활성화 함수에 따른 유출량 산정 인공신경망 모형의 성능 비교
Comparison of Artificial Neural Network Model Capability for Runoff Estimation about Activation Functions
0
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
Analysis of runoff is substantial for effective water management in the watershed. Runoff occurs by reaction of a watershed to the rainfall and hasnon-linearity and uncertainty due to the complex relation of weather and watershed factors. ANN (Artificial Neural Network), which learns from thedata, is one of the machine learning technique known as a proper model to interpret non-linear data. The performance of ANN is affected by the ANN’sstructure, the number of hidden layer nodes, learning rate, and activation function. Especially, the activation function has a role to deliver theinformation entered and decides the way of making output. Therefore, It is important to apply appropriate activation functions according to the problemto solve. In this paper, ANN models were constructed to estimate runoff with different activation functions and each model was compared and evaluated. Sigmoid, Hyperbolic tangent, ReLU (Rectified Linear Unit), ELU (Exponential Linear Unit) functions were applied to the hidden layer, and Identity,ReLU, Softplus functions applied to the output layer. The statistical parameters including coefficient of determination, (Nash and SutcliffeEfficiency), ln (modified ), and (Percent BIAS) were utilized to evaluate the ANN models. From the result, applications ofHyperbolic tangent function and ELU function to the hidden layer and Identity function to the output layer show competent performance rather thanother functions which demonstrated the function selection in the ANN structure can affect the performance of ANN.
Abstract
Analysis of runoff is substantial for effective water management in the watershed. Runoff occurs by reaction of a watershed to the rainfall and hasnon-linearity and uncertainty due to the complex relation of weather and watershed factors. ANN (Artificial Neural Network), which learns from thedata, is one of the machine learning technique known as a proper model to interpret non-linear data. The performance of ANN is affected by the ANN’sstructure, the number of hidden layer nodes, learning rate, and activation function. Especially, the activation function has a role to deliver theinformation entered and decides the way of making output. Therefore, It is important to apply appropriate activation functions according to the problemto solve. In this paper, ANN models were constructed to estimate runoff with different activation functions and each model was compared and evaluated. Sigmoid, Hyperbolic tangent, ReLU (Rectified Linear Unit), ELU (Exponential Linear Unit) functions were applied to the hidden layer, and Identity,ReLU, Softplus functions applied to the output layer. The statistical parameters including coefficient of determination, (Nash and SutcliffeEfficiency), ln (modified ), and (Percent BIAS) were utilized to evaluate the ANN models. From the result, applications ofHyperbolic tangent function and ELU function to the hidden layer and Identity function to the output layer show competent performance rather thanother functions which demonstrated the function selection in the ANN structure can affect the performance of ANN.
참고문헌 (36)
- Ahn, S. J., "Forecasting of runoff hydrograph using neural network algorithms", Journal of Korean Water Resources Association, 33(4), pp.505~515 (2000)
- Haniyeh Asadi, "Rainfall-Runoff Modelling Using Hydrological Connectivity Index and Artificial Neural Network Approach", Water, 11(2), pp.212~ (2019), MDPI AG[DOI]
- Marina Campolo, "River flood forecasting with a neural network model", Water Resources Research, 35(4), pp.1191~1197 (1999), American Geophysical Union (AGU)[DOI]
- Mohammad Ali Ghorbani, "A comparative study of artificial neural network (MLP, RBF) and support vector machine models for river flow prediction", Environmental Earth Sciences, 75(6), pp.476~ (2016), Springer Science and Business Media LLC[DOI]
- Glorot, X., "Deep sparse rectifier neural networks", Proceedings of the 14th International Conference on Artificial Intelligence and Statistics, 15, pp.315~323 (2011)
- Gecynalda S. da S. Gomes, "Comparison of new activation functions in neural network for forecasting financial time series", Neural Computing and Applications, 20(3), pp.417~439 (2011), Springer Science and Business Media LLC[DOI]
- Günther, Frauke, "Neuralnet : training of neural networks", Journal of Hydrologic Engineering, 2(1), pp.30~38 (2010)[DOI]
- Gupta, H. V., "Status of automatic calibration for hydrologic models : comparison with multilevel expert calibration", Journal of Hydrologic Engineering, 4(2), pp.135~143 (1999)
