ارزیابی کارایی چندین روش داده‌کاوی برای پیش‌بینی تبخیر(مطالعة موردی: ایستگاه سینوپتیک یزد)

نوع مقاله: مقاله پژوهشی

نویسندگان

1 دکتری علوم و مهندسی آبخیزداری، دانشکدة منابع طبعی و کویرشناسی، دانشگاه یزد

2 استاد دانشکدة منابع طبعی و کویرشناسی، دانشگاه یزد

چکیده

تبخیر یکی از پارامترهای اقلیمی مهم در مناطق خشک است و نقش مهمی را در مدیریت منابع آب بازی می‌کند، به همین جهت آگاهی از مقدار تبخیر و مدل‌سازی آن به عنوان یکی از متغیرهای مهم هیدرولوژیکی در تحقیقات کشاورزی و حفاظت آب و خاک از اهمیت زیادی برخوردار است. در دهه‌های اخیر روش‌های هوش مصنوعی در تخمین و پیش‌بینی پدیده‌های غیرخطی توانایی بالایی از خود نشان داده است. در این تحقیق از سه روش مهم داده‌کاوی شامل شبکة عصبی مصنوعی، شبکه‌های استنتاج فازی و درخت تصمیم رگرسیونی جهت پیش‌بینی تبخیر ماهانه در ایستگاه سینوپتیک یزد استفاده شد. برای این منظور از 8 متغیر هواشناسی در مقیاس ماهانه (متوسط کمینة دما، متوسط بیشینة دما، میانگین دما، ساعات آفتابی، سرعت باد، جهت باد، میانگین رطوبت نسبی و تبخیر) به عنوان ورودی مدل استفاده گردید. نتایج به‌دست‌آمده نشان داد هر سه مدل نامبرده قادرند با استفاده از پارامترهای اقلیمی مذکور به پیش‌بینی مقدار تبخیر ماهانه 12 ماه بعد از وقوع بپردازند ولی در میان سه مدل مورد استفاده، شبکة عصبی مصنوعی با ضریب همبستگی برابر با 97/0­r=، 1/5RMSE=­،3/36­MAE=­ و 48/0-­ME= بهترین کارایی را از خود نشان داد. همچنین نتایج نشان داد در پیش‌بینی تبخیر، تفاوت قابل‌ملاحظه‌ای در زمان استفاده از داده‌های خام و داده‌های نرمال شده وجود ندارد و پردازش داده‌ها تأثیر چندانی در بهبود نتایج مدل‌ها نخواهد داشت.

