ارزیابی دقت الگوریتم‌های مختلف یادگیری ماشین در پیش‌بینی تبخیر- تعرق محصول خیار گلخانه‌ای

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه مهندسی آبیاری و آبادانی، دانشکده کشاورزی، پردیس کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران.

2 گروه علوم باغبانی و فضای سبز، دانشکده کشاورزی، پردیس کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران.

3 گروه علوم و مهندسی خاک، دانشکده کشاورزی، پردیس کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران.

10.22059/jwim.2025.399994.1251

چکیده

در این پژوهش، تبخیر- تعرق گیاه (ETc) خیار گلخانه‌ای (رقم تالیسیا) در شرایط کنترل‌شده گلخانه‌ای در دو دوره رشد پاییز- زمستان 1401 و بهار- تابستان 1402 واقع در پردیس کشاورزی و منابع طبیعی دانشگاه تهران، با هدف مدل‌سازی و پیش‌بینی ETc موردبررسی قرار گرفت. ETc با پایش روزانه رطوبت خاک با استفاده از سنسورهای TDR نصب‌شده در عمق صفر تا 30 سانتی‌متر و به‌کارگیری معادله بیلان آب خاک و ETo از طریق میکرولایسیمتر چمن با دقت دو گرم اندازه‌گیری شد. داده‌های ورودی مدل شامل 10 ویژگی فصل، روز پس از انتقال نشا، دمای حداقل، حداکثر و میانگین، رطوبت نسبی حداقل، حداکثر و میانگین، تابش خورشیدی و تبخیر- تعرق مرجع بود. همبستگی پیرسون نشان داد که متغیرهای روز پس از انتقال نشا، میانگین دما و تابش بیش‌ترین ارتباط مثبت با ETc داشتند. برای پیش‌بینی ETc، شش الگوریتم یادگیری ماشین شامل رگرسیون مؤلفه‌های اصلی (PCR)، حداقل مربعات جزئی (PLS)، ماشین بردار پشتیبان (SVM)، جنگل تصادفی (RF)، گرادیان بوستینگ (GB) و اکستریم گرادیان بوستینگ (XGB) در زبان برنامه‌نویسی پایتون پیاده‌سازی شد. بهینه‌سازی هایپرپارامترها با بهره‌گیری از الگوریتم TPE کتابخانه Optuna با 200 تکرار صورت گرفت. ارزیابی عملکرد مدل‌ها براساس شاخص‌های آماری R2، RMSE، MAE و NSE با اعتبارسنجی متقابل پنج بخشی و سه تکرار انجام شد. نتایج مدل‌سازی نشان داد الگوریتم GB با میانگین R2، RMSE، MAE و NSE به‌ترتیب برابر با 90/0، 59/0 میلی‌متر در روز، 41/0 میلی‌متر در روز و 89/0 بالاترین دقت و عملکرد را نسبت به سایر مدل‌ها دارد. به‌دنبال آن، الگوریتم‌های XGB، RF و SVM نیز با عملکرد نزدیک به GB قرار گرفتند و با آن اختلاف معنی‌دار نداشتند. تحلیل SHAP به‌عنوان ابزار تفسیر مدل نشان داد که ویژگی‌های روز پس از انتقال نشا، تبخیر- تعرق مرجع و تابش بیش‌ترین سهم را در مدل‌سازی ETc دارند. در مجموع، نتایج این پژوهش نشان داد که الگوریتم‌های یادگیری ماشین مبتنی بر درخت تصمیم می‌توانند ابزار دقیقی برای پیش‌بینی نیاز آبی خیار گلخانه‌ای باشند و در مدیریت بهینه آبیاری و بهره‌وری منابع آب نقش به‌سزایی ایفا نمایند.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Evaluation of the accuracy of different machine learning algorithms in predicting greenhouse cucumber crop evapotranspiration

نویسندگان [English]

