類神經網路預測三維開口槽溝對震波垂直振幅阻隔效果

Using Artificial Neural Network to Estimate the Screening Effect of Vertical Amplitude of Seismic Waves on Open Trenches of 3-Dimension

10.6665/JLYIT.2010.9.38

洪昌祺(Chang-Chi Hung)

類神經網路 ； 震波 ； 振幅 ； 槽溝 ； neural network ； seismic waves ； amplitude ； trench

蘭陽學報

9期（2010 / 06 / 01）

38 - 49

繁體中文

本研究提出一個以前向式倒傳遞類神經網路為基礎模式，來預測三維開口槽溝對震波垂直振幅之阻隔效果。輸入前人對於槽溝阻隔震波振幅研究之相關重要參數，先經前處理之處理分析程序篩選後，選擇了槽溝尺寸、形狀、面積及振動基礎與三維開口槽溝相關位置、振動基礎沉埋、土壤性質等具推廣性之輸入研究參數。在網路訓練過程中，使用Cascade Correlation 學習程序，和藉由自動調整學習速率與慣性因子的 Extended-Delta-Bar-Delta (EDBD)演算法，來改善前人採試誤法來決定隱藏層節點個數和網路的學習參數的缺點，建立以三維開口槽溝對震波阻隔成效所代表的平均垂直振幅降低比 (Ā(下標 ry))(下標 open)為其輸出值的網路。經研究後，依網路訓練後的驗證程序，結果顯示類神經網路模式在此方面具有相當良好的學習及預測能力。

This paper presents a back-propagation neural network model to estimate the screening effects of vertical amplitude of seismic waves on 3D open trenches. Input parameters are selected by using genetic algorithm with important parameter analysis and pre-processing transfer functions. The parameters are the size of trench dimension, the relative position of vibrating foundation and trenches, the surrounding environment of soil material property and so on. The number of hidden layer nodes in the network is determined by Cascade Correlation learning processing. The learning parameters of network are determined by Extended-Delta-Bar-Delta algorithm to regulate learning-rate and momentum constant automatically. The output parameter of network is the average vertical amplitude reduction rate. According to the results of this study, the neural network model is a very good approach for 3D open trench in estimating the screening effect of vertical amplitude of seismic wave.

1. Beskos, D.E.,Dasgupta, B.,Vardoulakis, I.G.(1986).Vibration isolation using open or filled trenches. Part 1: 2D-homogeneous soil.Computational Mechanics,1(1),43-63.
2. Chatterjee, S.,Price, B.(1991).Regression analysis by example.John Wiley and Sons.
3. Cybenko, G.(1989).Approximation by superpositions of a sigmoidal function.Mathematics of Control, Signals, and Systems,2,303-314.
4. Fahlman, S. E.,Lebiere, C.(1988).The Cascade-Correlation Learning Architecture.Advances in Neural Information Processing Systems
5. Funahashi, K.(1989).On the approximate realization if continuous mappings by neural networks.Neural Networks,2,183-192.
6. Gately, E.(1996).Neural Networks for Financial Forecasting.New York:Wiley.
7. Haykin, S.(1994).Neural Networks.New York:Macmillan College Publishing Company.
8. Hornik, K.,Stinchcombe, M.,White, H.(1990).Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks.Neural Networks,3,551-560.
9. Koza, J. R.(1993).Genetic Programming II.MIT Press.
10. Minia, A.A.,Williams, R.D.(1990).Acceleration of Back-propagation through learning rate and momentum adaptation.Proceeding of the International Joint Conference on Neural Networks
11. Plutowski, M.,White, H.(1992).Selecting concise training sets from clean data.IEEE Trans Neural Netw.,4(2),305-18.
12. Turkey, J. W.(1977).Exploratory data analysis.Addison Wesley.
13. Yang, Y.B.,Hung, H.H.(1997).A parametric study of wave barriers for reduction of train-induced vibrations.International Journal for Numerical Methods in Engineering,40,3729-3747.
14. 倪勝火、洪昌祺(1998)。矩形條狀填充槽溝對表面波振動阻隔效應之分析。中國土木水利工程學刊，十(三)，457-465。