پیوندهای مفید
اخبار و اعلانات
آمار
| تعداد نشریات | 127 |
| تعداد شمارهها | 7,238 |
| تعداد مقالات | 77,638 |
| تعداد مشاهده مقاله | 161,190,096 |
| تعداد دریافت فایل اصل مقاله | 120,834,810 |
ارزیابی عملکرد مدلهای مبتنی بر یادگیری ماشین در برآورد ابعاد الگوی توزیع رطوبت تحت آبیاری قطرهای | ||
| تحقیقات آب و خاک ایران | ||
| دوره 57، شماره 5، مرداد 1405، صفحه 1165-1185 اصل مقاله (1.72 M) | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22059/ijswr.2026.413169.670123 | ||
| نویسندگان | ||
| سودابه گلستانی کرمانی* 1؛ میلاد سعیدی عباس آباد2 | ||
| 1استادیار و عضو هیات علمی، گروه علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه شهید باهنر کرمان، کرمان، ایران. | ||
| 2دانشجوی دکتری، گروه علوم و مهندسی آب، دانشگاه آزاد اسلامی واحد کرمان، کرمان، | ||
| چکیده | ||
| امروزه مدلهای یادگیری ماشین به عنوان جایگزین مناسب مدلهای تحلیلی، عددی و تجربی جهت تخمین ابعاد الگوی توزیع رطوبت در خاک بدون محدودیتهایی مانند تعریف شرایط مرزی یا بازکالیبراسیون مورد توجه قرار گرفتهاند. از این رو در تحقیق حاضر با استفاده از اطلاعات هشت متغیر ورودی (درصد شن، درصد سیلت، درصد رس، شوری آب، زمان تجمعی، حجم آب تجمعی، وزن مخصوص ظاهری و هدایت هیدرولیکی اشباع) به ارزیابی عملکرد پنج مدل یادگیری ماشین (CatBoost, XGBoost, RF, SVR, Elnet) برای تخمین عمق نفوذ و عرض سطحی پیاز رطوبتی در دو بافت مختلف خاک (Loamy Sand, Sandy Clay) تحت سیستم آبیاری قطرهای سطحی پرداخته شد. در مدلهای مذکور از روش اعتبارسنجی متقاطع استاندارد با اختلاط تصادفی استفاده شد و عملکرد مدلها با استفاده از شاخصهای آماری R2, RMSE , MAE ارزیابی گردید. نتایج بدست آمده نشان داد که دقیقترین تخمین عمق نفوذ (94/0=R2،cm 27/0=RMSE،cm 10/0=MAE ) و عرض سطحی (98/0=R2،cm 58/1=RMSE،cm 91/0=MAE ) در بافت Sandy Clay و در مدل CatBoost مشاهده شد. ضعیفترین نتایج نیز در مدل Elnet مشاهده شد. بهطوریکه پایینترین دقت تخمین عمق نفوذ و عرض سطحی با مدل مذکور به ترتیب در بافتهای (16/0=R2،cm 16/6=RMSE،cm 22/3=MAE ) Loamy Sandو (76/0=R2،cm 09/6=RMSE،cm 67/4=MAE ) Sandy Clay مشاهده گردید. در مجموع نتایج بدست آمده توانایی مدلهای یادگیری ماشین مبتنی بر روشهای درخت-پایه را که قادر به یادگیری رفتار پیچیده و غیرخطی بین متغیرها هستند، در برآورد ابعاد پیاز رطوبتی به ویژه در خاک با درصد رس بالا تایید میکند. | ||
| کلیدواژهها | ||
| ابعاد الگوی خیس شدگی؛ الگوریتم CatBoost؛ شبیهسازی؛ مدلهای داده محور | ||
| عنوان مقاله [English] | ||
| Performance Evaluation of Machine Learning-Based Models in Estimating Soil Moisture Distribution Dimensions under Drip Irrigation | ||
| نویسندگان [English] | ||
| Soudabeh Golestani Kermani1؛ Milad Saeidi Abbasabad2 | ||
| 1Assistant professor, Department of Water Science and Engineering, Faculty of Agriculture, Shahid Bahonar University of Kerman, Kerman, Iran | ||
| 2PhD Student, Department of Water Science and Engineering, Ke.C., Islamic Azad University of Kerman, Kerman | ||
| چکیده [English] | ||
| In recent years, machine learning models have been increasingly recognized as effective alternatives to analytical, numerical, and empirical models for estimating the dimensions of soil moisture distribution patterns, without requiring constraints such as boundary condition specification or recalibration. Therefore, in the present study, the performance of five machine learning models (CatBoost, XGBoost, RF, SVR, Elnet) was evaluated for estimating infiltration depth and surface width of the wetting bulb in two soil textures (Loamy Sand, Sandy Clay) under surface drip irrigation system. Eight input variables including sand percentage, silt percentage, clay percentage, water salinity, cumulative time, cumulative water volume, bulk density and saturated hydraulic conductivity were used