کلیدواژه‌ها


[1] Afkhami, H., Dastorani, M. T., Malekinezhad, H. and Mobin, M. H. (2010). Effect of climatic elements in Increasing the accuracy using artificial neural network on forcasting of drought in Yazd. Jornal of Sience and Technology of Agriculture and Natural Resources, Water and Soil Science, 51(14),157-170.
[2] Alizade, A. and Khalili, N. (2009). Estimation of angstrom coefficient anddeveloping a regressionequation for solar radiation estimation (case study:Mashhad). Jornal of Water and Soil. 23(1), 229-238.
[3] Azamathulla , H. Md., Chang, C. K., Ghani, A. Ab., Ariffin, J., Azazi Zakaria, N. and Abu Hasan, Z. (2009). An ANFIS-based approach for predicting the bed load for moderately sized rivers. Jornal of Hydro-environment Research, 3(1),35-44.
[4] Bayat, K. and Mirlatifi, M. (2009). Estimation of daily solar radiation using regression models and artificial neural network. Jornal of Agricultural Sciences and Natural Resources.16(3), 270-279.
[5] Breiman, L., Friedman, J., Olshen, R. and Stone, C. (1984). Classification and Regression Trees. Chapman & Hall/CRC Press, Boca Raton, FL.
[6] Bruton, J. M., McClendon, R. W. and Hoogenboom, G. (2000). Estimating daily panEvaporation with artificial neural network.Trans, ASAE, 43(2), 492-4962.
[7] Chari, M. M., Afrasiab, P., Piri, J. and Delbari, M. (2011). Predicting evaporation from a shallow water table using artificial neural network and simiulations of Vayazy. Jornal of Water Engineering, 8(4),11-20.
[8] Chauhan, S.and Shrivastava, R. K. (2009). Reference evapotranspiration forecasting using different artificial neural networks algorithms. Jornal of Civil Engineering, 36 (9),1491-1505.
[9] Chow, V., Maidment, D. and Mays, L. (1988). Applied Hydrology, N.Y.:McGraw-Hill Pub, New York.
[10] Dastorani, M. T., Habibipur, A., Ekhtesasi, M. R., Talebi, A. and Mahjobi, J. (2012). Evluation of Performance of the decision tree model on prediction of rainfall. Jornal of Iran-Water Resources Research,3(8),14.
[11] Ebrahimian, H., Liaghat, A. and Bazrafshan, M. (2011). Estimation of some climatic factors Using transfer functions. Jornal of Iran-Watershed Management Science and Engineering, 14(5),77-80.
[12] Gupta. B. (1992). Engineering Hydrology, Jain, India: N.C.
[13] Habibipur, A., Dastorani, M. T., Ekhtesasi, M. R. and Afkhami, H. (2011). Evaluation of the Effects of Data range Modification on Efficiency of Regression Decision Tree and Artificial Neural Networks for DroughtPrediction. Jornal of Watershed Management Research , 2(3), 63-79.
[14] Jang, J. S. R., Sun, C. T. and Mizutani, E. (1997). Neuro-Fuzzy and Soft Computing, A Computational Approach to Learning and Machine Intelligence, Prentice Hall, NJ, USA ISBN. 0-13-261066-3.
[15] Keskin, M. E. and Terzi, O. (2006). Artificialneural networks models of daily pan evaporation. Jornal of Hydrolgical Engineering, 11(1), 65–70.
[16] khoshhal, J. and mokarram, M. (2012). Model for Prediction of Evapotranspiration Using MLP Neural Network. Jornal of Enviromental Sciences, Volume 3( 3),1000-1009.
[17] Kim,S. and Kim, H.S. (2008). Neural networks and genetic algorithm approach for nonlinear evaporation and evapotranspiration modeling. Jornal of Hydrology, 351(3-4), 299–317.
[18] Kisi, O. (2006). Generalized regression neural networks for evapotranspiration modeling. Jornal of Hydrolgical Sciences, 51(6), 1092–1105.
[19] Kisi, O. and Tombul, M.(2013). Modeling monthly pan evaporations using fuzzy genetic approach. Jornal of Hydrology, 477(16), 203–212.
[20] Kumar, M., Raghuwanshi, N., Singh, R., Wallender, W. and Pruitt, W. (2002). Estimating Evapotranspiration using Artificial Neural Network. Jornal of Irrigation and Drainage Engineering, 128(4), 224–233.
[21] Kumar, P., Kumar, D., Jaipaul, A. and Tiwari, K.(2012). Evaporation Estimation Using Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference System Techniques. Jornal of Meteorology, 8(16), 88-97.
[22] Mahdavi, M. (2003). Applied Hydrology. 5nd Ed., Tehran university.
