电梯交通流量特性分析及预测方法研究
|
摘要
在高层和中高层建筑中，为了提高电梯运行的效率，降低运行能耗，电梯群控系统的调度问题已经引起极大重视。提高电梯运行效率的关键因素之一是事先获取详细的电梯交通流量信息。电梯交通流作为电梯群控系统的调度对象，对群控系统的节能降耗起决定性的影响，电梯交通流研究的主要内容是特性分析和流量预测。本文全面、系统地对三种流量特性进行了详细的定性分析和定量描述，同时从三种流量特性出发分别建立了相应特性的流量预测模型，为电梯群控调度提供了理论依据。
主要研究内容及创新点是：
（1）首先对电梯交通流的典型特性即周期性和随机性，进行了定性分析和定量描述。其次，重点研究电梯交通流中存在混沌性。通过重构电梯交通流量序列的吸引子进行定性分析；计算数据序列的关联维数和最大李亚谱诺夫指数进行定量分析。结果表明：上、下高峰模式下多种交通流时间序列都呈现出了相似的低维混沌行为特征，即存在混沌现象。本结论将有助于电梯群控系统根据流量混沌特性进行流量预测或调整群控调度策略，从而提高各项性能指标。
（2）对办公大楼、教学楼及住宅楼等进行电梯交通流量分析表明，该类大楼内的交通流量具有明显的周期性，并且交通流量时间序列中存在线性和非线性特征数据。提出采用ARIMA模型与GP模型相结合的混合预测方法实现了电梯交通流量的混合预测。仿真结果表明，文中所提混合预测模型较其它单一预测模型都具有较好的预测精度。
（3）对医院就诊大楼、机场大楼以及宾馆等建筑物的电梯交通流量分析表明，这类大楼的交通流量具有明显的随机性。提出了采用模糊加权马尔可夫链模型对电梯交通流量交通状态进行预测。预测结果与实际电梯交通进行比较，结果表明所提方法的有效性和可行性。
（4）对于不具有周期性且随机性又不明显的不规则流量，已分析其具有的混沌特性。针对具有混沌特性的交通流量，提出了基于支持向量机的电梯交通流量混沌时间序列预测方法，并采用PSO寻优的方法确定了模型的最优参数。分析结果表明，基于SVM的预测模型，相比于其他两种单一的线性模型及非线性模型，更适合于电梯交通流这样的小样本预测，预测精度较高。
Group elevator scheduling is important to transportation efficiency for mid-riseand high-rise buildings, and how to improve the service efficiency of elevators hasreceived considerable attention. One important trend to improve elevator systems is tocollect advance traffic information. In this paper, the three characteristics have beenanalyzed and described systemly, and at the same time the three different predictmodels have been founded. It can provide the thoeriotical basis for the elvator groupcontrol system.
(1) At first, we have discussed the periodic, stochastic characteristics carefully.Second, The phase space is reconstructed to get the chaotic attractor of the flow. Wethen calculate the correlation dimension and the Largest Lyapunov Exponent aimingat the above data. Based on these analysis and calculation, these results indicate thatlow dimensional chaotic characteristics obviously exist in the up-peak and down-peaktraffic flow data. The result will help to adjust the group control scheduling methodsaccording to the chaotic behaviour of the peak flow so as to save the building energyconsuming.
(2) It is shown that there are periodic flow in the offce-building, dormitory, hoteland so on. So, we proposed the autoregressive integrated moving average (ARIMA)model has been one of the most widely used linear models in time series forecasting.Then combining genetic programming (GP), a hybrid forecasting model will be usedfor elevator traffic flow time series which can improve the accuracy. At last,simulations are adopted to demonstrate the advantages of the proposed ARIMA-GPforecasting model than any of the GP and the ARIMA forecasting models separately.
(3) There are the randomness in the traffic flow of the hospital building, airportbuilding and hotel. So a new model is proposed to predict the elevator traffic flowstate called fuzzy-weighed Markov model. And the final prediction model canreazlized. The simulation shows the effective of the proposed method.
(4) The traffic flow has the chaotic characteristic based on irregular feactures,which has not obvious periodic characteristic or randomness. Aiming at this chaotictraffic flow, the predict model by support vector machine is presented. We havediscussed the influence of the parameters to the predict result, then the parameters canbe optimized by PSO method. Finally, simulations results have shown theeffectiveness and advantages of the proposed SVM forecasting model to predict thesmall samples.
主要研究内容及创新点是：
（1）首先对电梯交通流的典型特性即周期性和随机性，进行了定性分析和定量描述。其次，重点研究电梯交通流中存在混沌性。通过重构电梯交通流量序列的吸引子进行定性分析；计算数据序列的关联维数和最大李亚谱诺夫指数进行定量分析。结果表明：上、下高峰模式下多种交通流时间序列都呈现出了相似的低维混沌行为特征，即存在混沌现象。本结论将有助于电梯群控系统根据流量混沌特性进行流量预测或调整群控调度策略，从而提高各项性能指标。
（2）对办公大楼、教学楼及住宅楼等进行电梯交通流量分析表明，该类大楼内的交通流量具有明显的周期性，并且交通流量时间序列中存在线性和非线性特征数据。提出采用ARIMA模型与GP模型相结合的混合预测方法实现了电梯交通流量的混合预测。仿真结果表明，文中所提混合预测模型较其它单一预测模型都具有较好的预测精度。
（3）对医院就诊大楼、机场大楼以及宾馆等建筑物的电梯交通流量分析表明，这类大楼的交通流量具有明显的随机性。提出了采用模糊加权马尔可夫链模型对电梯交通流量交通状态进行预测。预测结果与实际电梯交通进行比较，结果表明所提方法的有效性和可行性。
（4）对于不具有周期性且随机性又不明显的不规则流量，已分析其具有的混沌特性。针对具有混沌特性的交通流量，提出了基于支持向量机的电梯交通流量混沌时间序列预测方法，并采用PSO寻优的方法确定了模型的最优参数。分析结果表明，基于SVM的预测模型，相比于其他两种单一的线性模型及非线性模型，更适合于电梯交通流这样的小样本预测，预测精度较高。
Group elevator scheduling is important to transportation efficiency for mid-riseand high-rise buildings, and how to improve the service efficiency of elevators hasreceived considerable attention. One important trend to improve elevator systems is tocollect advance traffic information. In this paper, the three characteristics have beenanalyzed and described systemly, and at the same time the three different predictmodels have been founded. It can provide the thoeriotical basis for the elvator groupcontrol system.
