All Issue

2022 Vol.64, Issue 6 Preview Page
Development of a Stochastic Snow Depth Prediction Model Using a Bayesian Deep Learning Method

베이지안 딥러닝 기법을 이용한 확률적 적설심 예측 모델 개발

Youngjoon Jeonga
,
Sang-ik Leeb
,
Jonghyuk Leec
,
Byunghun Seod
,
Dongsu Kimd
,
Yejin Seod
,
Won Choie†
정 영준a
,
이 상익b
,
이 종혁c
,
서 병훈d
,
김 동수d
,
서 예진d
,
최 원e†
aPh.D. Student, Department of Rural Systems Engineering, Seoul National University
bResearcher, Department of Rural Systems Engineering, Research Institute of Agriculture and Life Sciences, Seoul National University
cPh.D. Student, Department of Rural Systems Engineering, Global Smart Farm Convergence Major, Seoul National University
dMS Student, Department of Rural Systems Engineering, Global Smart Farm Convergence Major, Seoul National University
eAssociate Professor, Department of Rural Systems Engineering, Research Institute of Agriculture and Life Sciences, Global Smart Farm Convergence Major, Seoul National University
a서울대학교 농촌시스템공학과 박사과정 학생
b서울대학교 농업생명과학연구원 농촌시스템공학과 연구원
c서울대 글로벌 스마트팜 융합전공 농촌시스템공학과 박사과정 학생
d서울대 글로벌스마트팜융합전공 농촌시스템공학과 MS학생
e서울대 농업생명과학연구원 농촌시스템공학과 글로벌스마트팜융합전공 부교수
† Corresponding author Tel.: +82-2-880-4715, Fax: +82-2-873-2087 E-mail: fembem@snu.ac.kr
2022. pp. 35-41
PDF XML Citation
Abstract

Heavy snow damage can be prevented in advance with an appropriate security system. To develop the security system, we developed a model that predicts snow depth after a few hours when the snow depth is observed, and utilized it to calculate a failure probability with various types of greenhouses and observed snow depth data. We compared the Markov chain model and Bayesian long short-term memory models with varying input data. Markov chain model showed the worst performance, and the models that used only past snow depth data outperformed the models that used other weather data with snow depth (temperature, humidity, wind speed). Also, the models that utilized 1-hour past data outperformed the models that utilized 3-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 compared the selected model and the shifting method, which uses present data as future data without prediction, and the model outperformed the shifting method when predicting data after 11-24 hours.

Keywords
Snow depth prediction
long short-term memory
Bayesian model
deep learning
time-series modeling
References

