머신러닝 오픈소스 플랫폼을 활용한 쇄파 예측
Prediction of Wave Breaking Using Machine Learning Open Source Platform
1
조회
0
다운로드
7
인용
0
참고문헌
키워드
초록
지금까지 연안에서 발생하는 쇄파에 대한 연구는 지속적으로 수행되었으며, 그에 따른 많은 실험자료가 축적되어 왔다. 또한, 다양한 실험자료로부터 공학적인 적용을 위한 쇄파 정보를 정량적으로 예측하기 위하여 회귀분석에 기반한 다양한 경험식이 제안되었다. 그러나 쇄파는 내재하고 있는 변동성이 있으므로 선형 회귀분석과 같은선형적 통계접근 방법에는 한계가 있다. 본 연구에서는 쇄파파고 및 쇄파수심을 예측하기 위하여 기계학습 중 하나인 신경망을 사용하는 비선형 방법을 제안하였다. 신경망은 구글에서 배포하고 있는 머신러닝 오픈소스 플랫폼인 텐서플로(Tensorflow)를 이용하여 구축하였다. 신경망 모델은 수집된 실험자료를 무작위로 선택하여 학습하였으며, 학습에 이용하지 않은 자료를 사용하여 학습된 신경망을 평가하였다. 학습된 신경망에 의해 예측된 쇄파파고와 쇄파수심에 대한 예측결과는 기존의 경험식에 의한 계산결과에 비해 높은 예측성능을 보였으며, 이는 충분히 학습된 신경망은 쇄파파고 및 수심을 예측하기 위한 유용한 도구로 사용될 수 있음을 보여준다.
Abstract
A large number of studies on wave breaking have been carried out, and many experimental data have been documented. Moreover, on the basis of various experimental data set, many empirical or semi-empirical formulas based primarily on regression analysis have been proposed to quantitatively estimate wave breaking for engineering applications. However, wave breaking has an inherent variability, which imply that a linear statistical approach such as linear regression analysis might be inadequate. This study presents an alternative nonlinear method using an neural network, one of the machine learning methods, to estimate breaking wave height and breaking depth. The neural network is modeled using Tensorflow, a machine learning open source platform distributed by Google. The neural network is trained by randomly selecting the collected experimental data, and the trained neural network is evaluated using data not used for learning process. The results for wave breaking height and depth predicted by fully trained neural network are more accurate than those obtained by existing empirical formulas. These results show that neural network is an useful tool for the prediction of wave breaking.
참고문헌 (28)
- Abadi, M., "Large-scale machine learning on heterogeneous distributed systems. Preliminary White Paper" (2015), Google Research
- Abadi, M., "Tensorflow : Large-scale machine learning on heterogeneous distributed systems", Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (2016), USENIX Association
- Deo, M. C., "Prediction of breaking waves with neural networks", Ocean Engineering, 30, pp.1163~1178 (2003)
- Goda, Y., "A synthesis of breaker indices", Transactions of the Japan Society of Civil Engineers, 2(2), pp.39~40 (1970)
- Goda, Y., "A new method of wave pressure calculation for the design of composite breakwaters", Report of the Port and Harbour Research Institute, 14(3), pp.59~106 (1973)
- Goda, Y., "Reanalysis of regular and random breaking wave statistics", Coastal Engineering Journal, 52(1), pp.71~106 (2010)
- Hinton, G. E., "A fast learning algorithm for deep belief nets", Neural Computation, 18, pp.1527~1554 (2006)
- 김동현, "파고계산 인공신경망을 이용한 방파제 기대피해도 산정", 한국해안·해양공학회논문집, 22(2), pp.126~132 (2010), 한국해안,해양공학회
- Kim, D. H., "Neural network for design and reliability analysis of rubble mound breakwaters", Ocean Engineering, 32, pp.1332~1349 (2005)
- 김승우, "인공신경망과 M5' model tree를 이용한 Tetrapod 피복블록의 안정수 예측", 한국해안·해양공학회논문집, 23(1), pp.109~117 (2011), 한국해안,해양공학회
- Komar, P. D., "Airy wave theory and breaker height prediction", Proceedings of 13th Conference on Coastal Engineering, pp.405~418 (1972)
- Lara, J. L., "Breaking waves over mild gravel slope : Experimental and numerical analysis", Journal of Geophysical Research, 111, pp.1~26 (2006)
- Le Mehaute, B., "On the breaking of waves arriving at an angle to shore", Journal of Hydraulic Research, 5, pp.67~88 (1967)
- Le Roux, J. P., "A simple method to determine breaking height and depth for different wave height", length ratios and floor slope, 2007, pp.54, 271~277
- Liu, Y., "A new predictive formula for inception of regular wave breaking", Coastal Engineering, 58, pp.877~889 (2011)
- Miche, R., "Mouvements ondulatoires de la mer en profondeur constante ou décroissante", Annles de Ponts et Chaussées, 19, pp.370~406 (1944)
- Michell, J. H., "On the highest wave in water", Philosophical Magazine, Edinburgh, Series, 5(36), pp.430~435 (1893)
- Nakamura, M., "Study on wave breaking", Proceedings of 13th Japanese Conference on Coastal Engineering, pp.71~75 (1966), JSCE
- An analysis of longshore current and associated sediment transport in the surf zone, "Ostendorf, D. W.", Department of Civil Engineering, MIT, 1165~1178, pp.Report No. 241 (1979)
- Sunamura, T., "Two-demensional beach transformation due to waves", Proceedings of 13th Conference on Coastal Engineering, pp.920~938 (1974)
- Rattanapitikon, W., "Verification and modification of breaker height formulas", Coastal Engineering Journal, 42(4), pp.389~406 (2000)
- Rattanapitikon, W., "Braking wave formulas for breaking depth and orbital to phase velocity ratio", Coastal Engineering Journal, 48(4), pp.395~416 (2006)
- Rattanapitikon, W., "A proposal of new breaking height formula", Coastal Engineering Journal, 45(1), pp.28~48 (2003)
- Laboratory study on macrofeatures of wave breaking over bars and artificial reefs, "Smith, E. R.", WES, 232~, pp.U.S. Army Corps of Engineers Technical Report CREC-90-12 (1990)
- Tieleman, T., "Coursera: Neural Networks for Machine Learning" (2012)
- Xie, W., "A semi-empirical formula for calculating the breaking depth of plunging waves", Coastal Engineering Journal, 61(2), pp.199~209 (2019)
- Yagci, O., "Artificial intelligence methods in breakwater damage ratio estimation", Ocean Engineering, 32, pp.2088~2106 (2005)
- 이진학, "자기회귀 모델과 신경망 모델을 이용한 복잡한 지형 내 항만에서의 파고 및 하역중단 예측", 한국해안·해양공학회논문집, 29(4), pp.180~188 (2017), 한국해안,해양공학회