(1) At first, we have discussed the periodic, stochastic characteristics carefully.Second, The phase space is reconstructed to get the chaotic attractor of the flow. Wethen calculate the correlation dimension and the Largest Lyapunov Exponent aimingat the above data. Based on these analysis and calculation, these results indicate thatlow dimensional chaotic characteristics obviously exist in the up-peak and down-peaktraffic flow data. The result will help to adjust the group control scheduling methodsaccording to the chaotic behaviour of the peak flow so as to save the building energyconsuming.
(2) It is shown that there are periodic flow in the offce-building, dormitory, hoteland so on. So, we proposed the autoregressive integrated moving average (ARIMA)model has been one of the most widely used linear models in time series forecasting.Then combining genetic programming (GP), a hybrid forecasting model will be usedfor elevator traffic flow time series which can improve the accuracy. At last,simulations are adopted to demonstrate the advantages of the proposed ARIMA-GPforecasting model than any of the GP and the ARIMA forecasting models separately.
(3) There are the randomness in the traffic flow of the hospital building, airportbuilding and hotel. So a new model is proposed to predict the elevator traffic flowstate called fuzzy-weighed Markov model. And the final prediction model canreazlized. The simulation shows the effective of the proposed method.
(4) The traffic flow has the chaotic characteristic based on irregular feactures,which has not obvious periodic characteristic or randomness. Aiming at this chaotictraffic flow, the predict model by support vector machine is presented. We havediscussed the influence of the parameters to the predict result, then the parameters canbe optimized by PSO method. Finally, simulations results have shown theeffectiveness and advantages of the proposed SVM forecasting model to predict thesmall samples.
引文
[1] Elevater,2012[Online]. Available: www.elevate.peters-research.com PetersResearch, Ltd..
[2] Jin Sun,Qian-Chuan Zhao,Peter B. Luh, Optimization of Group ElevatorScheduling With Advance Information, IEEE TRANSACTIONS ONAUTOMATION SCIENCE AND ENGINEERING,2010,7(2):352-363.
[3]万忠培,朱武标,何新民.关于电梯能耗的探讨.中国电梯,2007,18(7):28-31.
[4] EMSD: Code of practice for energy efficiency of lift and escalator installations.Hong Kong:EMSD,2007.
[5] EMSD: Guidelines on energy efficiency of lift and escalator installations.Hong Kong: EMSD,2007.
[6] G. C. Barney, S. M. Dos Santos. Elevator Traffic Analysis, Design andControl.Peter Peregrinus,1985.
[7]应晨耕,基于垂直交通客流分析的电梯群控优化研究,浙江工业大学硕士学位论文.2011.
[8]万健如,刘英培,沈虹等.电梯技术进步与节能方式.中国电梯,2008,19(16):42-43.
[9]孙立新.关于电梯能效评价的探讨.中国电梯,2008,19(4):43-45.
[10] G. F. Newell，Strategies for Serving Peak Elevator Traffic. Transpn Res.1998,32(8):583-588.
[11] G.C.Barney，S.M.Dos，Elevator Traffic Analysis，Design and Control. IEE，Peter Peregrinus，London，1985.
[12]徐国圣,朱昌明.一种电梯客流交通分布的求法.应用基础与工程科学学报.2004,12(1):86-92.
[13]商安娜.基于SARIMA的电梯交通流递归预测方法的研究.天津大学硕士学位论文，天津，2007.6.
[14]杨祯山,邵诚.基于双层规划模型的电梯交通系统群组优化.沈阳建筑大学学报.2007,23(1):170-176.
[15]王遵彤,孙栋，乔非，吴启迪.分布式电梯群控系统多目标调度算法.控制理论与应用.2010,27(5):602-608.
[16] Barney,G.C.,"Traffic design". In: Barney, G.C., Elevator Technology. EllisHorwood, Chi Chester,1986:3-21.
[17]宗群，蔡昱，虚拟环境下的电梯配置，制造业自动化，2001,23(1):28-31.
[18] Sydney C.K. Chu, C.K.Y. Lin B.S.S. Lam c,"Hospital lift system simulator: Aperformance evaluator evaluator-predictor", European Journal of OperationalResearch,2003,146:156-180.
[19]王振世，基于马尔可夫排队论的电梯高峰期群控方法的研究:[硕士学位论文]，天津:天津大学，2002.1
[20] F. Luo, Y. G. Xu, J. Z. Cao. Elevator Traffic Flow Prediction with Least SquaresSupport Vector Machines. International Conference on Machine Learning andCybernetics. Guangzhou,2005:4266-4270.
[21]杨广全,朱昌明,张亚刚等基于电梯交通实测数据的乘客O-D矩阵推算.上海交通大学学报.2005,39(6):849-852.
[22]王遵彤，乔非等，基于MAS技术的电梯群控系统建模及agent协商机制与梯群调度算法，2007,10(10):1184-1188.
[23]丁洁,林建素,刘忠.基于网络排队模型的Monte Carlo多线程电梯交通流优化设计.计算机应用.2008,28:396-398.
[24] Yutae Lee, Tai Suk Kim, Ho-Shin Cho. Performance Analysis of An ElevatorSystem During Up-peak, Mathematical and Computer Modelling,2009,49:423-431.
[25] Thorsten Poschel, Jason A. C. Gallas. Synchronization e ects in the dynamicalbehavior of elevators. Phys. Rev. E.,1994,50:2654–2659.
[26] Nagatani, T. Dynamical Behavior in the Nonlinear-map Model of an Elevator.Physica a-Statistical Mechanics and Its Applications,2002,310,(1-2):67-77.
[27] Nagatani Takashi. Complex Behavior of Elevators in Peak Traffic, Physicaa-Statistical Mechanics and Its Applications,2003,326(3-4):556-566,.
[28] Nagatani T. Dynamical Transitions in Peak Elevator Traffic,Physica a-StatisticalMechanics and Its Applications,2004,333:441-452.
[29] Takashi Nagatani, Complex Motion in Nonlinear-map Model of Eevators inEnergy-saving Traffic, Physics Letters A,2011,375:2047-2050,.
[30] Takashi Nagatani, Effect of periodic inflow on elevator traffic,Physica A，2012,391:4397-4405.
[31] Marja-Liisa Siikonen,Elevator Trafic Simulation,Simulation,October1993:257-267.
[32] Thangavelu,Kameli. Predictor elevator for traffic during peak conditions. UnitedStates Patent5276295,1994.
[33] M. L. Siikonen. Customer Service in an Elevator System duringUp-peak.Transpn Res.1997,31(2):127-139.
[34]郑延军,张惠侨,朱昌明等，电梯群控系统客流分析与仿真，计算机工程与应用，2001:139-141.
[35]宗群,宋远军,蔡昱.基于电梯交通流概率仿真模型的模拟交通流发生器.制造业自动化,2002,(10):28-32.
