Journal of Digital Convergence (디지털융복합연구)

The Society of Digital Policy and Management (한국디지털정책학회)
권호 표지
DOI QR코드

DOI QR Code

Production of agricultural weather information by Deep Learning

심층신경망을 이용한 농업기상 정보 생산방법

  • Yang, Miyeon (Dept. of Statistics, Daegu University) ;
  • Yoon, Sanghoo (Division of Mathematics and big data science, Daegu University)
  • 양미연 (대구대학교 통계학과) ;
  • 윤상후 (대구대학교 수리빅데이터학부)
  • Received : 2018.09.11
  • Accepted : 2018.12.20
  • Published : 2018.12.28
Download PDF
⟨ Previous Next ⟩

Abstract

The weather has a lot of influence on the cultivation of crops. Weather information on agricultural crop cultivation areas is indispensable for efficient cultivation and management of agricultural crops. Despite the high demand for agricultural weather, research on this is in short supply. In this research, we deal with the production method of agricultural weather in Jeollanam-do, which is the main production area of onions through GloSea5 and deep learning. A deep neural network model using the sliding window method was used and utilized to train daily weather prediction for predicting the agricultural weather. RMSE and MAE are used for evaluating the accuracy of the model. The accuracy improves as the learning period increases, so we compare the prediction performance according to the learning period and the prediction period. As a result of the analysis, although the learning period and the prediction period are similar, there was a limit to reflect the trend according to the seasonal change. a modified deep layer neural network model was presented, that applying the difference between the predicted value and the observed value to the next day predicted value.

기상은 농작물 재배에 많은 영향을 미친다. 농작물 재배지의 기상정보는 효율적인 농작물 재배 및 관리에 필수적이다. 농업기상 정보의 높은 수요에도 불구하고 이에 대한 연구는 부족하다. 본 연구는 중장기 계절예측정보인 GloSea5와 심층 신경망을 통해 양파의 주산지인 전라남도의 농업기상 정보 생산 방법을 다룬다. 연구방법으로는 매일 생산되는 GloSea5 기상정보를 훈련시키기 위해 슬라이딩 창 방법을 활용한 심층신경망 모형이 사용되었다. 모형의 정확도평가는 농업기상관측소의 일 평균기온과 GloSea5 예측값 그리고 딥러닝 예측값 차이의 RMSE와 MAE로 계산하였다. 심층신경망 모형은 학습기간이 늘어날수록 정확도가 향상되므로 학습기간과 예측기간에 따른 예측성능을 비교하였다. 분석결과 학습기간과 예측기간은 비례하지만 계절변화에 따른 추세성이 반영되는 한계점이 있었다. 이를 보안하기 위해 예측값과 관측값의 차이를 다음날 예측값에 적용시킨 후보정 심층신경망 모형을 제시하였다.

Keywords

DJTJBT_2018_v16n12_293_f0001.png 이미지

Fig. 1. Structure diagram of DNN

DJTJBT_2018_v16n12_293_f0003.png 이미지

Fig. 2. Sliding Window.

DJTJBT_2018_v16n12_293_f0004.png 이미지

Fig. 3. Production areas and Agricultural weather stations

DJTJBT_2018_v16n12_293_f0005.png 이미지

Fig. 4. The location of agricultural weather station and GloSea5

DJTJBT_2018_v16n12_293_f0006.png 이미지

Fig. 5. Boxplot of RMSE and MAE by training days- predicted days : 30 days.

DJTJBT_2018_v16n12_293_f0007.png 이미지

Fig. 6. Boxplot of RMSE and MAE by training days- predicted days : 210 days.

DJTJBT_2018_v16n12_293_f0008.png 이미지

Fig. 7. The result of DNN - predicted days : 30days, training days : 30days.

DJTJBT_2018_v16n12_293_f0009.png 이미지

Fig. 8. The result of DNN - predicted days : 210days, training days : 90days.