  • Morteza khoshsimaie chenar 1
  • Hamideh Noory 1
  • Abdolmajid Liaghat 1
  • Forouzandeh Soltani Salehabadi 2
  • Babak Motesharezadeh 3
1 Department of Irrigation and Reclamation Engineering, Faculty of Agriculture, College of Agriculture and Natural Resources, University of Tehran, Karaj, Iran.
2 Department of Horticultural Science and Landscape Architecture, Faculty of Agriculture, College of Agriculture and Natural Resources, University of Tehran, Karaj, Iran.
3 Department of Soil Science, Faculty of Agriculture, College of Agriculture and Natural Resources, University of Tehran, Karaj, Iran.
چکیده [English]

In this study, the crop evapotranspiration (ETc) of greenhouse cucumber was investigated under controlled greenhouse conditions during two growing periods: autumn – winter 2022 and spring – summer 2023, at the College of Agriculture and Natural Resources, University of Tehran. The objective was to model and accurately predict ETc. Daily soil moisture was measured using TDR sensors installed at a depth of 0–30 cm and the soil water balance equation was applied. Reference evapotranspiration (ETo) was measured using a grass micro-lysimeter with an accuracy of 2 grams. The input data for modeling included 10 meteorological variables recorded during the growing period: days after transplanting, minimum, maximum, and average temperature, minimum, maximum, and average relative humidity, solar radiation, and reference evapotranspiration. Pearson correlation analysis revealed that days after transplanting, average temperature, and solar radiation had the strongest positive correlations with ETc. To predict ETc, six machine learning algorithms were implemented in Python: Principal Component Regression (PCR), Partial Least Squares Regression (PLS), Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting (GB) and Extreme Gradient Boosting (XGB). Hyperparameter optimization was conducted using the Tree-structured Parzen Estimator (TPE) algorithm from the Optuna library with 200 iterations. Model performance was evaluated based on R², RMSE, MAE, and NSE metrics using five-fold cross-validation with three repetitions. The modeling results indicated that the GB algorithm achieved the highest accuracy and performance, with average R², RMSE, MAE, and NSE values of 0.90, 0.59 mm/day, 0.41 mm/day, and 0.89, respectively. Following GB, the XGB, RF, and SVM models also performed well, with no statistically significant differences compared to GB. SHAP analysis, used as a model interpretability tool, revealed that days after transplanting, reference evapotranspiration, and solar radiation were the most influential features in predicting ETc. Overall, the results demonstrated that tree-based machine learning algorithms can serve as accurate tools for forecasting the water requirements of greenhouse cucumber and can play a key role in optimizing irrigation management and improving water use efficiency.

کلیدواژه‌ها [English]

  • Reference evapotranspiration
  • SHAP analysis
  • Soil water balance
  • XGBoost algorithm

