인공신경망과 M5' model tree를 이용한 Tetrapod 피복블록의 안정수 예측
Prediction of Stability Number for Tetrapod Armour Block Using Artificial Neural Network and M5' Model Tree
1
조회
0
다운로드
6
인용
0
참고문헌
키워드
초록
국내 경사식 방파제의 대표적인 피복재인 Tetrapod는 대부분 경험식을 사용하여 중량을 산정한다. 경험식은 수리 실험의 결과를 곡선맞춤(curve-fitting)하여 제안되기 때문에 실험 오차에 따른 불확실성이 내포되어 있다. 이런 불확실성을 최소화하기 위해 인경신경망과 M5' model tree를 사용하여 피복재 안정수를 예측하였다. 각 모형의 불확실성의 정도는 예측된 안정수와 수리실험의 안정수 사이의 일치지수(index of agreement)를 사용하여 평가하였다. 일치지수가 가장 큰 인공신경망은 우수한 예측 능력을 가지고 있지만 일반 설계자들이 쉽게 사용할 수 없는 큰단점이 있다. 반면에 M5' model tree는 인공신경망보다는 예측 능력이 조금 떨어지지만 기존의 경험식보다는 예측 능력이 우수하고 또한 일반 설계자들이 쉽게 사용할 수 있는 공식의 형태로 주어지는 장점이 있다.
Abstract
It was calculated using empirical formulas for the weight of Tetrapod, which was a representative armor unit in the rubble mound breakwater in Korea. As the formulas were evaluated from a curve-fitting with the result of hydraulic test, the uncertainty of experimental error was included. Therefore, the neural network and M5' model tree were used to minimize the uncertainty and predicted the stability number of armor block. The index of agreement between the predicted and measured stability number was calculated to assess the degree of uncertainty for each model. While the neural network with the highest index of agreement have an excellent prediction capability, a significant disadvantage exists that general designers can not easily handle the method. However, although M5' model tree has a lower prediction capability than the neural network, the model tree is easily used by the designers because it has a good prediction capability compared with the existing empirical formula and can be used to propose the formulas like an empirical formula.
참고문헌 (18)
- 김승우, "국내에서 시공된 Tetrapod 피복재에 대한 Hudson 공식의 부분안전계수 산정", 한국해안·해양공학회논문집, 21(5), pp.345~356 (2009), 한국해안,해양공학회
- 해양수산부, "경사식 방파제의 최적설계 기술개발 (I)" (2001), 한국해양연구원
- 해양수산부, "경사식 방파제의 최적설계 기술개발 (II)" (2001), 한국해양연구원
- Burcharth, H. F, "The PIANC safety factorsystem for breakwaters", Proc. Coastal Structures '99, pp.1125~1144 (2000)
- Stability of rubble mound breakwaters, "Delft hydraulics", ~, pp.H462 (1987)
- De Jong,R.J, "Wave transmission at low-crested structure", 석사, pp.1996
- Erdik,T, "Fuzzy logic approach to conventional rubblemound structures design", Expert Systems with Applications, 36, pp.4162~4170 (2009)
- Etemad-Shahidi, A, "Design of rubblemoundbreakwaters using M5' machine learning method", Applied Ocean Research, 31, pp.197~201 (2009)
- "M5' regresion tree and model tree toolbox forMatlab"[URL]
- Kaku, S., "Design formulas forhydraulic stability of rock slopes under irregular wave attack", Proc. 38th Japanese Conf. on Coast. Engrg, pp.661~665 (1991)
- Kim, D. H, "Neural network for design andreliability analysis of rubble mound breakwaters", Ocean Engineering, 32, pp.1332~1349 (2005)
- Kim, D., "Application of probabilisticneural network to design breakwater armor blocks", Ocean Engineering, 35, pp.294~300 (2008)
- Mase, H., "Neural network forstability analysis of rubble-mound breakwaters", J. Wtrwy. Port.Coast. Ocean Eng, 121(6), pp.294~299 (1995)
- Quinlan,J.R, "Learning with continuous classes", Adams,Sterling, editors. Proceedings of AI'92, pp.343~348 (1992)
- Smith, W. G., "Profile changesof rock slopes by irregular waves", Proc. 23rd Int. Conf. onCoast. Engrg, pp.1559~1572 (1992)
- US Army Corps of Engineers, "Shore Protection Manual" (1987), U.S. Army Corps of Engineers
- van der Meer,J.W, "Stability of Cubes, Tetrapods andAccropods", Proc. of the Breakwaters '88 Conference; Design ofBreakwaters, Institution of Civil Engineering, pp.71~80 (1988)
- Willmott,C.J, "On the validation of models", Phys. Geogr, 2(2), pp.184~194 (1981)