REFERENCES

1
C. Blundell, J. Cornebise, K. Kavukcuoglu and D. Wierstra, arXiv preprint arXiv:1505.05424, Weight Uncertainty in Neural Networks (2015)10.48550/arXiv.1505.05424
2
C. Choi, J. Kim, J. Kim, D. Kim, Y. Bae and H. S. Kim, Advances in Meteorology, Development of heavy rain damage prediction model using machine learning based on big data, 2018 (2018)10.1155/2018/5024930
3
P. Hajikhodaverdikhan, M. Nazari, M. Mohsenizadeh, S. Shamshirband and K. W. Chau, Engineering Applications of Computational Fluid Mechanics, Earthquake prediction with meteorological data by particle filter-based support vector regression, 12(1); 679-688 (2018)10.1080/19942060.2018.1512010
4
O. Hamidi, L. Tapak, H. Abbasi and Z. Maryanaji, Theoretical and Applied Climatology, Application of random forest time series, support vector regression and multivariate adaptive regression splines models in prediction of 365 snowfall (a case study of Alvand in the middle Zagros, Iran), 134(3-4); 769-776 (2018)10.1007/s00704-017-2300-9
5
E. Hart, K. Sim, K. Kamimura, C. Meredieu, D. Guyon and B. Gardiner, Agricultural and Forest Meteorology, Use of machine learning techniques to model wind damage to forests, 265; 16-29 (2019)10.1016/j.agrformet.2018.10.022
6
Y. Hu, T. Che, L. Dai and L. Xiao, Remote Sensing, Snow depth fusion based on machine learning methods for the Northern Hemisphere, 13(7); 1250 (2021)10.3390/rs13071250
7
D. Kim, Y. Jeong, S I Lee, J Lee, K. Hwang and W. Choi, Journal of The Korean Society of Agricultural Engineers, Development of Snow Load Sensor and Analysis of Warning Criterion for Heavy Snow Disaster Prevention Alarm System in Plastic Greenhouse, 63(2); 75-84, (in Korean) (2021)10.5389/KSAE.2021.63.2.075
8
Y. Kim, N. Kang, S. Kim and H. Kim, Korean Society of Hazard Mitigation, Evaluation for Snowfall Depth Forecasting using Neural Network and Multiple Regression Models, 13(2); 269-280, (in Korean) (2013)10.9798/kosham.2013.13.2.269
9
Y. Kim, S. Kim, N. Kang, T. Kim and H. Kim, Journal of Korean Society of Hazard Mitigation, Estimation of Frequency Based Snowfall Depth Considering Climate Change Using Neural Network, 14(1); 93-107, (in Korean) (2014)10.9798/kosham.2014.14.1.93
10
S. H. Kwon and G. Chung, KSCE Journal of Civil and Environmental Engineering Research, Estimation of Snow Damages using Multiple Regression Model - The Case of Gangwon Province -, 37(1); 61-72, (in Korean) (2017)10.12652/ksce.2017.37.1.0061
11
H. J. Lee and G. Chung, Journal of Wetlands Research, Categorical Prediction and Improvement Plan of Snow Damage Estimation using Random Forest, 21(2); 157-162, (in Korean) (2019)10.17663/JWR.2019.21.2.157
12
S. I. Lee, Y. J. Jeong, J. H Lee, G. Chung and W. Choi, Biosystems Engineering, Development of a heavy snowfall alarm model using a Markov chain for disaster prevention to greenhouses, 200; 353-365 (2020)10.1016/j.biosystemseng.2020.10.019
13
H. Liang, M. Zhang and H. Wang, IEEE Access, A neural network model for wildfire scale prediction using meteorological factors, 7; 176746-176755 (2019)10.1109/access.2019.2957837
14
H. Lu, M. Chen and W. Kuang, Transport Policy, The impacts of abnormal weather and natural disasters on transport and strategies for enhancing ability for disaster prevention and mitigation, 98; 2-9 (2020)10.1016/j.tranpol.2019.10.006
15
Ministry of the Interior and Safety, 2020 Disaster Report; 276-352 (2021)
16
Y. Oh and G. Chung, KSCE Journal of Civil and Environmental Engineering Research, Estimation of Snow Damage and Proposal of Snow Damage Threshold based on Historical Disaster Data, 37(2); 325-331, (in Korean) (2017)10.12652/ksce.2017.37.2.0325
17
Y. Oh, G. Lee, K. S. Jun, W. Sunwoo, S. Baek and G. Chung, Journal of the Korean Society of Hazard Mitigation, A Study on the Prediction of 390 Daily Snowmelt Depth using Multiple Linear Regression, 20(6); 311-321, (in Korean) (2020)10.9798/kosham.2020.20.6.311
18
H. Park, S. Jeong and G. Chung, Journal of Korean Society of Hazard Mitigation, Frequency Analysis of Future Fresh Snow Days and Maximum Fresh Snow Depth using Artificial Neural Network under Climate Change Scenarios, 14(6); 365-377, (in Korean) (2014)10.9798/kosham.2014.14.6.365
19
A. T. D. Perera, V. M. Nik, D. Chen, J. L. Scartezzini and T. Hong, Nature Energy, Quantifying the impacts of climate change and extreme climate events on energy systems, 5(2); 150-159 (2020)10.1038/s41560-020-0558-0
20
T. Röösli, C. Appenzeller and D. N. Bresch, Meteorological Applications, Towards operational impact forecasting of building damage from winter windstorms in Switzerland, 28(6); e2035 (2021)10.1002/met.2035
21
W. T. Scherer and D. M. Glagola, Journal of Transportation Engineering, Markovian models for bridge maintenance management, 120(1); 37-51 (1994)10.1061/(asce)0733-947x(1994)120:1(37)
22
M. S. Song, H. S. Yun, J. J. Lee and S. G. Yum, A Comparative Analysis of Machine Learning Algorithms for Snowfall Prediction Models in South Korea, Natural Hazards and Earth System Sciences, preprint (2022)10.5194/nhess-2022-118
23
X. Zhang, X. Li, L. Li, S. Zhang and Q. Qin, Journal of Arid Land, Environmental factors influencing snowfall and snowfall prediction in the Tianshan Mountains, Northwest China, 11(1); 15-28 (2019)10.1007/s40333-018-0110-2
Information
  • Publisher :The Korean Society of Agricultural Engineers
  • Publisher(Ko) :한국농공학회
  • Journal Title :Journal of Korean Society of Agricultural Engineers
  • Journal Title(Ko) :한국농공학회논문집
  • Volume : 64
  • No :6
  • Pages :35-41
  • Received Date : 2022-08-05
  • Revised Date : 2022-09-15
  • Accepted Date : 2022-09-20
  • DOI :https://doi.org/10.5389/KSAE.2022.64.6.035
Journal Informaiton Journal of Korean Society of Agricultural Engineers Journal of Korean Society of Agricultural Engineers
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crossref funder-registry
  • crosscheck
  • orcid
  • open access
Journal Informaiton Journal Informaiton - close