- 정성호, "딥러닝 오픈 라이브러리를 이용한 하천수위 예측", 한국방재학회논문집, 18(1), pp.1~11 (2018), 한국방재학회[DOI]
- Nachimuthu Karunanithi, "Neural Networks for River Flow Prediction", Journal of Computing in Civil Engineering, 8(2), pp.201~220 (1994), American Society of Civil Engineers (ASCE)[DOI]
- Kim, S. W., "A study on the forecasting of daily streamflow using the multilayer neural networks model", Journal of Korean Water Resources Association, 33(5), pp.537~550 (2000)
- Kingma, D. P., "Adam: a method for stochastic optimization", 3rd International Conference for Learning Representations (2015)
- Özgür Kişi, "Streamflow Forecasting Using Different Artificial Neural Network Algorithms", Journal of Hydrologic Engineering, 12(5), pp.532~539 (2007), American Society of Civil Engineers (ASCE)[DOI]
- A. R. Senthil Kumar, "Rainfall-runoff modelling using artificial neural networks: comparison of network types", Hydrological Processes, 19(6), pp.1277~1291 (2005), Wiley[DOI]
- Suraj Kumar, "A Comparison of Emotional Neural Network (ENN) and Artificial Neural Network (ANN) Approach for Rainfall-Runoff Modelling", Civil Engineering Journal, 5(10), pp.2120~2130 (2019), Ital Publication[DOI]
- Lee, K. S., "The study on the forecasting of runoff applied the B. P. Algorithm of the artificial neural network in the Young-San river", Journal of The Korean Society of Civil Engineers, 20(5B), pp.679~688 (2000)
- Yong Lin, "Surface runoff response to climate change based on artificial neural network (ANN) models: a case study with Zagunao catchment in Upper Minjiang River, Southwest China", Journal of Water and Climate Change, 10(1), pp.158~166 (2019), IWA Publishing[DOI]
- Mishra, P. K., "Performance of optimum neural network in rainfall-runoff modeling over a river basin", International Journal of Environmental Science and Technology, 16, pp.1289~1302 (2019)[DOI]
- Moriasi, D. N., "Hydrologic and Water Quality Models: Performance Measures and Evaluation Criteria", Transactions of the ASABE, 58(6), pp.1763~1785 (2015), American Society of Agricultural and Biological Engineers (ASABE)[DOI]
- Nash, J. E., "River flow forecasting through conceptual models partⅠ-a discussion of principles", Journal of Hydrology, 10(3), pp.282~290 (1970)[DOI]
- Patel, A. B., "Modeling of rainfallrunoff correlations using artificial neural network-a case study of Dharoi watershed of a Sabarmati river basin, India", Civil Engineering Journal, 3(2), pp.78~87 (2017)[DOI]
- Poonia, V., "Rainfall-runoff modeling for the Hoshangabad basin of Narmada river using artificial neural network", Arabian Journal of Geosciences, 13, pp.944~ (2020)[DOI]
- Pushpalatha, R., "A review of efficiency criteria suitable for evaluating low-flow simulations", Journal of Hydrology, 420, pp.171~182 (2012)[DOI]
- Ramachandran, P., "Searching for activation functions", 6th International Conference on Learning Representations (2017)
- Sarkar, A., "Artificial networks for event based rainfall-runoff modeling", Journal of Water Resource and Protection, 4(10), pp.891~897 (2012)[DOI]
- Shamseldin, A. Y., "Artificial neural network model for river flow forecasting in a developing country", Journal of Hydroinformatics, 12(1), pp.22~35 (2010)[DOI]
- Shamseldin, A. Y., "Comparison of differeny forms of the multi-layer feed-forward neural network method used for river flow forecasting", Hydrology and Earth System Sciences, 6(4), pp.671~684 (2002)[DOI]
- Sharma, S., "Activation functions in neural networks", International Journal of Engineering Applied Sciences and Technology, 4(12), pp.310~316 (2020)
- Sibi, P., "Analysis of different activation functions using back propagation neural networks", Journal of Theoretical and Applied Information Technology, 47(3), pp.1344~1348 (2013)
- 송철민, "Curve Number 및 Convolution Neural Network를 이용한 유출모형의 적용성 평가", Ecology and Resilient Infrastructure, 7(2), pp.114~125 (2020), 응용생태공학회[DOI]
- Sudheer, K. P., "Improving peak flow estimation in artificial neural network river flow models", Hydrological Processes, 17(3), pp.677~686 (2003)[DOI]
- Tokar, A. S., "Precipitation-runoff modeling using artificial neural networks and conceptual models", Journal of Hydrologic Engineering, 5(2), pp.156~161 (2000)[DOI]
- Yadav, A. K., "Rainfall-runoff modelling using artificial neural networks(ANNs)model", International Journal of Current Microbiology and Applied Sciences, 9(3), pp.127~135 (2020)[DOI]
- 여운기, "인공신경망과 유전자알고리즘의 결합모형을 이용한 수위예측에 관한 연구", 한국수자원학회 논문집, 43(8), pp.721~731 (2010), 한국수자원학회
- 윤강훈, "신경망을 이용한 낙동강 유역 홍수기 댐유입량 예측", 한국수자원학회 논문집, 37(1), pp.67~75 (2004), 한국수자원학회
- Zadeh, M. R., "Daily outflow prediction by multi layer perceptron with logistic sigmoid and tangent sigmoid activation functions", Water Resources Management, 24, pp.2673~2688 (2010)[DOI]