[36]徐国圣,朱昌明.一种电梯客流交通分布的求法.应用基础与工程科学学报.2004,12(1):86-92
[37]杨广全,朱昌明,张亚刚等,基于电梯交通实测数据的乘客O-D矩阵推算.上海交通大学学报.2005,39(6):849-852.
[38]宗群,岳有军.电梯群控系统交通流的预测方法.系统工程与电子技术,2001,23(7):103-105.
[39] Zhonghua Li, Shaobai Chen,Rirong Zheng, A Novel Approach to ClusteringAnalysis Based on Support Vector Machine, Proceedings of the2004International Symposium on Neural Network,2004, Part1:565-571.
[40]黄敏,崔宝同,顾树生.基于小波神经网络的电梯交通流预测.控制与决策.2006,21(5):589-592.
[41]靳小威,万健如.基于小波网络电梯交通流预测方法.中国电工技术学会电力电子学会第十一届年会.2006:325-329.
[42]三菱电气AI2200电梯监控系统介绍,2007. In: http://www.mitsubishi-elevator. com/innovations/control_system_b.html#pagetop.
[43]杨祯山,邵诚.基于双层规划模型的电梯交通系统群组优化.沈阳建筑大学学报.2007,23(1):170~176.
[44] Q. Zong, L. J. Wei, Y. J. Cheng. Elevator Configuration Based on theMarkovNetwork Queuing Model. Proceedings of World Congress onIntelligentControl and Automation. Hangzhou,2004:236-239
[45] Q. Zong, W. J. Wang, A. N. Shang. A Multi-mode Prediction Method forElevator Traffic Flow Based on Classification Off-line. Proceedings of ChineseControl Conference. Kunming,2008,522~526.
[46] Zong Qun, Zhao Zhan Shan, Shang An-Na. Recursive Forecasting Method forElevator Traffic Flow Based on SARIMA.Tianjin DaxueXuebao,2008,41(6):653-659.
[47] George E.P.Box, Gwilym M.Jenkins, Gregory C.Reinsel,时间序列分析,预测与控制，中国统计出版,2001.
[48] Hansen, J., Nelson, R."Time-series analysis with neural networks andARIMA-neural network hybrids"， Journal of Experimental and TheoreticalArtificial Intelligence,2003,15(3),315-330.
[49]陈淑燕,王炜.交通量的灰色神经网络预测方法.东南大学学报(自然科学版).2004,34(4):540-544.
[50] J.M.Górriz Genetic algorithm in parallel Neural Network..In proceeding ofWorld Automation Congress,2004.(17):7-12.
[51] Wei-Chiang Hong, Yucheng Dong, Feifeng Zheng,Forecasting urban traffic flowby SVR with continuous ACO, Applied MathematicalModelling,2011:1282-1291.
[52] Wei-Chiang Hong, Yucheng Dong, Feifeng Zheng,Hybrid evolutionaryalgorithms in a SVR traffic flow forecasting model, Applied Mathematics andComputation,2011,217:6733–6747.
[53] Xiping Wang; Ming Meng, A Hybrid Neural Network and ARIMA Model forEnergy Consumption Forecasting, Journal of Computers,2012,7(5):1184-1100.
[54] Yi-Shian Lee, Lee-Ing Tong, Forecasting time series using a methodology basedon autoregressive integrated moving average and genetic programming,Knowledge-Based Systems,2011,24:66–72.
[55] Clark, S. D., Dougherty, M. S., and Kirby, H. R., Statistical Models forPredicting Short-Term Changes in Traffic Flows: A Comparison with NeuralNetworks, ITS Technical Note317, Institute for Transport Studies, University ofLeeds.1992.
[56] Clark, S. D., Chen, H., and Grant-Muller, S., Artificial Neural Network andStatistical Modeling of Traffic Flows–the Best of Both Worlds, Paper No.630,presented at the8th World Congress on Transport Research, July12-17.1998.
[57] Davis,Gary A.，Nihan, Nancy L.,“Nonparametric Regression and Short-TermFreeway Traffic Forecasting”, Journal of Transportation Engineering,1991,117(2):178-188.
[58] Smith, Brian L.， Demetsky, Michael J.,“Multiple-Interval Freeway TrafficFlow Forecasting”, Transportation Research Record,1996,1554:136-141.
[59] Smith, B. L., Williams, B. M., and Oswald, R. K.,“Comparison of Parametricand Non-parametric Models for Traffic Flow Forecasting”, TransportationResearch Part C,2002,(10):303-321.
[60] Sun, Hongyu, Liu, Henry X., Xiao, Heng, He, Rachel R., and Ran, Bin,Short-term traffic forecasting using the local linear regression model,Transportation Research Record,2003,1836:143-150.
[61] Sun, Hongyu, Zhang, Chunming, Ran, Bin, Prediction interval for traffic timeseries,83rd TRB CD Washington DC,2004.
[62] Sudheer,C,Anand,N,Panigrahi,BK,Stream flow forecasting by SVM withquantum behaved particle swarm optimization. NEUROCOMPUTING,2013,101:18-23.
[63]朱昌明,洪致育,张惠桥.电梯与自动扶梯-原理结构安装测试,上海:上海交通大学出版社,2000.
[64]孙立新.关于电梯能效评价的探讨.中国电梯,2008,19(4):43-45.
[65] Sun Hongyu, Liu Henry X, Xiao HengHe, Short-term traffic forecasting usingthe local linear regression model,Transportation Research Record,2003,(1836):143-150.
[66] Barney G C. Discrete event simulation event on lift traffic control systems.Elevator，2005：68-73.
[67]罗斯,应用随机过程,人民邮电出版社，2007,7.
[68](美)Gregory F. Lawler,随机过程导论,机械工业出版社,2010,9.
[69] Tie Liu, Application of Markov Chains to Analyze and Predict the Time Series,Modern Applied Science,2010,4(5):162-166.
[70] Nagatani T, Nagai R. Chaotic and periodic motions of two competing vehiclescontrolled by traffic lights[J]. Chaos Solitons&Fractals,2005,25:245-253.
[71] Nagatani T.Chaos control and schedule of shuttle buses. Physica a-StatisticalMechanics and Its Applications,2006,.371(2):683-691.
[72] Nagatani J H C, H H C Iu, et al. Analysis and control of bifurcation and chaosin average queue length in TCP/RED model. International Journal ofBifurcation and Chaos2008,18(8):2449-2459.
[73] Bigdeli, N. and M. Haeri.Time-series analysis of TCP/RED computer networks,an empirical study. Chaos Solitons&Fractals.2009,39(2):784-800
[74]吕金虎,陆君安，陈士华.混沌时间序列分析及其应用.武汉:武汉大学出版社，2002.