مراجع

  1. Akiba, T., Sano, S., Yanase, T., Ohta, T., & Koyama, M. (2019). Optuna: A Next-generation Hyperparameter Optimization Framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2623–2631. https://doi.org/10.1145/3292500.3330701
  2. Alfieri, J. G., Blanken, P. D., Smith, D., & Morgan, J. (2009). Concerning the Measurement and Magnitude of Heat, Water Vapor, and Carbon Dioxide Exchange from a Semiarid Grassland. Journal of Applied Meteorology and Climatology, 48(5), 982–996. https://doi.org/10.1175/2008JAMC1873.1
  3. Allen, R. G., Pereira, L. S., Raes, D., Smith, M., & others. (1998). Crop evapotranspiration-Guidelines for computing crop water requirements-FAO Irrigation and drainage paper 56. Fao, Rome, 300(9), D05109.
  4. Assari, M., Kouchakzadeh, M., Shahabifar, M., & Bayat, K. (2012). Estimation of Reference Evapotranspiration in Greenhouse  by Artificial Neural Network. Journal of Water and Soil Conservation, 16(1), 107–121. https://jwsc.gau.ac.ir/article_582.html (In Persian)
  5. Averbuch, N., & Moshelion, M. (2024). Evaluating Evapotranspiration in a Commercial Greenhouse: A Comparative Study of Microclimatic Factors and Machine-Learning Algorithms. https://doi.org/10.1101/2024.01.11.575151
  6. Borg, H., & Grimes, D. W. (1986). Depth Development of Roots with Time: An Empirical Description. Transactions of the ASAE, 29(1), 194–197. https://doi.org/10.13031/2013.30125
  7. Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
  8. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
  9. Dhaliwal, D. S., & Williams, M. M. (2024). Sweet corn yield prediction using machine learning models and field-level data. Precision Agriculture, 25(1), 51–64. https://doi.org/10.1007/s11119-023-10057-1
  10. Eslamian, S. S., Abedi-Koup, J., Amiri, M. J., & Gohari, S. A. (2009). Estimation of Daily Reference Evapotranspiration Using Support Vector Machines and Artificial Neural Networks in Greenhouse. Research Journal of Environmental Sciences, 3(4), 439–447. https://doi.org/10.3923/rjes.2009.439.447
  11. Fan, J., Wu, L., Zhang, F., Cai, H., Wang, X., Lu, X., & Xiang, Y. (2018). Evaluating the effect of air pollution on global and diffuse solar radiation prediction using support vector machine modeling based on sunshine duration and air temperature. Renewable and Sustainable Energy Reviews, 94, 732–747. https://doi.org/10.1016/j.rser.2018.06.029
  12. Fernández, M. D., Bonachela, S., Orgaz, F., Thompson, R., López, J. C., Granados, M. R., Gallardo, M., & Fereres, E. (2010). Measurement and estimation of plastic greenhouse reference evapotranspiration in a Mediterranean climate. Irrigation Science, 28(6), 497–509. https://doi.org/10.1007/S00271-010-0210-Z
  13. Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29(5), 1189–1232. http://www.jstor.org/stable/2699986
  14. Gallegos, R. K. B., Saludes, R. B., & Paras Jr, F. O. (2013). Bowen ratio estimates in green house and open field conditions of the University of the Philippines Los Banos. ASIA LIFE SCIENCES, 22(2), 699–712.
  15. Ge, J., Zhao, L., Yu, Z., Liu, H., Zhang, L., Gong, X., & Sun, H. (2022). Prediction of Greenhouse Tomato Crop Evapotranspiration Using XGBoost Machine Learning Model. Plants, 11(15), 1923. https://doi.org/10.3390/plants11151923
  16. Ghorbani, M. A., Shamshirband, S., Zare Haghi, D., Azani, A., Bonakdari, H., & Ebtehaj, I. (2017). Application of firefly algorithm-based support vector machines for prediction of field capacity and permanent wilting point. Soil and Tillage Research, 172, 32–38. https://doi.org/10.1016/j.still.2017.04.009
  17. Incrocci, L., Thompson, R. B., Fernandez-Fernandez, M. D., De Pascale, S., Pardossi, A., Stanghellini, C., Rouphael, Y., & Gallardo, M. (2020). Irrigation management of European greenhouse vegetable crops. Agricultural Water Management, 242, 106393. https://doi.org/10.1016/j.agwat.2020.106393
