DOI QR코드

DOI QR Code

A study on frost prediction model using machine learning

머신러닝을 사용한 서리 예측 연구

  • Kim, Hyojeoung (Department of Applied Statistics, University of Chung-Ang) ;
  • Kim, Sahm (Department of Applied Statistics, University of Chung-Ang)
  • 김효정 (중앙대학교 응용통계학과) ;
  • 김삼용 (중앙대학교 응용통계학과)
  • Received : 2022.04.30
  • Accepted : 2022.05.26
  • Published : 2022.08.31

Abstract

When frost occurs, crops are directly damaged. When crops come into contact with low temperatures, tissues freeze, which hardens and destroys the cell membranes or chloroplasts, or dry cells to death. In July 2020, a sudden sub-zero weather and frost hit the Minas Gerais state of Brazil, the world's largest coffee producer, damaging about 30% of local coffee trees. As a result, coffee prices have risen significantly due to the damage, and farmers with severe damage can produce coffee only after three years for crops to recover, which is expected to cause long-term damage. In this paper, we tried to predict frost using frost generation data and weather observation data provided by the Korea Meteorological Administration to prevent severe frost. A model was constructed by reflecting weather factors such as wind speed, temperature, humidity, precipitation, and cloudiness. Using XGB(eXtreme Gradient Boosting), SVM(Support Vector Machine), Random Forest, and MLP(Multi Layer perceptron) models, various hyper parameters were applied as training data to select the best model for each model. Finally, the results were evaluated as accuracy(acc) and CSI(Critical Success Index) in test data. XGB was the best model compared to other models with 90.4% ac and 64.4% CSI, followed by SVM with 89.7% ac and 61.2% CSI. Random Forest and MLP showed similar performance with about 89% ac and about 60% CSI.

서리는 표면 근처의 공기의 이슬점 온도가 빙점 이하일 때 수증기가 승화, 응축되어 땅이나 물체에 얼게 되는 작은 얼음 결정체이다. 서리가 내리면 농작물이 직접 피해를 입는다. 농작물이 낮은 온도에 접촉하면 조직이 얼어서 세포막이나 엽록체가 딱딱해지고 파괴되거나 건조한 세포가 죽습니다. 2020년 7월, 세계 최대 커피 생산국인 브라질 미나스제라이스 주에 갑작스러운 영하의 날씨와 서리가 내려 지역 커피 나무의 약 30%가 피해를 입었다. 이로 인해 피해로 커피값이 크게 올랐고, 피해가 심각한 농가는 농작물이 회복되기까지 3년이 걸리기 때문에 2024년에야 커피를 생산할 수 있다. 본 논문에서는 심한 서리가 내리는 것을 방지하기 위해 기상청이 제공하는 서리 발생 데이터와 기상관측 데이터를 이용해 서리를 예측하려고 했다. 관측 지점의 고도 및 풍속, 온도, 습도, 강수량, 흐림 등의 기상 요인을 반영하여 모델을 구축하였다. XGB, SVM, Random Forest, MLP 모델을 사용하여 다양한 하이퍼 파라미터를 학습 데이터로 적용하여 각 모델에 가장 적합한 모델을 선택하였다. 마지막으로, 결과는 테스트 데이터에서 정확도(acc)와 중요 성공 지수(CSI)로 평가되었다. XGB는 90.4%의 acc와 64.4%의 CSI로 다른 모델에 비해 최고의 모델이었고, SVM은 89.7%의 acc와 61.2%의 CSI로 그 뒤를 이었다. 랜덤 포레스트와 MLP는 약 89%의 acc와 약 60%의 CSI로 비슷한 성능을 보였다.

Keywords

Acknowledgement

이 논문은 2018년도 대학원생지원장학금의 지원에 의해 작성되었음.

References

  1. Cao Z, Han H, Gu B, and Ren N (2009). A novel prediction model of frost growth on cold surface based on support vector machine. Applied Thermal Engineering, 29(11-12), 2320-2326. https://doi.org/10.1016/j.applthermaleng.2008.11.015
  2. Castaneda-Miranda A and Castano VM (2017). Smart frost control in greenhouses by neural networks models. Computers and Electronics in Agriculture, 137, 102-114. https://doi.org/10.1016/j.compag.2017.03.024
  3. Diedrichs AL, Bromberg F, Dujovne D, Brun-Laguna K, and Watteyne T (2018). Prediction of frost events using machine learning and IoT sensing devices, IEEE Internet of Things Journal, 5(6), 4589-4597. https://doi.org/10.1109/JIOT.2018.2867333
  4. Ding L, Noborio K, and Shibuya K (2019). Frost forecast using machine learning-from association to causality, Procedia Computer Science, 159, 1001-1010. https://doi.org/10.1016/j.procs.2019.09.267
  5. Ghielmi L and Eccel E (2006). Descriptive models and artificial neural networks for spring frost prediction in an agricultural mountain area. Computers and Electronics in Agriculture, 54(2), 101-114. https://doi.org/10.1016/j.compag.2006.09.001
  6. Halil RA??O and Demirci M (2019). Predicting the turkish stock market bist 30 index using deep learning. International Journal of Engineering Research and Development, 11(1), 253-265.
  7. Lee H, Chun JA, Han HH, and Kim S (2016). Prediction of frost occurrences using statistical modeling approaches. Advances in Meteorology.
  8. Lee YB and Ro ST (2002). Frost formation on a vertical plate in simultaneously developing flow. Experimental Thermal and Fluid Science, 26(8), 939-945. https://doi.org/10.1016/S0894-1777(02)00216-9
  9. Rajaei P and Baladi GY (2015). Frost depth: general prediction model. Transportation Research Record, 2510(1), 74-80. https://doi.org/10.3141/2510-09
  10. Rozante JR, Gutierrez ER, da Silva Dias PL, de Almeida Fernandes A, Alvim DS, and Silva VM (2020). Development of an index for frost prediction: Technique and validation. Meteorological Applications, 27(1), e1807.
  11. Sallis P, Jarur M, and Trujillo M (2008, November). Frost prediction characteristics and classification using computational neural networks, In International Conference on Neural Information Processing, 1211-1220.
  12. Tamura Y, Ding L, Noborio K, and Shibuya K (2020, December). Frost prediction for vineyard using machine learning. In 2020 Joint 11th International Conference on Soft Computing and Intelligent Systems and 21st International Symposium on Advanced Intelligent Systems (SCIS-ISIS), 1-4. IEEE.
  13. Wassan S, Xi C, Jhanjhi NZ, and Binte-Imran L (2021). Effect of frost on plants, leaves, and forecast of frost events using convolutional neural networks. International Journal of Distributed Sensor Networks, 17(10), 15501477211053777.
  14. Zendehboudi A and Li X (2017). Robust predictive models for estimating frost deposition on horizontal and parallel surfaces, International Journal of Refrigeration, 80, 225-237. https://doi.org/10.1016/j.ijrefrig.2017.05.013
  15. Zheng H and Wu Y (2019). A xgboost model with weather similarity analysis and feature engineering for short-term wind power forecasting. Applied Sciences, 9(15), 3019.