DJTJBT_2018_v16n12_293_f0010.png 이미지

Fig. 9. Comparison of obs, GloSea5, DNN1 and DNN2

Table 1. Variables of GloSea5

DJTJBT_2018_v16n12_293_t0001.png 이미지

Table 2. The result of validation by number of grid and predicted days - training days : 60days

DJTJBT_2018_v16n12_293_t0002.png 이미지

Table 3. Comparison of RMSE by GloSea5, DNN1 and DNN2 - training days : 30days

DJTJBT_2018_v16n12_293_t0003.png 이미지

Table 4. Comparison of MAE by GloSea5, DNN1 and DNN2 - training days : 30days

DJTJBT_2018_v16n12_293_t0004.png 이미지

References

  1. G. V. Oldenborgh, F. Doblas-Reyes, B. Wouters & W. Hazeleger. (2012). Skill in the trend and internal variability in a multi-model decadal prediction ensemble. Climate Dynamics, 38(7), 1263-80. https://doi.org/10.1007/s00382-012-1313-4
  2. M. I. Jung, S. W. Son, J. Choi & H. S. Kang. (2015). Assessment of 6-Month Lead Prediction Skill of the GloSea5 Hindcast Experiment. Atmosphere, 25(2), 323-337. https://doi.org/10.14191/Atmos.2015.25.2.323
  3. K. H. Son, D. H. Bae & H. S. Cheong. (2015). Construction & Evaluation of GloSea5-Based Hydrological Drought Outlook System. Atmosphere, 25(2), 271-281. https://doi.org/10.14191/Atmos.2015.25.2.271
  4. J. S. Min, M. H. Lee, J. B. Jee & M. Jang. (2016). A Study of the Method for Estimating the Missing Data from Weather Measurement Instruments. Journal of Digital Convergence, 14(8), 245-252. https://doi.org/10.14400/JDC.2016.14.8.245
  5. J. H. Ha, Y. H. Lee & Y. H. Kim. (2016). Forecasting the precipitation of the next day using deep learning. Journal of Korean Institute of Intelligent Systems, 26(2), 93-98. https://doi.org/10.5391/JKIIS.2016.26.2.093
  6. Q. K. Tran & S. K. Song. (2017). Water Level Forecasting based on Deep Learning: A Use Case of Trinity River-Texas-The United States. Journal of KIISE, 44(6), 607-612. https://doi.org/10.5626/JOK.2017.44.6.607
  7. I. H. Ryu. (2006). A Comparative Study of Time Series Forecasting By Artificial Neural Networks. Master dissertation. Yonsei University, Seoul.
  8. E. M. Yang, H. Jae. Lee & C. H. Seo. (2017). Comparison of Detection Performance of Intrusion Detection System Using Fuzzy and Artificial Neural Network. Journal of Digital Convergence, 15(6), 391-398. https://doi.org/10.14400/JDC.2017.15.6.391
  9. H. S. Song. (2017). Comparison of Performance between MLP and RNN Model to Predict Purchase Timing for Repurchase Product. Journal of information technology applications & management, 24(1), 111-128. https://doi.org/10.21219/jitam.2017.24.1.111
  10. J. S. Kim. (2013). Long-Term Runoff Prediction Using Artificial Neural Network in the Bocheong-Cheon. Master dissertation. Kyung Hee University, Seoul.
  11. K. T. Bae. (2016) Development of a Price Prediction Model of Agricultural Product using Artificial Neural Networks. Master dissertation. Soongsil University, Seoul.
  12. K. K. Seo. (2015). Sales Prediction of Electronic Appliances using a Convergence Model based on Artificial Neural Network and Genetic Algorithm. Journal of Digital Convergence, 13(9), 177-182. https://doi.org/10.14400/JDC.2015.13.9.177
  13. Y. Cho & I. Kim. (2010). Predicting the Performance of Recommender Systems through Social Network Analysis and Artificial Neural Network. Journal of Intelligence and Information Systems, 16(4), 159-172.
  14. M. Kim & C. Hong. (2016). The Artificial Neural Network based Electric Power Demand Forecast using a Season and Weather Informations. Journal of The Institute of Electronics and Information Engineers, 53(1), 71-78. https://doi.org/10.5573/ieie.2016.53.1.071
  15. B. W. Chio. (2006). Prediction of Swell Index of Clay Using the Artificial Neural Networks. Master dissertation. Kyungpook National University, Daegu.
  16. N. R. Jo. (2017). Design and Implementation of criminal Identification System Based on Deep Learning. Master dissertation, Gachon University, Gyeonggi.
  17. S. Moon, S. Han, K. Choi & J. Song. (2016). Data processing system and spatial-temporal reproducibility assessment of GloSea5 model. Journal of Korea Water Resources Association, 49(9), 761-771. https://doi.org/10.3741/JKWRA.2016.49.9.761
  18. Y. Shin & S. Yoon. (2016). Electricity forecasting model using specific time zone. Journal of the Korean Data and Information Science Society, 27(2), 275-284. https://doi.org/10.7465/jkdi.2016.27.2.275