کاربرد مدل‌های هوش مصنوعی و سری زمانی در تخمین رواناب (مطالعه موردی: قسمتی از حوضه آبریز رودخانه هلیل)

محورهای موضوعی : برگرفته از پایان نامه

الهه فرودی صفات ¹ , محمد مهدی احمدی ² , کورش قادری ³ , سودابه گلستانی کرمانی ⁴

1 - فارغ التحصیل مقطع کارشناسی ارشد رشته سازه‌های آبی، بخش علوم و مهندسی آب، دانشگاه شهید باهنر کرمان، کرمان، ایران
2 - دانشیار بخش علوم و مهندسی آب دانشگاه شهید باهنر کرمان، کرمان، ایران
3 - دانشیار بخش علوم و مهندسی آب دانشگاه شهید باهنر کرمان، کرمان، ایران
4 - استادیار بخش علوم و مهندسی آب دانشگاه شهید باهنر کرمان، کرمان، ایران

تاریخ دریافت : 1400/09/13 تاریخ پذیرش : 1402/08/24 تاریخ انتشار : 1402/08/01

کلید واژه: مدیریت منابع آب, بارش- رواناب, خودهمبسته با میانگین متحرک یکپارچه (ARIMA), شبکه عصبی مصنوعی (ANN), کنترل گروهی داده‌ها (GMDH),

چکیده مقاله :

چکیده
مقدمه: پیش­بینی دقیق رواناب و سیلاب برای جلوگیری از خسارتهای جانی و مالی یکی از چالش برانگیزترین کارها در مطالعات هیدرولوژیکی یک منطقه می باشد. از این رو، توسعه مدل های دقیق پیشبینی از قبیل روش های هوش مصنوعی مورد توجه بیشتر محققین قرار گرفته است.
روش­: در این تحقیق به بررسی کارآیی 3 مدل ANN، GMDH و ARIMA جهت شبیه ­سازی سیلاب قسمتی از حوضه رودخانه هلیل رود در استان کرمان پرداخته شد. مدل ANN یک روش مدل­سازی غیرخطی است که به مرور عملکرد خود را بهبود می ­بخشد. GMDH یک مدل هوش مصنوعی با ویژگی­ های قابلیت خودسازماندهی اکتشافی است که در انتهای آن سیستمی پیچیده با عملکرد مطلوب شکل می­ گیرد. کد نوشته شده ARIMA به ساخت مدلی برای توصیف ساختار داده ­ها و سپس پیش­بینی سری زمانی می ­پردازد. داده­های ورودی به مدل­ های فوق شامل دبی، بارش، دما، باد و رطوبت ماهانه بودند و مقادیر رواناب شبیه ­سازی شده با مقادیر مشاهداتی مقایسه گردید.
یافته ­ها: جهت ارزیابی دقت مدل­ها از شاخص ­های آماری RMSE، MSD و MAE استفاده شد و نتایج بدست آمده نشان داد مدل ((m³/s)042/0RMSE=، ²(m³/s)001/0MSD= و (m³/s)027/0MAE=) ANN توانست رواناب را با دقت بالاتری  در مقایسه با مدل (068/0RMSE=، 005/0 MSD= و 056/0MAE=) GMDH و سری زمانی (096/0RMSE=، 009/0MSD= و 063/0MAE=) ARIMA در حوضه مورد مطالعه برآورد کند. میانگین خطا در تخمین رواناب با مدل ANN در مقایسه با مقادیر تخمین زده شده با مدل GMDH و ARIMA به ترتیب 23/38 و 25/56 درصد کاهش یافت.
نتیجه­ گیری: باتوجه به نتایج بدست آمده در این مطالعه، مدل شبکه عصبی مصنوعی به سبب توانایی ساختاری مناسب جهت پیدا کردن رابطه غیرخطی بین داده های ورودی و خروجی، توانسته است کارایی بهتری نسبت به دو مدل دیگر از خود نشان دهند. 