[23] Moghaddamnia, A., Gousheh, M. G., Piri, J., Amin, S. and Han, D. (2009). Evaporation estimation using artificial neural networks and adaptive neuro-fuzzy inference system techniques. Advances in Water Resources, 32(1), 88-97
[24] Mosaedi, A. and Ghobani sogh, M. (2010). Modification of RDI index Selecting the most appropriate distribution function In arid and semi-arid regions of Iran, 1st National Conference on Agrometeorology and Agricultural Water Management, Karaj.
[25] Nourani, V. and Sayyah Fard, M.(2012). Sensitivity analysis of the artificial neural network outputs in simulation of the evaporation process at different climatologic regimes. Jornal of Advances in Engineering Software ,47(1), 127–146.
[26] Piri, J. and Ansari, H. (2013). Daily pan evaporation modelling with ANFIS and NNARX. Iran Agricultural Research, 31(2), 51-64.
[27] Piri, J., Mohammadi, K., Shamshirband, S. and Akib, S. (2016). Assessing the suitability of hybridizing the Cuckoo optimization algorithm with ANN and ANFIS techniques to predict daily evaporation. Environmental Earth Sciences, 75(3), 246.
[28] Sayadi, H., Oulad ghfari,A., Faalian, A., and Sadroddini, A.A. (2010). Comparison of MLP and RBF neural networks for estimation of reference evapotranspiration. Journal of Soil and Water.19(1)1-12.
 [29] Serrano, M.L.A., Ruiz, A., Garcia, J.A., Anton, M. and Vaquero, J.M. (2005). Solar global radiation and sunshine duration in Extremadura (Spain). Jornal of Physica Scripta,118,24-28.
[30] Shafizade, F. and Mobin, M. H. (2009). Evaluation of drought in southern Iran Using the RDI index and Mann – Kendall, SecondNationalConference onWater, Behbahan.
[31] Shayannejad, M. (2007). Comparative accuracy of artificial neural networks and Penman - mantis methods in calculating Potential evapotranspiration. National Conference on Irrigation and Drainage Networks. Shahid chamran university,ahvaz.Iran
[32] Shiri ,J., Nazemi, A. H., Sadraddini, A., Landeras, G., Kisi, O., Fakheri Fard, A. and Marti, P.(2013). Global cross-station assessment of neuro-fuzzy models for estimating daily reference evapotranspiration. Jornal of Hydrology, 480, 46–57
[33] Singh,V. (1992). Elementary Hydrology, NJ, U.S.A.: Prentice Hall Inc.
[34] Sudheer, K. P., Gosain, A. K., Mohana Rangan, D. and Saheb, S. M.(2002). Modelling evaporation using an artificial neural network algorithm. Jornal of Hydrological Processes, 16, 3189–3202.
[35] Takagi, T. and Sugeno, M. (1985). Fuzzy identification of systems and its application to modeling and control, IEEE Transactions on Systems, Man, and Cybernetics,15, 116–132.
[36] Trajkovic, S., Todorovic, B. and Stankovic, M. (2003). Forecasting of Reference Evapotranspiration by Artificial Neural Networks. Jornal of Irrigation and Drainage Engineering, 129(6), 454–457.
[37] Traore, S., Wang,Y.M. and Kerh,T. (2010). Artificial neural network for modeling reference evapotranspiration complex process in Sudano-Sahelian zone. Jornal of Agricultural Water Management, 97 (5), 707–714.
[38] Tsakiris, G. and Vangelis, H. (2005). Establishing a drought index incorporating evapotranspiration. Jornal of European water, 9(10), 3-11.
[39] Tsakiris, G., Pangalou, D. and Vangelis, H. (2006). Regional Drought Assessment Base on the Reconnaissance Drought Index (RDI). Jornal of Water Resource Management, 21(5), 821-833.
[40] Xu, C.Y. and Singh, V. P. (1998). Dependence of evaporation on meteorological variables at different time-scales and intercomparison of estimation methods. Jornal of Hydrol. Proc, 12(3), 429–442.
[41] Zanetti, S.S., Sousa, E.F., Oliveira,V.P.S., Almeida, F.T. and Bernardo,S. (2007). Estimating evapotranspiration using artificial neural network andminimum climatological data. Jornal of Irrigation and Drainage Engineering, 133 (2), 83–89.
[42] Zare Abyaneh, H., Bayat Varkeshi, M., Marofi, S. and Amiri Chayjan, R. (2010). Evaluation of Artificial Neural Network and Adaptive Neuro Fuzzy Inference System in Decreasing of Reference Evapotranspiration Parameters. Jornal of Water and Soil, 24(2),297-305.