[75] L1T Y,Yorke J A. Period three imPlies ehaos. American Mathematica Monthly，1975,82(10):985-992.
[76] Devaney R. An introduction to chaotic dynamical systems. NewYork，Addivon-Weslay，1989.
[77]黄润生,黄浩.混沌及其应用.武汉:武汉大学出版社.2005
[78] Grassberger P,Procaccial. Measuring the strangeness of strange attraetors.Physica D,1983,9:189-208.
[79]张国华，张文娟，薛鹏翔.小波分析与应用基础.西安:西北工业大学出版社，2006.
[80] Packard N H，Crutchifield J P，Farme J D，Geometry from a time series. Phy RevLett.,1980,45(6):712-716.
[81] Takens F, Mane. Detecting strange attractors in fluid turbulence. In:Rand D A，Young L S, Dynamical Systems and Turbulence.Vol，898of Lecture Notes inMathematics. Berlin:Springer，1986，366.
[82] Broomherd D S.Extracting qualitare dynamics from expetimentai data. PhysicaD,1987,20(11):217-236.
[83]关新平,范正平,陈彩莲.混沌控制及其在保密通信中的应用.北京:国防工业出版社.2002.
[84] Fraser A M. Information and entropy instrange atrractors. IEEE Transactions onInformation Theory,1989,35(2):245-262.
[85] Kennel M，Brown R，Abarbanel H. Determining embedding dimension forPhase-space reconstruction using a geometrical construction. Physics ReviewA，1992,45:3403-3411.
[86]马军海.复杂非线性系统的相空间重构技术，天津：天津大学出版社，2005.
[87] Schreiber T, Kantz H, Hegger R. Practical implementation of nonlinear timeseries methods: The TISEAN package [J]. Chaos,1999,9:413~435
[88] Iba，H.，Sasaki T.Using Genetic Programming to Prediet financial data. Evalutionand Computation,1999(1):251-256.
[89]鞠平,李靖霞,陆晓涛.电力负荷预测的遗传规划方法.电力系统自动化，2000,24(11):35-38.
[90] L E Mei-long, F AN G Yi. Application of Genetic Programming to Prediction ofContainers Handling Capacity. Journal of Shanghai Jiaotong University.2003,37(8):1246-1250.
[91] Matthew Streeter, Lee A, Beeker. Automated Discovery of NumeriealApproximation Formulae via Genetic Programming. Genetic Programming andEvolvable Machines,2003,(4):255-286.
[92] Kubota N，Hashimoto S， Kojima F. GP-Preprocessed fuzzy inference for theenergy load predietion.Evolutionary Computation,2000,(l):l-6.
[93] Yi-Shian Lee, Lee-Ing Tong. Forecasting time series using a methodology basedon autoregressive integrated moving average and genetic programming.Knowledge Based Systems,2011,24:66-72.
[94]时间序列的理论与方法，第二版，Brockwell, Richard A.Davis著，田铮译，2001,高等教育出版社.
[95]时间序列分析、预测与控制，第三版，(美)George E.P.Box,(英)GwilymM.Jenkins,(美)Gregory C.Reinsel著，顾岚主译，2001,中国统计出版.
[96] Gooijer De, Abaham B, Ann Gould, Methods for Deternining the order of anAutoregressive-Moving Average Process: A Survey. International StatisticalReview,1985,53(3):301-329.
[97] W. S. Chan. A comparison of some of pattern identification methods for orderdetermination of mixed ARMA models. Statistics&Probability Letters,1999,(42):69-79.
[98]张蔚，张彦琦，杨旭，时间序列资料ARIMA季节乘积模型及其应用，第三军医大学学报，2002,24(8):955-957.
[99] Guoqiang Yu, Changshui Zhang, Switching ARIMA model based forecasting fortraffic flow, International Conference on Acoustics, Speech, and SignalProcessing, May17-21,2004.
[100] Smith, B. L.Billy M. Williams, Modeling and forecasting vehicular traffic flowas a seasonal ARIMA process,, A theoretical basis and empirical results,ASCE Journal of Transportation Engineering.2003.
[101] Brockwell, P.J., Davis, R.A.,,Introduction to time series and forecasting,Springer-Verlag New York, Inc., New York.1996.
[102] A.C. Harvey, Time Series Models, John Wiley&Sons, Inc., New York,1981.
[103] H. Akaike, A new look at the statistical model identification, IEEE Transactionon Automatic Control AC-19,1974:716–723.
[104] Koza JR，Genetic Programming: on the Programming of computers by meansof natural selection. The MIT Press,Cambridge,Massaehusetts,1993:27-35
[105]崔晓玲.我国上市公司财务危机预鳌模型的研究,沈阳:东北大学，2005.
[106] G.P. Zhang, Time series forecasting using a hybrid ARIMA and neural networkmodel, Neurocomputing50,2003:159–175.
[107] C.H. Aladag, E. Egrioglu, C. Kadilar, Forecasting nonlinear time series with ahybrid methodology, Applied Mathematics Letters22(9),2009:1467–1470.
[108] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32(1),2007:254–264.
[109] James R. Broyles, Jeffery K. Cochran，A statistical Markov chainapproximation of transient hospital inpatient inventory，European Journal ofOperational Research207,2010:1645–1657.
[110]房光友.基于马尔可夫链的资产质量预测建模研究.计算机仿真，2010，27(12):354-357．
[111]刘剑锋，蒋瑞波.中国证券市场弱有效性检验.金融理论与实践，2010，4:83-87．
[112] N Bauerle，U Rieder．Portfolio optimization with Markov-modulated stockprices and interest rates[J].Automatic Control，2004，49(3):442-447．
[113] Zhang Yi,Gao Rong. Grey Markov Forecasting for Real Time Traffic Volume.Journal of nanjing normal university (Natural Science Edition).32(2),2009:41-45.
[114]陈萍,候传志,冯予,随机数学.北京:国防工业出版社,2008.
[115]应坚刚，金蒙伟,随机过程基础.上海：复旦大学出版社，2005.2:68-74.
[116]柳金莆,李学伟.应用随机过程.北京:中国铁道出版社，2000.7:75-88.
[117]林元烈.应用随机过程.北京:清华大学出版社,2002:78-113.
[118] XIA Le-tian. Prediction of plum rain intensity based on index weightedMarkov Chain[J]. Journal of Hydraulic Engineering,2005,36(8):988993.
[119]青悦红，刘嘉堑.马氏链在国际工程投标风险预测中的应用.华侨大学学报(自然科学版)，1999.10，433-436
[120]曹升乐，王艳玲.山东旱情评价预测方法研究.山东大学学报，2006，36(2):58-61.