چکیده انگلیسی:

Abstract
Introduction: Accurate forecasting of runoff and flooding to avoid human and financial losses is one of the most challenging tasks in hydrological studies of a given locale. Therefore, researchers have paid more attention to the development of accurate flood forecasting models, including the use of artificial intelligence methods.
Methods: In this investigation, the efficiency of 3 models, ANN, GMDH and ARIMA, has been investigated in order to simulate the flood of a part of Halil river basin in Kerman province. ANN model is a non-linear modeling method that improves its performance over time. The GMDH composed code is an artificial intelligence model with exploratory self-organizing features, at the conclusion of which a complex system with optimal performance is formed. Composed ARIMA code builds a model to describe the structure of the data and then predict the time series. The input data to the above models included discharge, precipitation, temperature, wind and monthly humidity, and the simulated runoff values ​​were compared with the observed values.
Findings: In order to evaluate the accuracy of the models in this research, statistical indices were used and the results showed that the ANN model (RMSE=0.042, MSD=0.001, MAE=0.027) had the possibility to estimate the runoff with higher accuracy compared to the GMDH model (RMSE=0.068, MSD=0.005, MAE=0.056) and the ARIMA time series (RMSE=0.096, MSD=0.009, MAE=0.063) in the studied basin. The mean error in runoff estimation with ANN model has been reduced by 38.23% and 56.25%, respectively, compared to the values estimated with GMDH and ARIMA models. According to the results obtained in this study, the artificial neural network model has been able to show a better performance than the other two models in predicting the outputs due to its suitable structural ability to find the nonlinear relationship between the input and output data.

منابع و مأخذ:

1.       Lee, E.H., Kim, J.H. 2018. Development of a Flood-Damage-Based Flood Forecasting Technique. Journal of Hydrology, 563, pp 181-194.

2.       Jain, SH.K., Mani, P., Jain, S.K., Prakash, P., Singh, V.P., Tullos, D., Kumar, S., Agarwal, S.P., Dimri, A.P. 2018. A Brief review of flood forecasting techniques and their applications. International Journal of River Basin Management, 16(3), pp 329-    344.

3.       Le, X.H., Ho, H.V., Lee, G., Jung, S. 2019. Application of Long Short-Term Memory (LSTM) Neural Network for Flood Forecasting. Water, 11(7), 1387.

4.       Li, Y., Shi, H., Liu, H. 2020. A hybrid model for river water level forecasting: Cases of Xiangjiang River and Yuanjiang River. China. Journal of Hydrology, 587, pp 1-13.

5.       Wu, W., Emerton, R., Duan, Q., Wood, A., W., Wetterhall, F., Robertson, D., E. 2020. Ensemble flood forecasting: Current status and future opportunities. WILEY, 3(29), pp 1-32.

6.       Jun, C.L., Mohamed, Z.S., peik, A.L., Razali, S.F., Sharil, S. 2016. Flood forecasting model using empirical method for a small catchment area. Journal of Engineering Science and Technology, 11(5), pp 666-672.

7.       Rodriguez Rivero, C., Patiño, H.D., Pucheta, J., Sauchelli, V. 2016. A new Approach for Time Series Forecasting: Bayesian Enhanced by Fractional Brownian Motion with application to rainfall series. International Journal of advanced Computer Science and Applications, 7(2), pp 1-8.

8.       Kumar, S., Roshni, T., Himayoun, D. 2019. A Comparison of Emotional Neural Network (ENN) and Artificial Neural Network (ANN) Approach for Rainfall-Runoff Modelling. Civil Engineering Journal, 5(10), pp 2120-2130.

9.       Pournemat Roudsari, A., Qaderi, K., Karimi-Googhari, SH. 2014. Rainfall Runoff Modeling using Group Method of Data Handling (GMDH) and Artificial Neural Network (ANN) In Polrood Basin. Journal of Watershed Management Research, 5(10), pp 68-84. [In Persian].

10.   Isazadeh, M. Ahmadzadeh, H., Ghorbani, M.A., Fazelifard, M.H. 2018. An Assessment of Time Series and Autoregressive Artificial Neural Network Models, Support Vector Machine and Gene Expression Programming Models Performance in Monthly River Flow Simulation (Case Study: Kherkherechi River Basin).  Irrigation Sciences and Engineering. 40(4), pp 91-107. [In Persian].

11.   Behyan Motlagh, S., Honarbakhsh, A., Abdolahi, KH., Pajouhesh, M. 2020. Evaluation of the Efficiency of Time Series and Fuzzy Models in Monthly Discharge Modeling (Case Study: Kohsukhteh Watershed). Hydrogeomorphology, 6(21), pp 65-86. [In Persian].

12.   Tsakiri, K., Marsellos, A., Kapetanakis, S. 2018. Artificial Neural Network and Multiple Linear Regression for Flood Prediction in Mohawk River, New York. Water, 10(9), 1158.

13.   Tabbussum, R., Dar, A.Q. 2020. Comparative analysis of neural network training algorithms for the flood forecast modelling of an alluvial Himalayan river. J Flood Risk Management, 13: e12656.

14.   Dodangeh, E., Panahi, M., Rezaie, F., Lee, S., Tien Bui, D., Lee, C.W., Pradhan, B. 2020. Novel hybrid intelligence models for flood-susceptibility prediction: Meta optimization of the GMDH and SVR models with the genetic algorithm and harmony search. Journal of Hydrology, 590, 125423.

