KIPS Transactions on Software and Data Engineering (정보처리학회논문지:소프트웨어 및 데이터공학)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Machine Learning-Based Greenhouse VPD Prediction Models

머신러닝 기반의 온실 VPD 예측 모델 비교

  • 장경민 (순천대학교 정보통신공학전공) ;
  • 이명배 (순천대학교 정보통신공학전공) ;
  • 임종현 (순천대학교 정보통신공학전공) ;
  • 오한별 (순천대학교 정보통신공학전공) ;
  • 신창선 (순천대학교 인공지능공학부) ;
  • 박장우 (순천대학교 인공지능공학부)
  • Received : 2022.11.30
  • Accepted : 2023.01.25
  • Published : 2023.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we compared the performance of machine learning models for predicting Vapor Pressure Deficits (VPD) in greenhouses that affect pore function and photosynthesis as well as plant growth due to nutrient absorption of plants. For VPD prediction, the correlation between the environmental elements in and outside the greenhouse and the temporal elements of the time series data was confirmed, and how the highly correlated elements affect VPD was confirmed. Before analyzing the performance of the prediction model, the amount and interval of analysis time series data (1 day, 3 days, 7 days) and interval (20 minutes, 1 hour) were checked to adjust the amount and interval of data. Finally, four machine learning prediction models (XGB Regressor, LGBM Regressor, Random Forest Regressor, etc.) were applied to compare the prediction performance by model. As a result of the prediction of the model, when data of 1 day at 20 minute intervals were used, the highest prediction performance was 0.008 for MAE and 0.011 for RMSE in LGBM. In addition, it was confirmed that the factor that most influences VPD prediction after 20 minutes was VPD (VPD_y__71) from the past 20 minutes rather than environmental factors. Using the results of this study, it is possible to increase crop productivity through VPD prediction, condensation of greenhouses, and prevention of disease occurrence. In the future, it can be used not only in predicting environmental data of greenhouses, but also in various fields such as production prediction and smart farm control models.

본 연구에서는 식물의 영양분 흡수에 따른 식물 성장뿐만 아니라 기공 기능 및 광합성에도 영향을 끼치는 온실의 수증기압차(VPD, Vapor Pressure Deficit)예측을 위한 머신러닝 모델들의 성능을 비교해보았다. VPD 예측을 위해 온실 내·외부 환경요소 및 시계열 데이터의 시간적 요소들과의 상관관계를 확인하고 상관관계가 높은 요소들이 VPD에 어떤 영향을 미치는지 확인하였다. 예측 모델의 성능을 분석하기 전 분석 시계열 데이터의 양(1일, 3일, 7일), 간격(20분, 1시간)이 예측 성능에 미치는 영향을 확인하여 데이터의 양과 간격을 조절하였다. 마지막으로 4개의 머신러닝 예측 모델(XGB Regressor, LGBM Regressor, Random Forest Regressor 등)을 적용하여 모델별 예측 성능을 비교했다. 모델의 예측 결과로 20분 간격의 1일의 데이터를 사용했을 때 LGBM에서 MAE는 0.008, RMSE는 0.011의 가장 높은 예측 성능을 보였다. 또한 20분 후 VPD 예측에 가장 큰 영향을 미치는 요소는 환경적 요인보다는 과거 20분 전의 VPD(VPD_y__71)임을 확인하였다. 본 연구의 결과를 활용하여 VPD 예측을 통해 작물의 생산성을 높이고, 온실의 결로, 병 발생 예방 등이 가능하다. 향후 온실의 환경 데이터 예측뿐만 아니라 더 나아가 생산량 예측, 스마트팜 제어 모델 등 다양한 분야에 활용할 수 있을 것이다.

Keywords

Acknowledgement

This research was supported by "Regional Innovation Strategy (RIS)" through the National Research Foundation of Korea funded by the Ministry of Education(MOE)(2021RIS-002).