[121]孙才志,张戈,林学钰.加权马尔可夫模型在降水丰枯状况预测中的应用.系统工程理论与实践.2003,(4):100-105.
[122] XIE Dao-wen, SHI Shi-liang,Mine water inrush prediction based on cloudmodel theory and Markov model，Journal of Central South University (Scienceand Technology),2012,43(6):2309-2315.
[123] LI De-ren, WANG Shu-liang, LI De-yi. Spatial data mining theories andapplications [M]. Beijing: Science Press,2006:337349.
[124]陈守煌.系统模糊决策理论与应用.大连理工大学出版社，1994:74-82.
[125]胡迪鹤.随机过程论.武汉:武汉大学出版社,2005:199-216.
[126]赵亚萍,张和生，周卓楠.基于最小二乘支持向量机的交通流量预测模型.北京交通大学学报.2011,35(2):114-117,136.
[127] C.L. Wu, K.W. Chau. Data-driven models for monthly streamflow time seriesprediction. Engineering Applications of Artificial Intelligence.23,2010:1350-1367.
[128] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32,2007:254-264.
[129] Z.J. Zhou, C.H. Hu, An effective hybrid approach based on grey and ARMAfor forecasting gyro drift, Chaos, Solitons and Fractals35,2008:525-529.
[130] T. Taskaya, M.C. Casey, A comparative study of autoregressive neural networkhybrids, Neural Networks18,2005:781-789.
[131] P. Chakradhara, V. Narasimhan, Forecasting exchange rate better with artificialneural network, Journal of Policy Modeling29,2007:227-236.
[132] G. Zhang, B.E. Patuwo, M.Y. Hu, Forecasting with artificial neural networks:the state of the art, International Journal of Forecasting14,1998:35-62.
[133] M. Khashei, S.R. Hejazi, M. Bijari, A new hybrid artificial neural networksand fuzzy regression model for time series forecasting, Fuzzy Sets andSystems159,2008:769-786.
[134] Victor Gomez, Agustin Maravall, Estimation, prediction, and interpolation fornonstationary series with the Kalman filter. Journal of the American statisticalassociation,1994,Vol.89,NO.426.
[135] Beatrix Izsák,The use of Kalman filter for improving the short range forecasttemperature, ALADIN Newsletter21,2002:76-79.
[136] P.F. Pai, C.S. Lin, A hybrid ARIMA and support vector machines model instock price forecasting, Omega33(2005)497-505.
[137] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32(2007)254-264.
[138] HUANG Cun-han, FENG Tao, WANG Wei-jun. Mine water inrush predictionbased on fractal and support vector machines[J]. Journal of China Coal Society,2010,35(5):806810.
[139]韩敏.混沌时间序列预测理论与方法.中国水利水电出版社,2007:28-34.
[140] Nello Cristianini,John Shawe-Taylor译者,支持向量机导论，电子工业出版社，2004.
[141]吕金虎,陆君安,陈士华.混沌时间序列分析及其应用.武汉大学出版社,2001:85-87
[142]崔建国,赵云龙,董世良.基于遗传算法和ARMA模型的航空发电机寿命预测.航空学报.2011.1(32):1506-1511.
[143] B. H. Asa, H. David, T. S. Hava and V. Vapnik. Support vector clustering.Journal of Machine Learning Research,2001(2):125-137
[144] Van der Merwe D W, Engelbrecht A P. Data Clustering Using Particle SwarmOptimization. In: The2003Congress on Evolutionary Computation. Canberra,2003:215-220.
[145] B. Scholkopf, P. Simard and A. J. Smola. Prior knowledge in support vectorkernels. Advanced in Neural Information Processing Systems. MA: MIT Press,1998,640-646.
[146] He Wang; Zhijian Hu; Mengyue Hu，Short-Term Prediction of Wind FarmPower Based on PSO-SVM, Asia-Pacific Power and Energy EngineeringConference,2012:1455-1459.
[147] Lin, LL, Weng, CC, Wang, CW, SVM and BPN models for predicting SoilErosion Degree in921Earthquake Slopeland Region in Taiwan, DISASTERADVANCES,2012,10,5(4):278-284.
[2] Jin Sun,Qian-Chuan Zhao,Peter B. Luh, Optimization of Group ElevatorScheduling With Advance Information, IEEE TRANSACTIONS ONAUTOMATION SCIENCE AND ENGINEERING,2010,7(2):352-363.
[3]万忠培,朱武标,何新民.关于电梯能耗的探讨.中国电梯,2007,18(7):28-31.
[4] EMSD: Code of practice for energy efficiency of lift and escalator installations.Hong Kong:EMSD,2007.
[5] EMSD: Guidelines on energy efficiency of lift and escalator installations.Hong Kong: EMSD,2007.
[6] G. C. Barney, S. M. Dos Santos. Elevator Traffic Analysis, Design andControl.Peter Peregrinus,1985.
[7]应晨耕,基于垂直交通客流分析的电梯群控优化研究,浙江工业大学硕士学位论文.2011.
[8]万健如,刘英培,沈虹等.电梯技术进步与节能方式.中国电梯,2008,19(16):42-43.
[9]孙立新.关于电梯能效评价的探讨.中国电梯,2008,19(4):43-45.
[10] G. F. Newell，Strategies for Serving Peak Elevator Traffic. Transpn Res.1998,32(8):583-588.
[11] G.C.Barney，S.M.Dos，Elevator Traffic Analysis，Design and Control. IEE，Peter Peregrinus，London，1985.
[12]徐国圣,朱昌明.一种电梯客流交通分布的求法.应用基础与工程科学学报.2004,12(1):86-92.
[13]商安娜.基于SARIMA的电梯交通流递归预测方法的研究.天津大学硕士学位论文，天津，2007.6.
[14]杨祯山,邵诚.基于双层规划模型的电梯交通系统群组优化.沈阳建筑大学学报.2007,23(1):170-176.
[15]王遵彤,孙栋，乔非，吴启迪.分布式电梯群控系统多目标调度算法.控制理论与应用.2010,27(5):602-608.
[16] Barney,G.C.,"Traffic design". In: Barney, G.C., Elevator Technology. EllisHorwood, Chi Chester,1986:3-21.
[17]宗群，蔡昱，虚拟环境下的电梯配置，制造业自动化，2001,23(1):28-31.
[18] Sydney C.K. Chu, C.K.Y. Lin B.S.S. Lam c,"Hospital lift system simulator: Aperformance evaluator evaluator-predictor", European Journal of OperationalResearch,2003,146:156-180.
[19]王振世，基于马尔可夫排队论的电梯高峰期群控方法的研究:[硕士学位论文]，天津:天津大学，2002.1
[20] F. Luo, Y. G. Xu, J. Z. Cao. Elevator Traffic Flow Prediction with Least SquaresSupport Vector Machines. International Conference on Machine Learning andCybernetics. Guangzhou,2005:4266-4270.