as model inputs. A standard shuffled cross-validation approach was applied and model performance was assessed using statistical indices including R², RMSE, and MAE. The results indicated that the most accurate estimates of infiltration depth (R² = 0.94, RMSE = 0.27 cm, MAE = 0.10 cm) and surface width (R² = 0.98, RMSE = 1.58 cm, MAE = 0.91 cm) were obtained for the Sandy Clay soil using the CatBoost model. The weakest performance was observed for the Elnet model. Specifically, the lowest accuracy for estimating infiltration depth and surface width using this model was obtained in Loamy Sand (R² = 0.16, RMSE = 6.16 cm, MAE = 3.22 cm) and Sandy Clay (R² = 0.76, RMSE = 6.09 cm, MAE = 4.67 cm), respectively. Overall, the findings confirm the strong capability of tree-based machine learning models which are able to capture complex and nonlinear relationships among variables in estimating the dimensions of wetting bulb, particularly in soils with higher clay content. | ||
| کلیدواژهها [English] | ||
| CatBoost algorithm, Data-driven models, Wetting pattern dimensions, Simulation | ||
| مراجع | ||
|
Alahmad, T., Neményi, M., Széles, A., Ali, NA., Hijazi, O., & Nyéki, A. (2025). Spatiotemporal prediction of soil moisture content at various depths in three soil types using machine learning algorithms. Frontiers in Soil Science, 5, 1612908. https://doi.org/10.3389/fsoil.2025.1612908. Al-Ogaidi, AAM., Wayayok, A., Kamal, MR., & Abdullah, AF. (2015). A modified empirical model for estimating the wetted zone dimensions under drip irrigation. Journal Teknologi, 76, 69–73. Al-Ogaidi, AAM., Wayayok, A., Rowshon, MK., & Abdullah, AF. (2016). Wetting patterns estimation under drip irrigation systems using an enhanced empirical model. Agricultural Water Management, 176, 203–213. Amin, MSM., & Ekhmaj, AIM. (2006). DIPAC-Drip irrigation water distribution pattern calculator. 7th International micro irrigation congress, 10–16 September, PWTC, Kuala Lumpur, Malaysia. Arbat, G., Puig-Bargués, J., Duran-Ros, M., Barragán, J., & Ramírez de Cartagena, F. (2013). Drip-Irriwater: computer software to simulate soil wetting patterns under surface drip irrigation. Computers and Electronics in Agriculture, 98, 183–192. Bentéjac, C., Csörgő, A., & Martínez‑Muñoz, G. (2021). A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review, 54, 1937–1967. 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. Cook, FJ., Thorburn, PJ., Fitch, P., & Bristow, KL. (2003). WetUp: a software tool to display approximate wetting patterns from drippers. Irrigation Science, 22, 129-134. Cristóbal-Muñoz, I., Prado-Hernández, JV., Martínez-Ruiz, A., Pascual-Ramírez, F., Cristóbal-Acevedo, D., & Cristóbal-Muñoz, D. (2022). An improved empirical model for estimating the geometry of the soil wetting front with surface drip irrigation. Water, 14(11), 1827. https://doi.org/10.3390/w14111827. Delgado, JA., Short, NM., Roberts, DP., & Vandenberg, B. (2019). Big data analysis for sustainable agriculture on a geospatial cloud framework. Frontiers in Sustainable Food Systems, 3, Article 469303. https://doi.org/10.3389/fsufs.2019.00054 Dhaliwal, JK., Panday, D., Saha, D., Lee, J., Jagadamma, S., Schaeffer, S., & Mengistu, A. (2022). Predicting and interpreting cotton yield and its determinants under long-term conservation management practices using machine learning. Computers and Electronics in Agriculture, 199, doi:10.1016/j.compag.2022.107107. Elmaloglou, S., Soulis, KX., & Dercas, N. (2013). Simulation of soil water dynamics under surface drip irrigation from equidistant