References

  1. S. J. Oh, "A design of intelligent information system for greenhouse cultivation," The Society of Digital Policy and Management, Vol.15, Iss.2, pp.183-190, 2017. https://doi.org/10.14400/JDC.2017.15.2.183
  2. M. B. Son and D. Y. Han, "Assessment of feature agricultural land use and climate change impacts on irrigation water requirement considering greenhouse culitvation," Journal of the Korean Association of Greographic Information Studies, Vol.23, No.4, pp.120-139, 2020.
  3. C. Amitrano, C. Arena, Y. Rouphael, S. Pascale, and V. De Micco, "Vapour pressure deficit: The hidden driver behind plant morphofunctional traits in controlled environments," Annals of Applied Biology, Vol.175, Iss.3, pp.313-325, 2019. https://doi.org/10.1111/aab.12544
  4. J. Ding, X. Jiao, P. Bai, Y. Hu, J. Zhang, and J. Li, "Effect of vapor pressure deficit on the photosynthesis, growth, and nutrient absorption of tomato seedlings," Scientia Horticulturae, Vol.293, pp.110736, 2022.
  5. T. Inoue, M. Sunaga, M. Ito, Q. Yuchen, Y. Matsushima, K. Sakoda, and W. Yamori, "Minimizing VPD fluctuations maintains higher stomatal conductance and photosynthesis, resulting in improvement of plant growth in lettuce," Frontiers in Plant Science, Vol.12, 2021.
  6. J. Ding, X. Jiao, P. Bai, Y. Hu, J. Zhang, and J. Li, "Effect of vapor pressure deficit on the photosynthesis, growth, and nutrient absorption of tomato seedlings," Scientia Horticulturae, Vol.293, 2022.
  7. H. W. Lee, Y. S. Kim, S. Y. Shim, and J. W. Lee, "Variation of vapor pressure deficit and condensation flux of air heating plastic greenhouse installed with two layers thermal curtain in winter," The Korean Society for Bio-Environment Control, Vol.22, No.1, pp.35-41, 2013. https://doi.org/10.12791/KSBEC.2013.22.1.035
  8. H. M. Noh and J. H. Lee, "The effect of vapor pressure deficit regulation on the growth of tomato plants grown in different planting environments," Applied Sciences, Vol.12, No.7, pp.3667, 2022.
  9. C. Grossiord et al., "Plant responses to rising vapor pressuredeficit," The New Phytologist, Vol.226, No.6, pp.1550-1566, 2020. https://doi.org/10.1111/nph.16485
  10. Heuvelink Ep, "Tomatoes," Cambridge MA: CABI Pub, 2005.
  11. H. C. Kim, S. G. Jung, J. H. Lee, and H. J. Bae, "Effects of greenhouse covering material on environment factors and fruit yield in protected cultivation of sweet pepper," Journal of Bio-Environment Control, Vol.18, No.3, pp.253-257, 2009.
  12. Perfect Grower Vapor Pressure Deficit Recommendations (kPa), Google Chrome [Internet], https://www.perfectgrower.com/knowledge/knowledge-base/vpd-chart-vapor-pressure-deficit/
  13. S. Redmond, C. M. Hasfalina, Z. Abdjamil, V. B. Peter, A. Desa, W. Ismail, and W. Ishak, "Membership function model for defining optimality of vapor pressure deficit in closed-field cultivation of tomato," Acta Horticulturae, 10.17660/ActaHortic.2017.1152.38, 2016.
  14. benro-type vinyl greenhouse, Google Chrome [Internet], http://www.jjn.co.kr/news/articleView.html?idxno=788176
  15. J. H. kim, C. H. Lee, and K. S. Shim, "Time series prediction using clustering algorithm," KIISE Transactions on Computing Practices, Vol.20, No.3, 191-195, 2014.
  16. J. H. Moon, S. W. Park, S. M. Rho, and E. J. Hwang, "A comparative analysis of artificial neural network architectures for building energy consumption forecasting," International Journal of Distributed Sensor Networks, Vol.15, No.9, 2019.
  17. T. Chen and C. Guestrin, "XGBoost: A scalable tree boosting system," arXiv:1603.02754v3 [cs.LG], 2016.
  18. S. Cheon, J. Yu, J. G. Kim, J. S. Oh, T.-H. Nam, and T. Lee, "Predicting deformation behavior of additively manufactured Ti-6Al-4V based on XGB and LGBM," Transactions of Materials Processing, Vol.31, No.4, 2022.
  19. XGBoost & LightGBM, slidshare [Internet], https://www.sli deshare.net/GabrielCyprianoSaca/xgboost-lightgbm
  20. L. Breiman, "Random Forests," Machine Learning, Vol.45, pp.5-32, 2001. https://doi.org/10.1023/A:1010933404324
  21. J. E. Ha, H. C. Shin, and J. G. Lee, "Korean text classification using randomforest and XGBoost focusing on Seoul metropolitan civil complaint data," The Journal of Bigdata, Vol.2, No.2, pp.95-104, 2017. https://doi.org/10.36498/kbigdt.2017.2.2.95
  22. S. W. Rye and D. Y. Park, "Economic analysis based on smart farm colling technology," KIEAE Journal, Vol.21, No.5, pp.55-65, 2021. https://doi.org/10.12813/kieae.2021.21.5.055