베이지안 딥러닝 기법을 이용한 확률적 적설심 예측 모델 개발
Development of a Stochastic Snow Depth Prediction Model Using a Bayesian Deep Learning Method
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
Heavy snow damage can be prevented in advance with an appropriate security system. To develop the security system, we developed a model thatpredicts snow depth after a few hours when the snow depth is observed, and utilized it to calculate a failure probability with various types ofgreenhouses and observed snow depth data. We compared the Markov chain model and Bayesian long short-term memory models with varying inputdata. Markov chain model showed the worst performance, and the models that used only past snow depth data outperformed the models that used otherweather data with snow depth (temperature, humidity, wind speed). Also, the models that utilized 1-hour past data outperformed the models that utilized3-hour data and 6-hour data. Finally, the Bayesian LSTM model that uses 1-hour snow depth data was selected to predict snow depth. We comparedthe selected model and the shifting method, which uses present data as future data without prediction, and the model outperformed the shifting methodwhen predicting data after 11-24 hours.
Abstract
Heavy snow damage can be prevented in advance with an appropriate security system. To develop the security system, we developed a model thatpredicts snow depth after a few hours when the snow depth is observed, and utilized it to calculate a failure probability with various types ofgreenhouses and observed snow depth data. We compared the Markov chain model and Bayesian long short-term memory models with varying inputdata. Markov chain model showed the worst performance, and the models that used only past snow depth data outperformed the models that used otherweather data with snow depth (temperature, humidity, wind speed). Also, the models that utilized 1-hour past data outperformed the models that utilized3-hour data and 6-hour data. Finally, the Bayesian LSTM model that uses 1-hour snow depth data was selected to predict snow depth. We comparedthe selected model and the shifting method, which uses present data as future data without prediction, and the model outperformed the shifting methodwhen predicting data after 11-24 hours.
참고문헌 (23)
- Blundell, C., "Weight Uncertainty in Neural Networks", arXiv preprint arXiv:1505.05424, ~ (2015)
- Changhyun Choi, "Development of Heavy Rain Damage Prediction Model Using Machine Learning Based on Big Data", Advances in Meteorology, 2018, pp.1~11 (2018), Hindawi Limited[DOI]
- Pouria Hajikhodaverdikhan, "Earthquake prediction with meteorological data by particle filter-based support vector regression", Engineering Applications of Computational Fluid Mechanics, 12(1), pp.679~688 (2018), Informa UK Limited[DOI]
- Omid Hamidi, "Application of random forest time series, support vector regression and multivariate adaptive regression splines models in prediction of snowfall (a case study of Alvand in the middle Zagros, Iran)", Theoretical and Applied Climatology, 134(3-4), pp.769~776 (2017), Springer Science and Business Media LLC[DOI]
- Emma Hart, "Use of machine learning techniques to model wind damage to forests", Agricultural and Forest Meteorology, 265, pp.16~29 (2019), Elsevier BV[DOI]
- Yanxing Hu, "Snow Depth Fusion Based on Machine Learning Methods for the Northern Hemisphere", Remote Sensing, 13(7), pp.1250~ (2021), MDPI AG[DOI]
- 김동수, "비닐온실 폭설 방재 예·경보 시스템을 위한 설하중 센서 개발과 적설 경보 기준 분석", 한국농공학회논문집, 63(2), pp.75~84 (2021), 한국농공학회[DOI]
- 김연수, "적설량 예측을 위한 신경망 및 다중회귀 모형의 적용성 평가", 한국방재학회논문집, 13(2), pp.269~280 (2013), 한국방재학회
- 김연수, "신경망 기법을 이용한 기후변화에 따른 미래 확률적설량 산정", 한국방재학회논문집, 14(1), pp.93~107 (2014), 한국방재학회
- 권순호, "대설피해액 추정을 위한 다중회귀 모형의 적용성 평가 - 강원도 지역을 중심으로 -", 대한토목학회논문집(국문), 37(1), pp.61~72 (2017), 대한토목학회[DOI]
- 이형주, "랜덤포레스트를 이용한 대설피해액에 대한 범주형 예측 및 개선방안 검토", 한국습지학회지, 21(2), pp.157~162 (2019), 한국습지학회
- Sang-ik Lee, "Development of a heavy snowfall alarm model using a Markov chain for disaster prevention to greenhouses", Biosystems Engineering, 200, pp.353~365 (2020), Elsevier BV[DOI]
- Hao Liang, "A Neural Network Model for Wildfire Scale Prediction Using Meteorological Factors", IEEE Access, 7, pp.176746~176755 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Huapu Lu, "The impacts of abnormal weather and natural disasters on transport and strategies for enhancing ability for disaster prevention and mitigation", Transport Policy, 98, pp.2~9 (2020), Elsevier BV[DOI]
- 2020 Disaster Report, "Ministry of the Interior and Safety", 276~352 (2020)
- 오영록, "재난통계를 활용한 대설피해 예측 및 대설 피해 적설심 기준 결정 방안", 대한토목학회논문집(국문), 37(2), pp.325~331 (2017), 대한토목학회[DOI]
- 오영록, "기상데이터 통계를 활용한 일 단위 융설 깊이 예측 가능성 검토", 한국방재학회논문집, 20(6), pp.311~321 (2020), 한국방재학회[DOI]
- 박희성, "기후변화 시나리오와 신경망 모형을 이용한 미래 신적설일수와최심신적설 빈도 분석", 한국방재학회논문집, 14(6), pp.365~377 (2014), 한국방재학회
- A. T. D. Perera, "Quantifying the impacts of climate change and extreme climate events on energy systems", Nature Energy, 5(2), pp.150~159 (2020), Springer Science and Business Media LLC[DOI]
- Thomas Röösli, "Towards operational impact forecasting of building damage from winter windstorms in Switzerland", Meteorological Applications, 28(6) (2021), Wiley[DOI]
- William T. Scherer, "Markovian Models for Bridge Maintenance Management", Journal of Transportation Engineering, 120(1), pp.37~51 (1994), American Society of Civil Engineers (ASCE)[DOI]
- Song, M. S., "A Comparative Analysis of Machine Learning Algorithms for Snowfall Prediction Models in South Korea", Natural Hazards and Earth System Sciences, pp.5194~ (2022)[DOI]
- Zhang, X., "Environmental factors influencing snowfall and snowfall prediction in the Tianshan Mountains, Northwest China", Journal of Arid Land, 11(1), pp.15~28 (2019)[DOI]