بررسی عملکرد الگوریتم جنگل تصادفی در پیش‌بینی نوسانات سطح ایستابی در مقایسه‏ با دو مدل‏ درخت تصمیم و شبکۀ عصبی مصنوعی ‌(مطالعۀ موردی: آبخوان آزاد دشت بیرجند)

نوع مقاله : پژوهشی

نویسندگان

1 دانشجوی دکترای مهندسی منابع آب، دانشکدۀ کشاورزی، دانشگاه بیرجند، بیرجند، ایران

2 استاد گروه علوم و مهندسی آب، دانشکدۀ کشاورزی، دانشگاه بیرجند، بیرجند، ایران

3 دانشیار گروه علوم و مهندسی آب، دانشکدۀ کشاورزی، دانشگاه بیرجند، بیرجند، ایران

10.22059/ije.2022.327263.1526

چکیده

امروزه به سبب برداشت‏های بی‏رویه از منابع آب زیرزمینی و افت تراز سطح ایستابی به‌ویژه در مناطق خشک و نیمه‏خشک، برنامه‏ریزی و مدیریت در مصرف این منابع باارزش اهمیت زیادی دارد که این امر نیازمند مطالعۀ رفتار آبخوان نسبت به تغییرات واردشده بر آن است. هدف از انجام این تحقیق، بررسی کارایی الگوریتم جنگل تصادفی در پیش‏بینی تراز سطح ایستابی آبخوان آزاد دشت بیرجند و مقایسۀ نتایج آن با دو مدل درخت تصمیم و شبکۀ عصبی مصنوعی است. در این راستا، ابتدا اطلاعات ورودی به مدل شامل تراز سطح ایستابی چاه‏های مشاهده‏ای، دما، بارندگی، رطوبت و تبخیر طی سال‏های آبی 1389ـ 1390 تا 1395 1396 به صورت ماهانه جمع‏آوری و پس از بررسی روند و حذف آن، برای ایجاد مدل‌های یادشده از بستۀ نرم‏افزاری Rattle در نرم‏افزار آماری R استفاده شد. نتایج حاصل از شبیه‏سازی با استفاده از الگوریتم جنگل تصادفی براساس معیارهای ارزیابی معادل 714/0=R2، 003/0=RMSE متر و 598/0=NS نشان می‏دهد این الگوریتم توانایی نسبتاً زیادی در شبیه‏سازی تراز سطح ایستابی آبخوان دارد. از مقایسۀ نتایج این الگوریتم با دو مدل درخت تصمیم و شبکۀ عصبی مصنوعی می‏توان دریافت که نتایج الگوریتم جنگل تصادفی نسبت به مدل درخت تصمیم با 5409/0=R2، 0072/0=RMSE متر و 0187/0-=NS تطابق بیشتری با تراز واقعی آبخوان دارد و با نتایج شبکۀ عصبی مصنوعی با 7055/0=R2، 003/0=RMSE متر و 6046/0=NS هم‌راستا است. همچنین، خروجی الگوریتم جنگل تصادفی نشان می‏دهد در بین پارامترهای ورودی، چاه‏های مشاهده‏ای واقع در نواحی مرکزی دشت و نیز پارامترهای هواشناسی بارندگی و رطوبت در شبیه‏سازی تراز سطح ایستابی آبخوان نقش مؤثرتری نسبت به سایر پارامترها ایفا می‏کنند.

کلیدواژه‌ها


عنوان مقاله [English]

Investigating the performance of random forest algorithm in predicting water table fluctuations Compared with two models of decision tree and artificial neural network (Case study: unconfined aquifer of Birjand plain)

نویسندگان [English]

  • fatemeh poursalehi 1
  • Abbas KhasheiSiuki 2
  • Seyyed Reza Hashemi 3
1 Water Resources Engineering PhD student of Birjand University
2 university of birjand, Avini street, birjand city, soth khorasan province,iran
3 Associate Professor Department of Water Engineering of University of Birjand
چکیده [English]

