傳統的歷時模擬方式通常牽涉複雜的數學運算；本文嘗試以類神經網路藉由學習來模擬具不確定性的近常態歷時資料，提供另一種歷時模擬的可能方法。首先透過組合演算法與最適化網路結構建立符合近常態分佈歷時資料的複製類神經網路(RNN)，獲得歷時資料的壓縮網路（RNNl）與解壓縮網路（RNNu），並求得歷時的壓縮資料。接著透過歷時的壓縮資料分別建立倒傳遞類神經網路（BNN）與隨機類神經網路（SNN），用以模擬壓縮的歷時資料。再利用RNNl、BNN與RNNu建立歷時產生器類神經網路（TGNN1）。同樣地，RNNl、SNN與RNNu組成另一個歷時產生器類神經網路（TGNN2）。建構出的兩個歷時產生器類神經網路，分別作為近常態分佈歷時資料的歷時產生器。訓練完成的TGNN1給予一筆歷時資料多次，可產生多筆歷時資料。TGNN2訓練完成後給予一筆歷時資料，則至少可獲得與輸入向量數相同筆數的歷時資料。TGNN2能夠產生與輸入的歷時資料相同頻率內涵的歷時資料，而且產生的歷時資料在前四個統計量與單邊頻譜密度函數上均具有相當不錯的準確性。TGNN1模擬結果的精度不如TGNN2來的佳。

In the past, engineers have used many different methods for modeling time series. These methods have many kinds of complicated nonlinear mathematics model. A lot of parameters of simulation model must be established via some procedures. When neural networks generate time series, which do not need define complicated mathematical model or equation. This article applies a suit of technique and mechanism of neural networks, which provides another time series modeling method, to generate specific time series data. In this article chooses nearly normal time series simulation as the example. Show a two-stage approach. In the first stage, a trained replicator neural network is used as a data compression tool. The replicator neural network compresses the vector of the discrete time series data to vectors of much smaller dimension. In the second stage, trained stochastic neural networks or trained back propagation neural networks learns to relate the compressed time series data to another compressed time series data. Then, the replicator neural network was combined with back propagation neural network or stochastic neural network, which trained to learn to associate the time series data with another time series data, to form TGNN1 and TGNN2. That two time series generator neural networks are becoming nearly normal time series generator. Finally, this article uses TGNN1 and TGNN2 to simulate time series data and analyses results.

Ackley, D. H., Hinton, G. E. and Sejnoewski, T. J. (1985), A Learning Algorithm for Boltzmann Machines, Cognitive Science 9, 147-169.

Akaike, H. (1970), Statistical Predictor Identification, Ann. Inst. Statist. Math., 22, 203-217.

Akaike, H. (1974), A New Look at the Statistical Model Identification, IEEE Trans. Autom. Control, AC19, 716-723, Dec.

Akiyama, Y., Yamashita, A., Kajiura, M. and Aiso, H. (1989), Combinatorial Optimization with Gaussian machines, IEEE IJCNN 1, 533-540.

Ash, T. (1989), Dynamic Node Creation in Backpropagation Networks, ICS Report 8901, Institude for Cognitive Science, University of California, San Diego, La Jolla.

Burg, J. P. (1975), Maximum Entropy Spectral Analysis, Ph.D. dissertation, Stanford University, May.

Cichocki, A. and Unbehauen, R. (1993), Neural networks for optimization and signal processing, John Wiley & Sons, Chichester.

Fahlman, S. E. (1988), Faster-learning Variations on Back-propagation: An Empirical Study, Proceeding of 1988 Connectionist Models Summer School, Morgan Kaufmann Los Altos CA, 38-51.

Gelenbe, E. (1989), Random Neural Networks with Negtive and Positive Signals and Product from Solution, Neural Computation 1, 502-510.

Ghaboussi, J. (1993), An Overview of the Potential Applications of Neural Networks in Civil Engineering, Proceedings, ASCE Structures Congress ’93, Irvine, California.

Ghaboussi, J. (1994), Some Applications of Neural Networks in Structural Engineering, Proceedings, Structures Congress, ASCE, Atlanta, GA.

Ghaboussi, J., Banan, M. R. and Florom, R. L. (1994), Application of Neural Networks in Acoustic Wayside Fault Detection in Railway Engineering, Proceedings of World Congress on Railway Research, Paris, France.

Ghaboussi, J., Garret, J. H., Jr. and Wu, X. (1990), Material Modeling with Neural Networks, Proceedings of the International Conference on Numerical Methods in Engineering: Theory and Application, Swansea, U. K., 701-717.

Ghaboussi, J., Garret, J. H., Jr. and Wu, X. (1990), Knowledge-based Modeling of Material Behavior with Neural Networks, Journal of Engineering Mechanics Division, ASCE 117(1), 132-153.

Ghaboussi, J. and Lin, C. C. J. (1998), New method of Generating Spectrum Compatible Accelerograms using Neural Networks, Earthquake Engng. Struct. Dyn. 27(4), 377-396.

Grossberg, S. (1976), Adaptive Pattern Classification and Universal Recording: Part 1. Parallel Development and Coding of Natural Features Detectors, Biological Cybernetics 23, 121-134.

Hecht Nielsen, R. (1995), Replicator Neural Networks for Optimal Source Coding, Science 269, 1860-1863.

Hecht Nielsen, R. (1996), Data manifolds, natural coordinates, replicator neural networks, and optimal source coding, ICONIP-96.

Hertz, J., Krogh, A. and Palmer, R. G. (1991), Introduction to the theory of neural computation, Addison-Wesley, Redwood City, Caligornia.

Hinton, G. E. and Sejnowski, T. J. (1983), Optimal Perceptural Infernce, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 448-453.

