Application of Combined Artificial Neural Network and Wavelet Theory for Flow Rate Prediction and Comparing the Results with those of ANFIS (Case Study: The Kor River)

Document Type : Research Paper

Authors

• A. YazdanPanah
• N. Barahmand

Abstract

Due to the prolonged droughts in the recent decades, the importance of predicting the flow rate of surface water in rivers for water resources management increases. In this regard, the flow rates in the natural water ways, which is the most important supplement source for water in dam storages, are considered as the most vital factors in predicting surface water. In this study, a combined powerful model, using the artificial neural networks and wavelet theory, was developed to predict the flow rate of the Kor River at the Dashtbal hydrometric station located at the upstream of the Doroudzan Dam. The comparison of the results obtained from this model with those predicted by ANFIS inculcate the superiority of the former providing that the employed parameters are adequately selected. The ANFIS model with 4 Gaussian functions, and the Daubechies 4 wavelet within the third decomposition level occupied the 2^nd and 3^rd positions in accurate prediction after the combined model developed in this study.

Keywords

• Artificial Neural Networks
• Wavelet theory
• ANFIS
• the Kor River
• the Doroudzan Dam

20.1001.1.20086377.1394.8.26.3.3

---

References

1. ابراهیمی، ل، و غ، بارانی. 1384. معرفی شبیه  تلفیقی تبدیل موجکی و شبکه های عصبی برای پیش بینی خشکسالی حوزه های آبخیز سدها. مجموعه مقالات دومین کنفرانس سراگروه  آبخیزداری مدیریت منابع آب و خاک. 2359-2354ص.
2. رضوی، س.س، م، کارآموز، 1382. استفاده از شبکه های عصبی مصنوعی در پیش بینی ماهانه جریان رودخانه. مجموعه مقالات دهمین کنفرانس دانشجویی عمران، تهران، دانشگاه صنعتی امیرکبیر.
3. گل محمدی، م.ح، ح، صفوی. 1389. پیش بینی گروه  های زمانی تک متغیره آبشناسی با استفاده از سامانه فازی بر پایه شبکه عصبی تطبیقی پنجمین کنگره ملی مهندسی عمران. دانشگاه فردوسی مشهد.
4. وفاه خواه م، 1391. تخمین بده  از روی اشل با استفاده از شبکه عصبی مصنوعی. اولین کنفرانس ملی راهکارهای دستیابی به توسعه پایدار. پژوهشکده سوانح طبیعی ایران. تهران.
   1. Donoho, D.L. 1995, De-noising by softthresholding: IEEE Trans. on Inf. Theory, 41: 613-627.
   2. Folorunsho, J.O, and E.O. Iguisi. 2012. Application of adaptive neuro fuzzy inference system in River Kaduna discharge forecasting, J Appl Sci. Eng Technol 4: 4275-4283.
   3. Jang, J.S.R. 1993. ANFIS: Adaptive network based fuzzy inference system, IEEE Trans. Sys, Man. Cybernetics 23: 665-685. 
   4. Keshavarzi, A., and S. H. Nabavi. 2006. Dominant discharge the kor River. Fars Province, Iran. Tenth Int Water Technol Conf. IWTC10, Alexandria, Egypt.
   5. Li X, J. Ding. and H. Li. 1997. Wavelet analysis and its potential application to hydrology and water resources. J Sichuan Union Univ Eng. Sci.52-49.
   6. Mallat, S. 1999. A wavelet tour of signal processing. Academic Press, New York.
   7. Marvdasht, Kharameh and Zarghan Plains, Shiraz, Iran. World Appl. Sci J 22: 380-388.
   8. Nourani, V, M.T. Alami, and M.H. Aminfar 2009. A combined neural-wavelet model for prediction of watershed precipitation. Lighvanchai. Iran, Eng Appl. Artificial Intelligence, 16:1–12.
   9. Nourani V, Kisi O. and Komasi M. 2011. Two hybrid artificial intelligence approaches for modeling rainfall–runoff process. J Hydrol. 402:41–59.
   10. Partal, T., and O. Kisi. 2007. Wavelet and neruro-fuzzy conjunction model for precipitation forecasting. J Hydrol. 342: 199-212.
   11. Polikar, R. 1994. The wavelet tutorial. Second Edition, Part 1.
   12. Resnikov, H. L., and R. W. Wells, 1998. Wavelet analysis: The scalable structure of Information. Springer.
   13. Samani, S, F, Boustani, and M. H. Hojati 2013. Screen for heavy metals from groundwater samples from industrialized zones in.
   14. Takagi, T, and M. Sugeno. 1985. Fuzzy identification of systems its application to modelling and control. IEEE Trans. Sys, Man, and Cybernetics 15: 116-132. 
   15. Wang,  W.,  J, Ding, and H. Xiang. 2002. The  multi-time scale analysis of hydrological time  series with wavelet transform. J Sichuan Univ. 35:14-7.
   16. Wang, R, and T. Lee. 1998. A study on the wavelet model of upland watersheds and its application to hydrological estimation. Proc. Stat Meth Hydrolo Sys . 12–21.
   17. Zadeh, L. A. 1965. Fuzzy sets. Inform contr. 8: 338–35.
   18. Zounemat- Kermani, M, and M. Teshnehlab. 2007. Using adaptive neuro-fuzzy inference system for hydrological time series prediction, Elsevier.  8: 928-936.