15.   Dhunny, A.Z., Seebocus, R.H., Allam, Z., Chuttur, M.Y., Eltahan, M., Mehta, H. 2020. Flood Prediction using Artificial Neural Networks: Empirical Evidence from Mauritius as a Case Study. Knowledge Engineering and Data Science, 3(1), pp 1-10.

16.   Bano, P., Singh, R., Aggarwal, G. 2021. Forecasting of Flood in Upper Yamuna Basin by Using Artificial Neural Network and Geoinformatics Techniques & Learning. Ilkogretim Online, 20(5), pp 3008-3021.

17.   Mandal, S., Biswas, S. 2021. Runoff Prediction Using Artificial Neural Network and SCS-CN Method: A Case Study of Mayurakshi River Catchment, India. In Water Security and Sustainability: Proceedings of Down to Earth 2019, Springer Singapore, pp 27-42.

18.   Khodakhah, H., Aghelpour, P., Hamedi, Z. 2022. Comparing linear and non-linear data-driven approaches in monthly river flow prediction, based on the models SARIMA, LSSVM, ANFIS, and GMDH. Environ Sci Pollut Res 29, 21935–21954.

19.   Motamednia, M., Karimi Zarchi, K., Nohegar, A., Saberi Anari, M., Malekian, A. 2019. An Assessment the Performance of Genetic Programming and Auto Regresive Moving Average on the Daily Discharge Prediction (Case study: The Amameh Watershed). Watershed Management Research, 32(2), pp 2-18. [In Persian].

20.   Azizi, H., Montazeri, M. 2015. Anticipated Monthly Temperatures for Selected Stations in Isfahan Province Using Artificial Neural Network Multi-Layer Perceptron. GeoRes, 30(3), 241-258. [In Persian].

21.   Abrishami Moghaddam, H., Abdoli, GH., Ahrari, M., Dolatabadi, S. 2012. Applying GMDH Algorithm for rule extraction from the oil price behavior.  Energy Economics Review, 9(32), pp 147-168. [In Persian].

22.   Eivani, Z., Ahmadi, M.M., Qaderi, K. 2016. Estimation of Suspended Sediment Load Concentration in River System using Group Method of Data Handling (GMDH). Journal of Watershed Management Research, 7(13), pp 218-229. [In Persian].

23.   Zhongda, T., Shujiang, L., Yanhong, W., Yi, S. (2017). A prediction method based on wavelet transform and multiple models fusion for chaotic time series. Chaos. Solitons and Fractals, 98, PP 158-172.

24.   Siami, Namini, S., Tavakoli, N., Siami, Namin, A. 2018. A Comparison of ARIMA and LSTM in Forecasting Time Series. 17th IEEE International Conference on Machine Learning and Applications, pp 1394-1401.

25.   Mehdiabadi, M., Mohammadi pour, R. 2019. Determining the nonlinear effect of the money market interest rate on the Tehran stock exchange by the means of generalized autoregressive conditional heteroskedasticity (GARCH) model and smooth transition regression (STR) model. Financial Engineering and Securities Management (Portfolio Management). 10(40), pp 126-151. [In Persian].

26.   Mazumder, M. T. R., Gupta, B. C. 2021. Flood Forecasting Using Artificial Neural Network (ANNs): A Case study of Jamuna River. In: AGU Fall Meeting 2021, held in New Orleans, LA, 13-17 December 2021.

27.   Sahoo, B., Nanda, T., Chatterjee, C. 2022. Flood Forecasting Using Simple and Ensemble Artificial Neural Networks. In: Pandey, A., Chowdary, V.M., Behera, M.D., Singh, V.P. (eds) Geospatial Technologies for Land and Water Resources Management. Water Science and Technology Library, vol 103. Springer, Cham.

28.   Wong, W. M., Subramaniam, S. K., Feroz, F. S., Ai Fen Rose, L. 2022. Short-term Water Level Forecast Using ANN Hybrid Gaussian-Nonlinear Autoregressive Neural Network. International Journal of Integrated Engineering, 14(4), 425–437.

29.   Ziari, H., Sobhani, J., Ayoubinejad, J., Hartmann, T. 2015. Prediction of IRI in short and long terms for flexible pavements: ANN and GMDH methods, International Journal of Pavement Engineering.

Rosamma, C.N. 2022. Applicability of Soft Computational Models for Integrated Water Resource Management. In: Modeling and Simulation of Environmental Systems. 1st Edition.