-225P
    [92] 杨绍清.一种最大李雅普诺夫指数估计的稳健算法.物理学报.2000,49(4):636-640页
    [93] A. Wolf, B. J. Swifi, L. H. Swinnery.Determining Lyapunov exponents from a time series. Physica D. 1985. 16: 285-317P
    [94] J. D. Framer, J. J. Sidorowich. Predicting chaotic time series, Phys. Rew. Lett. 1987, 2: 845-848P
    [95] 李恒超,张家树.混沌时间序列的局域零阶预测方法性能比较.西南交通大学学报.2004,39(3):328-331页
    [96] 温权,张勇传,程时杰.混沌时间序列预测技术.水电能源科学.2001,19(3):76-78页
    [97] R. Gencay. Nonlinear prediction of noise time series with feed-forward network. Physica D. 1995. 85: 225-238P
    [98] 蒋伟进.非线性混沌系统的神经网络预测预算法研究.南京大学学报.2002,38(11):33-36页
    [99] J. Deppisch, U. H. Bauser, T. Geisel. Hierarchical training of neural networks and prediction of chaotic time series. Phys. Lett A, 1991, 158:57-63P
    [100] L. Cao, Y. Hong, H. Fang, G. He. Predicting chaotic time series using wavelet network. Physica D. 1995, 85:225-238P
    [101] J. Zhang, K. EMan. Time series prediction using RNN in multi-dimension embedding phase space. IEEE International Conference on Systems, Man, and Cybernetics. San Diego, CA, USA. 1998: 1868-6873P
    [102] J. Zhang, K. S. Tang, K. EMan. Recurrent NN Model for chaotic time series prediction. 23rd Intemational Conference on Industrial Electronics, Control and Instrumentation. New Orleans, LA. 1997: 1108-1112P
    [103] K. Oliveira, A. Vannucci, E. Silva. Using artifical neural networks to forecast chaotic time series. Physica A. 2000, 284:393-404P
    [104] J. Kuo, J. C. Principe, B. Vries. Prediction of chaotic time series using recurrent neural networks. Proceedings of the 1992 IEEE-SP Workshop on??Neural Networks for Signal Processing. Helsingoer. 1992: 436-443P
    [105] H. D. Navone, H. A. Ceccatto. Learning chaotic dynamics by neural networks. Chaos, Solitons & Fractals. 1995, 6:383-387P
    [106] A. Aussem. Dynamical recurrent neural networks towards prediction and modeling of dynamical systems. Neurocomputing. 1999, 28: 207-232P
    [107] M. Han, J. Xi, S. Xu, F. Yin. Prediction of Chaotic time series based on the recurrent predictor neural network. IEEE Trans. on Signal Processing. 2004, 52(12): 3409-3416P
    [108] 郭双冰,肖先赐,基于Volterra 级数的神经网络实现混沌时间序列的快速预测.Proceedings of the 3rd World Congress on Intelligent Control andAutomation.Hefei,Chian,July,2000:1051-1054P
    [109] 崔甲武.混沌时间序列的非线性滤波预测模型的研究.河南大学学报(自然科学版).2004,34(2):24-27页
    [110] 张伟,吴智铭,杨根科.混沌时间序列的遗传演化建模.电子学报.2005,33(4):748-751页
    [111] 蒋伟进,孙星明,王润球.混沌序列优化预测模型研究.华北电力大学学报.2004,31(2):64-68页
    [112] 蒋伟进,许宇胜,彭召意等.混沌序列的优化预测模型及算法研究.吉林大学学报.2004,22(4):293-297页
    [113] S. A. Weigend. Paradigm change in prediction. Chaos and Forecasting. 1995, 145-160P
    [114] 刘涵,刘丁,李琦.基于支持向量机的混沌时间序列非线性预测.系统工程理论与实践.2005,(9):94-99页
    [115] 孙德山,吴今培,肖健华.基于线性规划支持向量回归的混沌系统预测.计算机工程与应用.2005,(19):35-37
    [116] 孙德山,吴今培,侯振挺,肖健华.基于SVR的混沌时间序列预测.计算机工程与应用.2004,(12):54-56页
    [117] 孙德山,吴今培,侯振挺,肖健华.单参数支持向量回归算法.系统工程学报.2005,20(1):109-112页
    [118] 孙德山,吴今培.LS-S VM在混沌时间序列预测中的应用.微机发展.2004,14(1):22-24页
    [119] 刘隽,周涛,周佩玲.GA优化支持向量机用于混沌时间序列预测.中国科学技术大学学报.2005,35(2):258-265页
    [120] M. Dominique, B. Alistair. Nonlinear blind source separation using kernels. IEEE Trans. on Neural Networks. 2003, 14(1):228-235P
    [121] B. Sholkopf, K. Sung, C. J. Burges. Comparing support vector machine with Gaussian Kernels to radial basis function classifiers. IEEE Trans. Signal Processing. 1997. 45: 2758-2765P[122] C. Cortes, V. Vapnik. Support-vector networks. Machine Leaming, 1995, 20: 273-297P
    [123] V. Vapnik, O. Chapelle. Bounds on error expectation for support vector machines. Neural Comoutation. 2000. 12(9): 2013-2036P
    [124] 董春曦,饶鲜,杨绍全,徐松涛.支持向量机参数选择方法研究.系统工程与电子技术.2004,26(8):1117-1120页
    [125] 张浩然,韩正之.回归支持向量机的改进序列最小优化学习算法.软件学报.2003,14(2):2006-2013页
    [126] J. C. Platt. Fast training of support vector machines using sequertial minimal optimization. Advan ces in Kernel Method: Support Vector achines. Cambridge: MIT Press. 1998. 185-208P
    [127] M. Martin. Online support vector machines for function approximation. Technical Report, LSI-02-11-R. Catalunya: Department of Software, Universitat Politecnica de Catalunya, 2002

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700