[21]杨广全,朱昌明,张亚刚等基于电梯交通实测数据的乘客O-D矩阵推算.上海交通大学学报.2005,39(6):849-852.
[22]王遵彤，乔非等，基于MAS技术的电梯群控系统建模及agent协商机制与梯群调度算法，2007,10(10):1184-1188.
[23]丁洁,林建素,刘忠.基于网络排队模型的Monte Carlo多线程电梯交通流优化设计.计算机应用.2008,28:396-398.
[24] Yutae Lee, Tai Suk Kim, Ho-Shin Cho. Performance Analysis of An ElevatorSystem During Up-peak, Mathematical and Computer Modelling,2009,49:423-431.
[25] Thorsten Poschel, Jason A. C. Gallas. Synchronization e ects in the dynamicalbehavior of elevators. Phys. Rev. E.,1994,50:2654–2659.
[26] Nagatani, T. Dynamical Behavior in the Nonlinear-map Model of an Elevator.Physica a-Statistical Mechanics and Its Applications,2002,310,(1-2):67-77.
[27] Nagatani Takashi. Complex Behavior of Elevators in Peak Traffic, Physicaa-Statistical Mechanics and Its Applications,2003,326(3-4):556-566,.
[28] Nagatani T. Dynamical Transitions in Peak Elevator Traffic,Physica a-StatisticalMechanics and Its Applications,2004,333:441-452.
[29] Takashi Nagatani, Complex Motion in Nonlinear-map Model of Eevators inEnergy-saving Traffic, Physics Letters A,2011,375:2047-2050,.
[30] Takashi Nagatani, Effect of periodic inflow on elevator traffic,Physica A，2012,391:4397-4405.
[31] Marja-Liisa Siikonen,Elevator Trafic Simulation,Simulation,October1993:257-267.
[32] Thangavelu,Kameli. Predictor elevator for traffic during peak conditions. UnitedStates Patent5276295,1994.
[33] M. L. Siikonen. Customer Service in an Elevator System duringUp-peak.Transpn Res.1997,31(2):127-139.
[34]郑延军,张惠侨,朱昌明等，电梯群控系统客流分析与仿真，计算机工程与应用，2001:139-141.
[35]宗群,宋远军,蔡昱.基于电梯交通流概率仿真模型的模拟交通流发生器.制造业自动化,2002,(10):28-32.
[36]徐国圣,朱昌明.一种电梯客流交通分布的求法.应用基础与工程科学学报.2004,12(1):86-92
[37]杨广全,朱昌明,张亚刚等,基于电梯交通实测数据的乘客O-D矩阵推算.上海交通大学学报.2005,39(6):849-852.
[38]宗群,岳有军.电梯群控系统交通流的预测方法.系统工程与电子技术,2001,23(7):103-105.
[39] Zhonghua Li, Shaobai Chen,Rirong Zheng, A Novel Approach to ClusteringAnalysis Based on Support Vector Machine, Proceedings of the2004International Symposium on Neural Network,2004, Part1:565-571.
[40]黄敏,崔宝同,顾树生.基于小波神经网络的电梯交通流预测.控制与决策.2006,21(5):589-592.
[41]靳小威,万健如.基于小波网络电梯交通流预测方法.中国电工技术学会电力电子学会第十一届年会.2006:325-329.
[42]三菱电气AI2200电梯监控系统介绍,2007. In: http://www.mitsubishi-elevator. com/innovations/control_system_b.html#pagetop.
[43]杨祯山,邵诚.基于双层规划模型的电梯交通系统群组优化.沈阳建筑大学学报.2007,23(1):170~176.
[44] Q. Zong, L. J. Wei, Y. J. Cheng. Elevator Configuration Based on theMarkovNetwork Queuing Model. Proceedings of World Congress onIntelligentControl and Automation. Hangzhou,2004:236-239
[45] Q. Zong, W. J. Wang, A. N. Shang. A Multi-mode Prediction Method forElevator Traffic Flow Based on Classification Off-line. Proceedings of ChineseControl Conference. Kunming,2008,522~526.
[46] Zong Qun, Zhao Zhan Shan, Shang An-Na. Recursive Forecasting Method forElevator Traffic Flow Based on SARIMA.Tianjin DaxueXuebao,2008,41(6):653-659.
[47] George E.P.Box, Gwilym M.Jenkins, Gregory C.Reinsel,时间序列分析,预测与控制，中国统计出版,2001.
[48] Hansen, J., Nelson, R."Time-series analysis with neural networks andARIMA-neural network hybrids"， Journal of Experimental and TheoreticalArtificial Intelligence,2003,15(3),315-330.
[49]陈淑燕,王炜.交通量的灰色神经网络预测方法.东南大学学报(自然科学版).2004,34(4):540-544.
[50] J.M.Górriz Genetic algorithm in parallel Neural Network..In proceeding ofWorld Automation Congress,2004.(17):7-12.
[51] Wei-Chiang Hong, Yucheng Dong, Feifeng Zheng,Forecasting urban traffic flowby SVR with continuous ACO, Applied MathematicalModelling,2011:1282-1291.
[52] Wei-Chiang Hong, Yucheng Dong, Feifeng Zheng,Hybrid evolutionaryalgorithms in a SVR traffic flow forecasting model, Applied Mathematics andComputation,2011,217:6733–6747.
[53] Xiping Wang; Ming Meng, A Hybrid Neural Network and ARIMA Model forEnergy Consumption Forecasting, Journal of Computers,2012,7(5):1184-1100.
[54] Yi-Shian Lee, Lee-Ing Tong, Forecasting time series using a methodology basedon autoregressive integrated moving average and genetic programming,Knowledge-Based Systems,2011,24:66–72.
[55] Clark, S. D., Dougherty, M. S., and Kirby, H. R., Statistical Models forPredicting Short-Term Changes in Traffic Flows: A Comparison with NeuralNetworks, ITS Technical Note317, Institute for Transport Studies, University ofLeeds.1992.
[56] Clark, S. D., Chen, H., and Grant-Muller, S., Artificial Neural Network andStatistical Modeling of Traffic Flows–the Best of Both Worlds, Paper No.630,presented at the8th World Congress on Transport Research, July12-17.1998.
[57] Davis,Gary A.，Nihan, Nancy L.,“Nonparametric Regression and Short-TermFreeway Traffic Forecasting”, Journal of Transportation Engineering,1991,117(2):178-188.
[58] Smith, Brian L.， Demetsky, Michael J.,“Multiple-Interval Freeway TrafficFlow Forecasting”, Transportation Research Record,1996,1554:136-141.