Today, due to uncontrolled withdrawal of groundwater resources and declining water table, especially in arid and semi-arid regions, planning and management in the consumption of these valuable resources are of great importance, which requires a study of the behavior of the aquifer in relation to the changes made on it. The purpose of this study is to investigate the efficiency of random forest algorithm in predicting the water table of the unconfined aquifer of Birjand plain and to compare the results with two models of decision tree and artificial neural network. In this regard, first, the input data to the model was collected on a monthly basis during 2010-2011 until 2016-2017 water years, and after checking the trend and removing it, to create the mentioned models, the rattle software package in the statistical software R was used. The results of simulation using the random forest algorithm based on evaluation criteria of R2=0.714, RMSE=0.003 and NS=0.598 (m) show that this algorithm has a relatively high ability to simulate the aquifer water table. Comparing the results of this algorithm with two decision tree and artificial neural network models, it can be seen that the results of the random forest algorithm compared to the decision tree model with R2 = 0.5409, RMSE = 0.0072 and NS = -0.0187 (m) is more consistent with the actual water table of the aquifer and is in line with results of the artificial neural network with R2 = 0.7055, RMSE = 0.003 and NS = 0.6046 (m).

کلیدواژه‌ها [English]

  • Evaluation criteria
  • Groundwater
  • Statistical software R
  • Water Table Simulation
[1]. Rajaee T, Zeynivand A. Modeling of groundwater level using ANN–Wavelet Hybrid model (Case Study: Sharif Abad Plain). Journal of Civil and Environmental Engineering. 2015; 44(4): 51-63. [Persian]
[2]. Mirmorsley N, Karbasi M. Comparison of the J48, Random Forest and Tree Random algorithms efficiency in predicting bed form in sandy rivers. Second Iranian National Hydrology Conference. 2017; Shahrekord University, Shahrekord, Iran. [Persian]
[3]. Rajaee T, Mirbagheri S.A. Suspended sediment model in rivers using artificial neural networks. J. Civil Engin. 2009; 21(1): 27-43. [Persian]
[4]. Rajaee T, Ebrahimi H. Application of wavelet-neural network model for forecasting groundwater level time series with non-stationary and nonlinear characteristics. J. of Water and Soil Conservation. 2016; 22(5): 99-115. [Persian]
[5]. Altunkaynak A. Forecasting surface water level fluctuations of Lake Van by artificial neural networks. Water Resour. Manage. 2007; 21 (2): 399-408.
[6]. Li Y, Zhang Q, Yao J, Werner A.D, Li X. Hydrodynamic and hydrological modeling of the Poyang Lake catchment system in China. J. Hydrol. Eng. 2013; 19 (3): 607–616.
[7]. Li B, Yang G, Wan R, Dai X, Zhang Y. Comparison of random forests and other statistical methods for the prediction of lake water level : a case study of the Poyang Lake in China. Hydrology Research. 2016; 69–83.
[8]. Khalili Naft Chali A, Shahidi A, khashei siuki A. Comparison of Lazy Algorithms and M5 model to estimate groundwater level (Case Study: Plain Neyshabur). JWSS. 2017; 21 (3): 15-26. [Persian]
[9].  Mirhashemi S.H, Haghighat jou P, Mirzaei F, Panahi M. Using CART algorithm in predicting groundwater table fluctuations inside and outside of an irrigation system (case study: irrigating area of Qazvin). Iranian Journal of Soil and Water Research. 2018; 49(2): 385-395. [Persian]
[10]. Mohtasham M, Dehghani A.A, Akbarpour A, Meftah M, Etebari B. Oundwater level determination by using Artificial Neural Network (Case study: Birjand Aquifer). Iranian Journal of lrrigation and drainage. 2010; 4(1): 1-10. [Persian]
 