  18. James, L. (1988). Principles of farm irrigation systems design. https://www.cabidigitallibrary.org/doi/full/10.5555/19891934086
  19. Jamieson, P. D., Porter, J. R., & Wilson, D. R. (1991). A test of the computer simulation model ARCWHEAT1 on wheat crops grown in New Zealand. Field Crops Research, 27(4), 337–350. https://doi.org/10.1016/0378-4290(91)90040-3
  20. Jolliffe, I. T. (1986). Principal Components in Regression Analysis (pp. 129–155). https://doi.org/10.1007/978-1-4757-1904-8_8
  21. Jones, H. G. (2004). Irrigation scheduling: advantages and pitfalls of plant-based methods. Journal of Experimental Botany, 55(407), 2427–2436. https://doi.org/10.1093/jxb/erh213
  22. JuarezMaldonado, A., BenavidesMendoza, A., deAlbaRomenus, K., & MoralesDiaz, A. (2014). Estimation of the water requirements of greenhouse tomato crop using multiple regression models. Emirates Journal of Food and Agriculture, 26(10), 885. https://doi.org/10.9755/ejfa.v26i10.18270
  23. Jung, D.-H., Lee, T. S., Kim, K., & Park, S. H. (2022). A Deep Learning Model to Predict Evapotranspiration and Relative Humidity for Moisture Control in Tomato Greenhouses. Agronomy, 12(9), 2169. https://doi.org/10.3390/agronomy12092169
  24. Khoshsimaie Chenar, M., Liaghat, A., Noory, H., Soltani Salehabadi, F., & Motesharezadeh, B. (2025). Impact of Different Salinity Levels and Irrigation Water Amounts on Yield and Water Productivity of Greenhouse Cucumber in Two Autumn-Winter and Spring-Summer Growing Periods. Water Management in Agriculture. https://wmaj.iaid.ir/article_217895.html (In Persian)
  25. Kramer, O. (2016). Scikit-Learn. Studies in Big Data, 20, 45–53. https://doi.org/10.1007/978-3-319-33383-0_5
  26. Kuhn, M., & Johnson, K. (2013). Linear Regression and Its Cousins. In Applied Predictive Modeling (pp. 101–139). Springer New York. https://doi.org/10.1007/978-1-4614-6849-3_6
  27. Lakhiar, I. A., Yan, H., Zhang, C., Zhang, J., Wang, G., Deng, S., Syed, T. N., Wang, B., & Zhou, R. (2025). A review of evapotranspiration estimation methods for climate-smart agriculture tools under a changing climate: vulnerabilities, consequences, and implications. Journal of Water and Climate Change, 16(2), 249–288. https://doi.org/10.2166/wcc.2024.048
  28. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30.
  29. Medrano, E., Lorenzo, P., Sánchez-Guerrero, M. C., & Montero, J. I. (2005). Evaluation and modelling of greenhouse cucumber-crop transpiration under high and low radiation conditions. Scientia Horticulturae, 105(2), 163–175. https://doi.org/10.1016/J.SCIENTA.2005.01.024
  30. Mehdizadeh, S., Behmanesh, J., & Khalili, K. (2017). Using MARS, SVM, GEP and empirical equations for estimation of monthly mean reference evapotranspiration. Computers and Electronics in Agriculture, 139, 103–114. https://doi.org/10.1016/j.compag.2017.05.002
  31. Merrill, S. D., Tanaka, D. L., & Hanson, J. D. (2002). Root Length Growth of Eight Crop Species in Haplustoll Soils. Soil Science Society of America Journal, 66(3), 913–923. https://doi.org/10.2136/SSSAJ2002.9130
  32. Natekin, A., & Knoll, A. (2013). Gradient boosting machines, a tutorial. Frontiers in Neurorobotics, 7. https://doi.org/10.3389/fnbot.2013.00021
  33. Nikolaou, G., Neocleous, D., Christou, A., Polycarpou, P., Kitta, E., & Katsoulas, N. (2021). Energy and Water Related Parameters in Tomato and Cucumber Greenhouse Crops in Semiarid Mediterranean Regions. A Review, Part I: Increasing Energy Efficiency. Horticulturae, 7(12), 521. https://doi.org/10.3390/horticulturae7120521