Hinton, G. E., Sejnowski, T. J. and Ackley, D. H. (1984), Boltzmann Machines: Constraint Satisfaction Networks That Learn, Technical Report No. CMU-CS-84-110, Department of Computer Science, Carnegie, Mellon University, Pittsburg, PA.
Hinton, G. E. and Sejnowski, T. J. (1986), Learning and Relearning in Boltzman Machines, Parallel Distributed Processing, 1, Chap. 7.

Hopfield, J. J. (1982), Neural Networks and Physical Systems with Emergent Collective Computational Abilities, Proceeding of the National Academy of 79, 2554-2558.

Hornik, K., Stinchcomebe, M. and White, H. (1989), Multilayer Feedforward Networks Ara Universal Approximators, Neural Networks 2, 359-366.

Jacobs, R. A. (1988), Increased Rates of Convergence Through Learning Rate Adaptation, Neural Networks 1, 295-307.

Joghataie, A., Ghaboussi, J. and Wu, X. (1995), Learning and architecture determination through automatic node generation, International Conference on Artificial Neural Networks in Engineering, ANNIE ’95, St. Louis, Missouri.

Kesten, H. (1958), Accelerated Stochastic Approximation, Annals of Mathematical Statistics 29, 41-59.

Kohonen, T. (1974), An Adaptive Associative Memory Principle, IEEE Transactions C-23, 444-445.

Kohonen, T. (1989), Self-Organization and Associative Memory (3rd ed.), Berlin: Springer-Verlag.

Lin, C. C. J. (1999), A neural network based methodology for generating spectrum compatible earthquake accelerograms, PhD thesis, Dept. of Civil and Environmental Engineering, Univ. of Illinois at Urbana Champaign, Urbana, Illinois.

Lin, C. C. J. and Ghaboussi, J. (1997a), A New Method of Generating Artificial Earth-quake Accelerograms Using Neural Networks, ICCCBE VII, Souel, Korea.

Lin, C. C. J. and Ghaboussi, J. (1997b), Replicator Neural Network in Generating Artificial Earthquake Accelerograms, Pro., ANNIE ‘97, St. Louis, Missouri, 377-396.

Lin, C. C. J. and Ghaboussi, J. (1999), Stochastic Neural Networks in Generating Multiple Artifical Earthquake Accelerograms, Proc., ANNIE ’99, Louis, Missouri, 1061-1066.

More, Anurag and Deo, M.C.( 2002), Forecasting wind with neural networks, Marine Structures, 16(1), January/February, 2003, 35-49

Parzen, E. (1976), An Approach to Time Series Modeling and Forecasting Illustrated by Hourly Electricity Demands, Tech. Rep. 37, Statistical Science Division, State University of New York, Jan.

Plaut, D., Nowlan, S. and Hinton, G. (1986), Experiments on Learning by Back Propagation, Technical Report CMU-CS-86-126, Deparment of Computer Science, Carnegie Mellon University, Pittsburgh, PA.

Rissanen, J. (1978), Modeling by Shortest Data Description, Automatica, 14, 465-471.

Rumelhart, D. E. and Zipser, D. (1985), Feature Discovery by Competitive Learning, Cognitive Science 9, 75-112.

Rumelhart, D. E. and McClelland, J. L. (Eds), (1986), Parallel Distributed Processing: Explorations in the Microstructure of Cognition; Vol. 1: Foundations, The MIT Press,Cambrige, MA.

Rumelhart, D. E., Hinton, G. E. and McClelland, J. L., (1986a), A General Framework for Parallel Distributed Processing, Parallel Distributed Processing, Vol. 1: Foundation, The MIT Press, MA.

Rumelhart, D. E., Hinton, G. E. and Williams, R. J. (1986b), Learning Internal Representations by Error Propagation, Parallel Distributed Processing, Vol. 1: Foundations, The MIT Press, MA.

Sardis, G. N. (1970), Learning Applied to Successive Approximation Algorithms, IEEE Transactions on Systems Science and Cybernetics, SSC-6, 97-103.

Sfetsos, A. ( 1999), A comparison of various forecasting techniques applied to mean hourly wind speed time series, Renewable Energy, 21(1), Sep, 23-35.

Shawe-Taylor, J., Jeavons, P. and van Daalen, M. (1991), Probabilistic Bit Stream Neural Clip: Theory, Connection Science 3(3), 317-328.

Silva, W. J. (1991), Global Characteristics and Site Geometry, Proceeding: NSF/EPRI Workshop on Dynamic Soil Properties and Site Characterization. Electric Power Research Institute, EPRI NP-7337.

Specch, L. F. (1990), Probability Neural Networks, Neural Networks 3, 109-118.

Steven, M. Kay (1988), Modern Spectral Estimation: Theory & Application.

Wu, X. (1991), Neural Network-Based Material Modeling, Ph.D. Thesis, University of Illinois at Urbana-Champaign, Urbana, Illinois.

Wu, X. and Ghaboussi, J. (1995), Neural Network Based Material Modeling, Cicil Engineering Studies, SRS 599, University of Illinois, Urbana, Illinois.

Yang, C. Y. (1985), Random Vibration of Structures.

陳鶴修，「台灣地區風速頻譜之初步研究」，台灣科技大學營建工程研究所碩士論文，陳瑞華教授指導（1998）。

張斐章，張麗秋與黃浩倫，類神經網路理論與實務，東華書局，民國 92 年 9 月。

葉怡成，類神經網路模式應用與實作，儒林圖書，民國 89年 7 月。

黃俊銘，數值方法-使用Matlab程式語言，全華科技圖書，民國90年10月。