[11]. Khashei-Siuki A, Ghahraman B, Kouchakzadeh M. Comparison of ANN, ANFIS and Regression models to estimate groundwater level of Neyshaboor Aquifer. Iranian Journal of lrrigation and Drainage. 2013; 1(7): 10-22. [Persian]
[12]. Saeedi Razavi B, Arab A. Groundwater Level Prediction of Ajabshir Plain using Fuzzy Logic, Neural Network Models and Time Series. Hydrogeology. 2019; 3(2): 69-81. [Persian]
[13]. Jabalbarezi B, Malekian A. Comparison of the performance of artificial neural networks and gene expression to predict the groundwater level in arid and semi-arid areas (Case study: Jiroft plain). Iranian Journal of Range and Desert Research. 2019; 26(2): 292-301. [Persian]
[14]. Nayak P.C, Satyaji Rao Y.R, Sudheer K.P. Groundwater level forecasting in a shallow aquifer using artificial neural network approach. Water Resources Management. 2006; 2(1): 77-99.
[15]. Sreekanth P.D, Geethanjali N, Sreedevi P.D, Ahmed S, Ravi Kumar N, Kamala Jayanthi P.D. Forecasting groundwater level using artificial neural networks. Current Science. 2009; 96: 933- 939.
[16]. Sun Y, Wendi D, Kim D.E, Liong S.Y. Technical note: Application of artificial neural networks in groundwater table forecasting - a case study in a Singapore swamp forest. Hydrology and Earth System Sciences. 2016; 20(4): 1405–1412.
[17]. Noruzi H, Nadiri A.A, Asgharimoghaddam A, Gharekhani M. Prediction of Transmissivity of Malikan Plain Aquifer Using Random Forest Method. Water and Soil Science. 2017; 27(2): 61-75. [Persian]
[18]. Norouzi H, Nadiri A. Groundwater Level Prediction of Boukan Plain using Fuzzy Logic, Random Forest and Neural Network Models. Journal of Range & Watershed Management. 2018; 71(3): 829-845. [Persian]
[19]. Wang X, Liu T, Zheng X, Peng H, Xin J, Zhang B. Short‑term prediction of groundwater level using improved random forest regression with a combination of random features. Applied Water Science. 2018; 8(5): 1–12.
[20]. Hamraz B.S, Akbarpour A, Pourreza Bilondi M. Assessment of parameter uncertainty of MODFLOW model using GLUE method (Case study: Birjand plain). Journal of Water and Soil Conservation. 2016; 22(6): 61-79. [Persian]
[21]. Farpoor A, Ramezani Y, Akbarpour A. Numerical Simulation of Chromium Changes Trend in Aquifer of Birjand Plain. Iranian Journal of Irrigation and Drainage. 2018; 12(5): 1203-1216. [Persian]
[22]. Breiman L. Random forests. Mach Learn. 2001; 45(1):5-32.
[23]. Kotsiantis S, Pintelas P. Combining bagging and boosting. Journal of Computational Intelligence. 2004; 1(4): 324-333.
[24]. Guo L, Chehata N, Mallet C, Boukir S. Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests. ISPRS J Photogramm Remote Sens. 2011; 66(1): 56-66.
[25]. Rodriguez-Galiano V, Mendes M.P, Garcia-Soldado M.J, Chica-Olmo M, Ribeiro L. Predictive modeling of groundwater nitrate pollution using Random Forest and multisource variables related to intrinsic and specific vulnerability: A case study in an agricultural setting (Southern Spain). Science of the Total Environment. 2014; 189–206.
[26]. Talebi A, Akbari Z. Investigation of ability of decision Trees model to estimate river suspended sediment (Case Study: Ilam Dam Basin). J. Sci. & Technol. Agric. & Natur. Resour. Water and Soil Sci. 2013; 17(63): 109-121. [Persian]
[27]. Fallahi M.R, Varvani H, Golian S. Precipitation prediction using tree regression model to flood control. Fifth National Conference on Watershed Management and Soil and Water Resources Management. 2012; Kerman, Iran. [Persian]
[28]. Ghafari G.A, Vafakhah M. Simulation of rainfall-runoff process using Artificial Neural Network and adaptive Neuro-Fuzzy Interface System (Case Study: Hajighoshan Watershed). Journal of Watershed Management Research. 2013; 4(8): 120-136. [Persian]
[29]. Yue, S, Wang C. Y. The Mann-Kendall test modified by effective sample size to detect trend in serially correlated hydrological series, Water Res.Manage. 2004; 18: 201-218.
[30]. Ghodoosi1 M, Morid S, Delavar M. Comparison of detrending methods for the temperature and precipitations time series. Journal of Agricultural Meteorology. 2013; 1(2): 32-45. [Persian]
[31]. Ahmadi F, Radmanesh F. Trend Analysis of Monthly and Annual Mean Temperature of the Northern Half of Iran Over the Last 50 Years. Journal of Water and Soil. 2014; 28(4): 855-865. [Persian]