[59] Smith, B. L., Williams, B. M., and Oswald, R. K.,“Comparison of Parametricand Non-parametric Models for Traffic Flow Forecasting”, TransportationResearch Part C,2002,(10):303-321.
[60] Sun, Hongyu, Liu, Henry X., Xiao, Heng, He, Rachel R., and Ran, Bin,Short-term traffic forecasting using the local linear regression model,Transportation Research Record,2003,1836:143-150.
[61] Sun, Hongyu, Zhang, Chunming, Ran, Bin, Prediction interval for traffic timeseries,83rd TRB CD Washington DC,2004.
[62] Sudheer,C,Anand,N,Panigrahi,BK,Stream flow forecasting by SVM withquantum behaved particle swarm optimization. NEUROCOMPUTING,2013,101:18-23.
[63]朱昌明,洪致育,张惠桥.电梯与自动扶梯-原理结构安装测试,上海:上海交通大学出版社,2000.
[64]孙立新.关于电梯能效评价的探讨.中国电梯,2008,19(4):43-45.
[65] Sun Hongyu, Liu Henry X, Xiao HengHe, Short-term traffic forecasting usingthe local linear regression model,Transportation Research Record,2003,(1836):143-150.
[66] Barney G C. Discrete event simulation event on lift traffic control systems.Elevator，2005：68-73.
[67]罗斯,应用随机过程,人民邮电出版社，2007,7.
[68](美)Gregory F. Lawler,随机过程导论,机械工业出版社,2010,9.
[69] Tie Liu, Application of Markov Chains to Analyze and Predict the Time Series,Modern Applied Science,2010,4(5):162-166.
[70] Nagatani T, Nagai R. Chaotic and periodic motions of two competing vehiclescontrolled by traffic lights[J]. Chaos Solitons&Fractals,2005,25:245-253.
[71] Nagatani T.Chaos control and schedule of shuttle buses. Physica a-StatisticalMechanics and Its Applications,2006,.371(2):683-691.
[72] Nagatani J H C, H H C Iu, et al. Analysis and control of bifurcation and chaosin average queue length in TCP/RED model. International Journal ofBifurcation and Chaos2008,18(8):2449-2459.
[73] Bigdeli, N. and M. Haeri.Time-series analysis of TCP/RED computer networks,an empirical study. Chaos Solitons&Fractals.2009,39(2):784-800
[74]吕金虎,陆君安，陈士华.混沌时间序列分析及其应用.武汉:武汉大学出版社，2002.
[75] L1T Y,Yorke J A. Period three imPlies ehaos. American Mathematica Monthly，1975,82(10):985-992.
[76] Devaney R. An introduction to chaotic dynamical systems. NewYork，Addivon-Weslay，1989.
[77]黄润生,黄浩.混沌及其应用.武汉:武汉大学出版社.2005
[78] Grassberger P,Procaccial. Measuring the strangeness of strange attraetors.Physica D,1983,9:189-208.
[79]张国华，张文娟，薛鹏翔.小波分析与应用基础.西安:西北工业大学出版社，2006.
[80] Packard N H，Crutchifield J P，Farme J D，Geometry from a time series. Phy RevLett.,1980,45(6):712-716.
[81] Takens F, Mane. Detecting strange attractors in fluid turbulence. In:Rand D A，Young L S, Dynamical Systems and Turbulence.Vol，898of Lecture Notes inMathematics. Berlin:Springer，1986，366.
[82] Broomherd D S.Extracting qualitare dynamics from expetimentai data. PhysicaD,1987,20(11):217-236.
[83]关新平,范正平,陈彩莲.混沌控制及其在保密通信中的应用.北京:国防工业出版社.2002.
[84] Fraser A M. Information and entropy instrange atrractors. IEEE Transactions onInformation Theory,1989,35(2):245-262.
[85] Kennel M，Brown R，Abarbanel H. Determining embedding dimension forPhase-space reconstruction using a geometrical construction. Physics ReviewA，1992,45:3403-3411.
[86]马军海.复杂非线性系统的相空间重构技术，天津：天津大学出版社，2005.
[87] Schreiber T, Kantz H, Hegger R. Practical implementation of nonlinear timeseries methods: The TISEAN package [J]. Chaos,1999,9:413~435
[88] Iba，H.，Sasaki T.Using Genetic Programming to Prediet financial data. Evalutionand Computation,1999(1):251-256.
[89]鞠平,李靖霞,陆晓涛.电力负荷预测的遗传规划方法.电力系统自动化，2000,24(11):35-38.
[90] L E Mei-long, F AN G Yi. Application of Genetic Programming to Prediction ofContainers Handling Capacity. Journal of Shanghai Jiaotong University.2003,37(8):1246-1250.
[91] Matthew Streeter, Lee A, Beeker. Automated Discovery of NumeriealApproximation Formulae via Genetic Programming. Genetic Programming andEvolvable Machines,2003,(4):255-286.
[92] Kubota N，Hashimoto S， Kojima F. GP-Preprocessed fuzzy inference for theenergy load predietion.Evolutionary Computation,2000,(l):l-6.
[93] Yi-Shian Lee, Lee-Ing Tong. Forecasting time series using a methodology basedon autoregressive integrated moving average and genetic programming.Knowledge Based Systems,2011,24:66-72.
[94]时间序列的理论与方法，第二版，Brockwell, Richard A.Davis著，田铮译，2001,高等教育出版社.
[95]时间序列分析、预测与控制，第三版，(美)George E.P.Box,(英)GwilymM.Jenkins,(美)Gregory C.Reinsel著，顾岚主译，2001,中国统计出版.
[96] Gooijer De, Abaham B, Ann Gould, Methods for Deternining the order of anAutoregressive-Moving Average Process: A Survey. International StatisticalReview,1985,53(3):301-329.
[97] W. S. Chan. A comparison of some of pattern identification methods for orderdetermination of mixed ARMA models. Statistics&Probability Letters,1999,(42):69-79.
[98]张蔚，张彦琦，杨旭，时间序列资料ARIMA季节乘积模型及其应用，第三军医大学学报，2002,24(8):955-957.
[99] Guoqiang Yu, Changshui Zhang, Switching ARIMA model based forecasting fortraffic flow, International Conference on Acoustics, Speech, and SignalProcessing, May17-21,2004.
[100] Smith, B. L.Billy M. Williams, Modeling and forecasting vehicular traffic flowas a seasonal ARIMA process,, A theoretical basis and empirical results,ASCE Journal of Transportation Engineering.2003.
[101] Brockwell, P.J., Davis, R.A.,,Introduction to time series and forecasting,Springer-Verlag New York, Inc., New York.1996.
