인공신경망을 이용한 현장지반의 장래 침하량 산정

Estimates of Settlement in Field Ground Using Neural Networks

  • 발행 : 2003.10.01

초록

본 연구는 기존의 침하량예측법의 단점을 극복하기 위한 방법으로 인공신경망의 적용성을 분석하였다. 연약지반을 개량하기 위해 사용되는 선행재하 공법에서 침하량의 산정은 매우 중요한 부분을 차지하는데, 현재 쌍곡선법, Hoshino법, Asaoka법이 침하량예측에 주로 사용되고 있다. 그러나 이들 방법들은 설계단계에서는 예측이 불가능하다는 단점을 가지고 있다. 반면 인공신경망은 축적된 자료들의 학습을 통해 설계단계에서 예측이 가능하며 비교적 용이하게 적용할 수 있다. 본 연구에서는 장래침하량을 산정하기 위하여 Elman 신경망을 사용하였다.

This study analyzed an application possibility of neural network to overcome problems of conventional settlement prediction. It is very important to estimate settlement in preloading method used to improve soft ground. At present, Hyperbolic method, Hoshino method and Asaoka method are used mostly in the prediction of settlement. But these methods can not predict settlement at the phase of design. On the other hand, neural networks are capable of predicting settlement through accumulated data in the phase of design and this method can be easily applied in practice. In this study Elman neural network is used to estimate future settlement.

키워드

참고문헌

  1. Proc., 9th Conf., ASCE Engrg. Mech. Neural Network modeling of the mechnical behavior of sand Ellis,G.W.;Yao,C.;Zhao,R.
  2. IEEE Trans. Neural Network v.5 no.2 Dynamic recurrent neural networks : theory and application Giles,C.L.;Kuhn,G.W.;Williams,R.J.
  3. J. Geotech. Engrg. Div. v.121 no.5 Stress-strain modeling of sands using artificial neural networks G.W.Ellis,;C.Yao;R.Zhao;D.Penumadu https://doi.org/10.1061/(ASCE)0733-9410(1995)121:5(429)
  4. Congnitive v.14 Finding structure in time Jeffery;L.Elman
  5. Int. J. Numer. Anal. Meth. Geomech. v.22 Modeling of shearing behaviour of a residual soil with recurrent neural network Jian-hua,Z.;Musharraf,M.;Scott,A. https://doi.org/10.1002/(SICI)1096-9853(199808)22:8<671::AID-NAG939>3.0.CO;2-Y
  6. IEEE Trans. Neural Networks v.5 no.2 Application of the recurrent multilayer perceptron in modeling complex process dynamics Parlos,A.g.;Chong,K.T.;Atiya,A.F. https://doi.org/10.1109/72.279189
  7. J. Geotech. Engrg. v.12 no.5 Stress-strain modeling of sands using artifical neural network Peumadu,D.;Zhao,R.;Yao,C.;Ellis,G.W.
  8. Electronic Journal of Geotechnical Engineering Artificial Intelligence Applications in Geotechnical Engineering Toll,D.