  34. Nikolaou, G., Neocleous, D., Katsoulas, N., & Kittas, C. (2017). Modelling transpiration of soilless greenhouse cucumber and its relationship with leaf temperature in a mediterranean climate. Emirates Journal of Food and Agriculture, 29(12), 911–920. https://doi.org/10.9755/EJFA.2017.V29.I12.1561
  35. Nikolaou, G., Neocleous, D., Katsoulas, N., & Kittas, C. (2019). Irrigation of Greenhouse Crops. Horticulturae, 5(1), 7. https://doi.org/10.3390/horticulturae5010007
  36. Nikolaou, G., Neocleous, D., Kitta, E., & Katsoulas, N. (2023). Estimating cucumber crop coefficients under different greenhouse microclimatic conditions. International Journal of Biometeorology, 67(11), 1745–1756. https://doi.org/10.1007/s00484-023-02535-y
  37. Pandorfi, H., Bezerra, A. C., Atarassi, R. T., Vieira, F. M. C., Barbosa Filho, J. A. D., & Guiselini, C. (2016). Artificial neural networks employment in the prediction of evapotranspiration of greenhouse-grown sweet pepper. Revista Brasileira de Engenharia Agrícola e Ambiental, 20(6), 507–512. https://doi.org/10.1590/1807-1929/agriambi.v20n6p507-512
  38. Rahimikhoob, H., Sohrabi, T., & Delshad, M. (2020). Assessment of reference evapotranspiration estimation methods in controlled greenhouse conditions. Irrigation Science, 38(4), 389–400. https://doi.org/10.1007/s00271-020-00680-5
  39. Rezvani, S., Zarei, G., & Salemi, H. (2022). Evapotranspiration and crop coefficient of greenhouse cucumber in the Hamedan region. Iranian Journal of Irrigation & Drainage, 16(5), 904–916. https://idj.iaid.ir/article_159921.html (In Persian)
  40. Rho, H., Su, J., Sim, H. S., Moon, Y. H., Woo, U. J., & Kim, S. K. (2023). Development of a Cucumber Transpiration Model Based on a Simplified Penman-Monteith Model in a Semi-closed Greenhouse. HortScience, 58(10), 1208–1216. https://doi.org/10.21273/HORTSCI17218-23
  41. Shahrajabian, M. H., & Sun, W. (2024). A Review of Lysimeter Studies and Experiments by Considering Agricultural Production. Journal of Stress Physiology & Biochemistry, 20(2), 114–132.
  42. Shi, W., Zhang, X., Xue, X., Feng, F., Zheng, W., & Chen, L. (2023). Analyzing Evapotranspiration in Greenhouses: A Lysimeter-Based Calculation and Evaluation Approach. Agronomy, 13(12), 3059. https://doi.org/10.3390/agronomy13123059
  43. Shrestha, N. K., & Shukla, S. (2015). Support vector machine-based modeling of evapotranspiration using hydro-climatic variables in a sub-tropical environment. Agricultural and Forest Meteorology, 200, 172–184. https://doi.org/10.1016/j.agrformet.2014.09.025
  44. Stanghellini, C. (1987). Transpiration of greenhouse crops: an aid to climate management. Wageningen University and Research.
  45. Vapnik, V. (1999). The nature of statistical learning theory. Springer science & business media.
  46. Wen, X., Si, J., He, Z., Wu, J., Shao, H., & Yu, H. (2015). Support-Vector-Machine-Based Models for Modeling Daily Reference Evapotranspiration with Limited Climatic Data in Extreme Arid Regions. Water Resources Management, 29(9), 3195–3209. https://doi.org/10.1007/s11269-015-0990-2
  47. Yin, Z., Wen, X., Feng, Q., He, Z., Zou, S., & Yang, L. (2017). Integrating genetic algorithm and support vector machine for modeling daily reference evapotranspiration in a semi-arid mountain area. Hydrology Research, 48(5), 1177–1191. https://doi.org/10.2166/nh.2016.205
  48. Yu, J., Zheng, W., Xu, L., Zhangzhong, L., Zhang, G., & Shan, F. (2020). A PSO-XGBoost Model for Estimating Daily Reference Evapotranspiration in the Solar Greenhouse. Intelligent Automation & Soft Computing, 26(5), 989–1003. https://doi.org/10.32604/iasc.2020.010130
  49. Zhang, Y., Chen, X., Geng, S., & Zhang, X. (2025). A review of soil waterlogging impacts, mechanisms, and adaptive strategies. Frontiers in Plant Science, 16. https://doi.org/10.3389/fpls.2025.1545912
مدیریت آب و آبیاری
دوره 15، شماره 3
آذر 1404
صفحه 563-583

فایل ها

هم رسانی

ارجاع به این مقاله

آمار