line sources. Water Resource Management, 27, 4131-4148. Hammami, M., & Zayani, K. (2016). An analytical approach to predict the moistened bulb volume beneath a surface point source. Agricultural Water Management, 166, 123-129. Hanson, B.R. Gratten, S.R. & Fulton, A. (2006). Agricultural salinity and drainage. Regents of the University of California, Oakland, 180. Hong, H., Pourghasemi, HR., & Pourtaghi, ZS. (2016). Landslide susceptibility assessment in lianhua county (China): a comparison between a random forest data mining technique and bivariate and multivariate statistical models, Geomorphology, 259, 105–118. Kandelous, M., Liaghat, A., & Abbasi, F. (2008). Estimation of soil moisture pattern in subsurface drip irrigation using dimensional analysis methods. The Journal of Agricultural Science, 39(2), 371-378. (In Persian). Kandelous, MM., & Šimůnek, J. (2010). Comparison of numerical, analytical, and empirical models to estimate wetting patterns for surface and subsurface drip irrigation. Irrigation Science, 28, 435–444. Karimi, B., Mohammadi, P., Sanikhani, H., Salih, SQ., & Yaseen, ZM. (2020). Modeling wetted areas of moisture bulb for drip irrigation systems: An enhanced empirical model and artificial neural network. Computers and Electronics in Agriculture, 178, 105767. Kheimi, M., Alotaibi, F., & Alqahtani, A. (2025). Conventional and advanced ai-based models in soil moisture prediction. Journal of Hydrology, (In press). https://doi.org/10.1016/j.jhydrol.2025.XXXXXX. Kisi, O., Khosravinia, P., Heddam, S., Karimi, B., & Karimi, N. (2021). Modeling wetting front redistribution of drip irrigation systems using a new machine learning method: adaptive neuro-fuzzy system improved by hybrid particle swarm optimization-gravity search algorithm. Agricultural Water Management, 256. 107067. Kusumavathi, K., Konatala, R., Lai, P., Sarkar, S., Banerjee, H., Bandopadhyay, P., Sethi, D., & Upendar, K. (2025). Artificial intelligence for fostering sustainable agriculture. Gurrent Plant Biology, 42, 100476. Malek, K., & Peters, RT. (2011). Wetting pattern models for drip irrigation, new empirical models. Journal of Irrigation and Drainage Engineering, 137, 530-536. Mirzaeitalarposhti, R., Shafizadeh-Moghadam, H., & Demyan, MS. (2022). Digital soil texture mapping and spatial transferability of machine learning models using sentinel-1, sentinel-2, and terrain-derived covariates. Remote Sensing, 14(23), 5909. https://doi.org/10.3390/rs14235909. Moncef, H., & Khemaies, Z. (2016). An analytical approach to predict the moistened bulb volume beneath a surface point source. Agricultural Water Management, 166, 123–129. Nikbakht, J., & Abdollahi Siahkalroudi, M. (2014). Effect of magnetization of irrigation water on the properties of soil wetting pattern in surface drip irrigation. Water and Soil Science, 24(4), 139-152. (In Persian). Nogueira, LSR., De Carvalho, MAS., Santos, BDO., Yonaba, R., Bamal, A., Uddin, MG., Bodini, M., & Goliatt, L. (2026). A comparative study of ensemble and non-ensemble machine learning methods for predicting river pollution index. Ecological Informatics, 81, 103617. https://doi.org/10.1016/j.ecoinf.2025.103617. Priyanka, P., Kumar, P., & Panda, S. (2024). Can machine learning models predict soil moisture evaporation rates? an investigation via novel feature selection techniques and model comparisons. Frontiers in Earth Science, 12, 1344690. https://doi.org/10.3389/feart.2024.1344690. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, AV., & Gulin, A. (2018). Catboost: unbiased boosting with categorical features. Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montréal, 3-8 December, 6639-6649. Rajhi, M., Deak, T., & Dobos, E. (2026). Non-invasive soil texture prediction psing machine learning and multi-source environmental data. Soil Systems, 10(1), 8. https://doi.org/10.3390/soilsystems10010008 Saeidi Abbasabad, M., Golestani Kermani, S., Mohayeji Nasrabadi, M., & Zounemat-Kermani, M. (2024). Effect of magnetic fields on saline water distribution pattern under time in drip irrigation. Iranian Journal of Irrigation and Drainage, 1(18), 185-204. (In Persian). Samadianfard, S., Sadraddini, AA., Nazemi, AH., Provenzano, G., & Kisi, O. (2012). Estimation soil wetting pattern for drip irrigation using genetic programming. Spanish Journal of Agricultural Research, 10(4), 1155-1166. Segovia, JA., Toaquiza, JF., Llanos, JR., & Rivas, DR. (2023). Meteorological variables forecasting system using machine learning and open‑source software. Electronics, 12(4), 1007. Seifu Majdar, R., Rahnamaei, A., & Babazadeh, V. (2025). Hybrid machine learning in hydrological runoff forecasting: an exploration of extreme gradient-boosting and categorical gradient boosting optimization in the russian river basin. Advances in Engineering and Intelligence Systems, 4(2). https://doi.org/10.22034/aeis.2025.509199.1293. Sejna, M., Simunek, J., & Van Genuchten, MT. (2014). The HYDRUS software package for simulating two and three dimensional movement of water, heat and multiple solutes in variably – saturated porous media, version 2-04. (PC Progress, Prague, Czech Republic). Shiri, J., Karimi, B., Karimi, N., Kazemi, MH., & Karimi, S. (2020). Simulating wetting front dimensions of drip irrigation systems: multi criteria assessment of soft computing models. Journal of Hydrology, 585, 124792. Sishodia, RP., Ray, RL., & Singh, SK. (2020). Applications of remote sensing in precision agriculture: a review. Remote Sensing, 12(19), 1–31. Smola, AJ., & Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3), 199–222. Taheri, M., Bigdeli, M., Imanian, H., & Mohammadiam, A. (2025). An overview of machine-learning methods for soil moisture estimation. Water, 17(11), 1638. https://doi.org/10.3390/w17111638 Vanwinckelen, G., & Blockeel, H. (2012). On estimating model accuracy with repeated cross-validation. Proceedings of the 21st Belgian-Dutch conference on machine learning, 39-44. Vapnik, V. (1984). Estimation of dependences based on empirical data. Springer-Verlag, 400 p. https://books.google.nl/books?id=wxFS0AEACAAJ Vapnik, V., & Chervonenkis, A. (1974). Theory of pattern recognition. Nauka, Moscow. 353 p. Wang, X., Liu, T., Zheng, X., Peng, H., Xin, J., & Zhang, B. (2018). Short‑term prediction of groundwater level using improved random forest regression with a combination of random features. Applied Water Science, 8(5), 1–12. Zhang, X., Sun, X., & Lin, Z. (2025). Improving soil moisture prediction using gaussian process regression. Smart Agricultural Technology, 11, 100905. https://doi.org/10.1016/j.atech.2025.100905. Zhu, Z., Waseem Rasheed, M., Safdar, M., Yao, B., Tumaerbai, H., Sarwar, A., & Zhu, L. (2024). Intermittent drip irrigation soil wet front prediction model and effective water storage analysis. Sustainability, 16, 9553. https://doi.org/10.3390/su16219553 Zou, H., & Hastie, T. (2005). Regularization and variable selection via the Elastic net. Journal of the Royal Statistical Society Series A, 67 (2), 301-320. | ||
|
آمار تعداد مشاهده مقاله: 46 تعداد دریافت فایل اصل مقاله: 38 |
||
سامانه مدیریت نشریات علمی. قدرت گرفته از سیناوب