[102] A.C. Harvey, Time Series Models, John Wiley&Sons, Inc., New York,1981.
[103] H. Akaike, A new look at the statistical model identification, IEEE Transactionon Automatic Control AC-19,1974:716–723.
[104] Koza JR，Genetic Programming: on the Programming of computers by meansof natural selection. The MIT Press,Cambridge,Massaehusetts,1993:27-35
[105]崔晓玲.我国上市公司财务危机预鳌模型的研究,沈阳:东北大学，2005.
[106] G.P. Zhang, Time series forecasting using a hybrid ARIMA and neural networkmodel, Neurocomputing50,2003:159–175.
[107] C.H. Aladag, E. Egrioglu, C. Kadilar, Forecasting nonlinear time series with ahybrid methodology, Applied Mathematics Letters22(9),2009:1467–1470.
[108] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32(1),2007:254–264.
[109] James R. Broyles, Jeffery K. Cochran，A statistical Markov chainapproximation of transient hospital inpatient inventory，European Journal ofOperational Research207,2010:1645–1657.
[110]房光友.基于马尔可夫链的资产质量预测建模研究.计算机仿真，2010，27(12):354-357．
[111]刘剑锋，蒋瑞波.中国证券市场弱有效性检验.金融理论与实践，2010，4:83-87．
[112] N Bauerle，U Rieder．Portfolio optimization with Markov-modulated stockprices and interest rates[J].Automatic Control，2004，49(3):442-447．
[113] Zhang Yi,Gao Rong. Grey Markov Forecasting for Real Time Traffic Volume.Journal of nanjing normal university (Natural Science Edition).32(2),2009:41-45.
[114]陈萍,候传志,冯予,随机数学.北京:国防工业出版社,2008.
[115]应坚刚，金蒙伟,随机过程基础.上海：复旦大学出版社，2005.2:68-74.
[116]柳金莆,李学伟.应用随机过程.北京:中国铁道出版社，2000.7:75-88.
[117]林元烈.应用随机过程.北京:清华大学出版社,2002:78-113.
[118] XIA Le-tian. Prediction of plum rain intensity based on index weightedMarkov Chain[J]. Journal of Hydraulic Engineering,2005,36(8):988993.
[119]青悦红，刘嘉堑.马氏链在国际工程投标风险预测中的应用.华侨大学学报(自然科学版)，1999.10，433-436
[120]曹升乐，王艳玲.山东旱情评价预测方法研究.山东大学学报，2006，36(2):58-61.
[121]孙才志,张戈,林学钰.加权马尔可夫模型在降水丰枯状况预测中的应用.系统工程理论与实践.2003,(4):100-105.
[122] XIE Dao-wen, SHI Shi-liang,Mine water inrush prediction based on cloudmodel theory and Markov model，Journal of Central South University (Scienceand Technology),2012,43(6):2309-2315.
[123] LI De-ren, WANG Shu-liang, LI De-yi. Spatial data mining theories andapplications [M]. Beijing: Science Press,2006:337349.
[124]陈守煌.系统模糊决策理论与应用.大连理工大学出版社，1994:74-82.
[125]胡迪鹤.随机过程论.武汉:武汉大学出版社,2005:199-216.
[126]赵亚萍,张和生，周卓楠.基于最小二乘支持向量机的交通流量预测模型.北京交通大学学报.2011,35(2):114-117,136.
[127] C.L. Wu, K.W. Chau. Data-driven models for monthly streamflow time seriesprediction. Engineering Applications of Artificial Intelligence.23,2010:1350-1367.
[128] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32,2007:254-264.
[129] Z.J. Zhou, C.H. Hu, An effective hybrid approach based on grey and ARMAfor forecasting gyro drift, Chaos, Solitons and Fractals35,2008:525-529.
[130] T. Taskaya, M.C. Casey, A comparative study of autoregressive neural networkhybrids, Neural Networks18,2005:781-789.
[131] P. Chakradhara, V. Narasimhan, Forecasting exchange rate better with artificialneural network, Journal of Policy Modeling29,2007:227-236.
[132] G. Zhang, B.E. Patuwo, M.Y. Hu, Forecasting with artificial neural networks:the state of the art, International Journal of Forecasting14,1998:35-62.
[133] M. Khashei, S.R. Hejazi, M. Bijari, A new hybrid artificial neural networksand fuzzy regression model for time series forecasting, Fuzzy Sets andSystems159,2008:769-786.
[134] Victor Gomez, Agustin Maravall, Estimation, prediction, and interpolation fornonstationary series with the Kalman filter. Journal of the American statisticalassociation,1994,Vol.89,NO.426.
[135] Beatrix Izsák,The use of Kalman filter for improving the short range forecasttemperature, ALADIN Newsletter21,2002:76-79.
[136] P.F. Pai, C.S. Lin, A hybrid ARIMA and support vector machines model instock price forecasting, Omega33(2005)497-505.
[137] K.Y. Chen, C.H. Wang, A hybrid SARIMA and support vector machines inforecasting the production values of the machinery industry in Taiwan, ExpertSystems with Applications32(2007)254-264.
[138] HUANG Cun-han, FENG Tao, WANG Wei-jun. Mine water inrush predictionbased on fractal and support vector machines[J]. Journal of China Coal Society,2010,35(5):806810.
[139]韩敏.混沌时间序列预测理论与方法.中国水利水电出版社,2007:28-34.
[140] Nello Cristianini,John Shawe-Taylor译者,支持向量机导论，电子工业出版社，2004.
[141]吕金虎,陆君安,陈士华.混沌时间序列分析及其应用.武汉大学出版社,2001:85-87
[142]崔建国,赵云龙,董世良.基于遗传算法和ARMA模型的航空发电机寿命预测.航空学报.2011.1(32):1506-1511.
[143] B. H. Asa, H. David, T. S. Hava and V. Vapnik. Support vector clustering.Journal of Machine Learning Research,2001(2):125-137
[144] Van der Merwe D W, Engelbrecht A P. Data Clustering Using Particle SwarmOptimization. In: The2003Congress on Evolutionary Computation. Canberra,2003:215-220.
[145] B. Scholkopf, P. Simard and A. J. Smola. Prior knowledge in support vectorkernels. Advanced in Neural Information Processing Systems. MA: MIT Press,1998,640-646.
[146] He Wang; Zhijian Hu; Mengyue Hu，Short-Term Prediction of Wind FarmPower Based on PSO-SVM, Asia-Pacific Power and Energy EngineeringConference,2012:1455-1459.
[147] Lin, LL, Weng, CC, Wang, CW, SVM and BPN models for predicting SoilErosion Degree in921Earthquake Slopeland Region in Taiwan, DISASTERADVANCES,2012,10,5(4):278-284.