مدل‌سازی رفتار جریان غلیظ با الگوریتم‌های یادگیری ماشین

فصلنامه علمی مهندسی منابع آب

دوره 15، شماره 54، آبان 1401، صفحه 29-42   

نوع مقاله: مقاله پژوهشی

شناسه دیجیتال (DOI): 10.30495/wej.2021.27117.2290

نویسندگان

مهدی درخشان نیا^{* 1}؛ مهدی قمشی²؛ سید سعید اسلامیان³؛ سید محمود کاشفی پور²

¹دانشجوی دکتری، گروه مهندسی عمران، واحد نجف آباد، دانشگاه آزاد اسلامی، نجف‌آباد، ایران.

²استاد، گروه سازه های آبی، دانشکده مهندسی آب و محیط زیست، دانشگاه شهید چمران اهواز، اهواز، ایران.

³استاد، گروه مهندسی عمران، واحد نجف آباد، دانشگاه آزاد اسلامی، نجف آباد، ایران. و گروه مهندسی آب، دانشکده کشاورزی، دانشگاه صنعتی اصفهان، اصفهان، ایران

تاریخ دریافت: 30 دی 1399،  تاریخ بازنگری: 26 اردیبهشت 1400،  تاریخ پذیرش: 10 شهریور 1400 

چکیده

چکیده
مقدمه :  جریان چگال یکی از عوامل موثر بر انتقال رسوبات به مخازن سدها می باشد. در این راستا یکی از روش های عملی برای کنترل رسوبات، ایجاد مانع در مسیر این جریان ها می­باشد.
روش :  در این تحقیق آزمایشگاهی، رفتار جریان چگالی تحت تأثیر موانع استوانه‌ای ساخته شده از چوب با قطر 1.5 سانتی‌متر و ارتفاع 30 سانتی‌متر (بیش از ارتفاع بدنه جریان چگالی) مورد ارزیابی قرار گرفت. بنابراین با در نظر گرفتن متغیرهایی مانند شیب کف، غلظت و دبی، مقادیر هد جریان غلیظ تعیین شد. همچنین در این مقاله از الگوریتم‌های یادگیری ماشین مانند سیستم استنتاج فازی-عصبی تطبیقی و شبکه عصبی مصنوعی برای مدل‌سازی نتایج استفاده شد.
یافته ­ها :  بر اساس نتایج، هد جریان نمک چگال با استفاده از الگوریتم‌های یادگیری ماشینی مانند سیستم استنتاج عصبی فازی تطبیقی و شبکه عصبی مصنوعی مدل‌سازی انجام شد و عملکرد این دو روش مقایسه شد. نتایج نشان داد که الگوریتم‌های یادگیری ماشین در مدل‌سازی هد جریان نمک چگالی مفید هستند و مقادیر رگرسیون سیستم استنتاج فازی عصبی تطبیقی برای داده های آموزش و آزمون 0.99 و رگرسیون شبکه عصبی مصنوعی به ترتیب 0.94 و 0.91 به دست آمد.
نتیجه ­گیری : با مقایسه این دو روش مشخص شد که سیستم استنتاج عصبی-فازی تطبیقی در مدل‌سازی درصد کاهش جریان سر چگالی نسبت به روش شبکه عصبی مصنوعی پیش‌خور مؤثرتر بوده است.

کلیدواژه‌ها

جریان‌ غلیظ؛ درصد کاهش هد؛ رسوب‌گذاری؛ سیستم استنتاج عصبی_ فازی تطبیقی؛ شبکه عصبی مصنوعی پیش‌خور

عنوان مقاله [English]

Modeling the Behavior of Density Current with Machine Learning Algorithms

نویسندگان [English]

Mehdi Derakhshannia¹؛ Mehdi Ghomeshi²؛ Seyed Saeid Eslamian³؛ Seyed Mahmood Kashefipour²

¹Ph.D. Candidate, Department of Civil Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran

²Professor, Department of Water Structures, Faculty of Water and Environmental Engineering, Shahid Chamran University of Ahvaz, Ahvaz, Iran

³Professor, Department of Civil Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran. and Department of Water Engineering, College of Agriculture, Isfahan University of Technology, Isfahan, Iran

چکیده [English]

Abstract
Introduction: Density current is one of the factors influencing the transfer of sediments to reservoirs of dams. One of the practical methods to control sediments is to build an obstacle in the path of these currents.
Methods: In this laboratory research, the behavior of the Density current under the effect of cylindrical obstacles made of wood with a diameter of 1.5 cm and a height of 30 cm (more than the height of the body of the Density current) was evaluated. Therefore, by considering variables such as floor slope, concentration and discharge, the values of the density current head were determined. Machine learning algorithms such as adaptive neural fuzzy inference system and artificial neural network were used to model the results.
Findings: Based on the results, the density salt flow head was modeled using machine learning algorithms such as adaptive fuzzy neural inference system and artificial neural network and the performance of these two methods were compared. The results showed that machine learning algorithms are useful in modeling the density salt flow head. And the regression of the adaptive neural fuzzy inference system for the training and test data was 0.99 and the regression of the artificial neural network was 0.94 and 0.91, respectively.
Conclusion: By comparing the two methods, it was found that the adaptive neural-fuzzy inference system is more effective in modeling the percent reduction of the head of Density current than the feed-forward artificial neural network method.

کلیدواژه‌ها [English]

Density current, head reduction percentage, sedimentation, adaptive neural-fuzzy inference system, feed-forward artificial neural network

مراجع

1.       Abbaspour A, Farsadizadeh D, Ghorbani MA (2013). Estimation of hydraulic jump on corrugated bed using artificial neural networks and genetic programming. Water Sci Eng 6:189–198.

2.       Abhari, M.N., Iranshahi, M., Ghodsian, M. and Firoozabadi, B., (2018). Experimental study of obstacle effect on sediment transport of turbidity currents. Journal of Hydraulic Research, 56(5), pp.618-629.

3.       Asghari Pari, S. A., Kashefipour, S. M., Ghomeshi,  M. (2017). An experimental study to determine the obstacle height required for the control of subcritical and supercritical gravity currents. European Journal of Environmental and Civil Engineering, 21(9): 1080-1092.

4.       Baghalian, S. and Ghodsian, M., (2020). Experimental study on the effects of artificial bed roughness on turbidity currents over abrupt bed slope change. International Journal of Sediment Research, 35(3), pp.256-268.

5.       Basson, G. R. (2009). Management of siltation in existing and new reservoirs. General report Q. 89, Proc. of the 23rd congress of the Int. Commission on Large Dams CIGBICOLD (vol. 2),

6.       Belvederesi, C., Dominic, J. A., Hassan, Q. K., Gupta, A., and Achari, G. (2020). Predicting river flow using an AI-based sequential adaptive neuro-fuzzy inference system. Water, 12(6), 1622.

7.       Beyer Portner, N., and Schleiss, A. (1998). Erosion des bassins versants alpins par ruissellement de surface (PhD thesis No. 1815, Communication No. 6) Constructions (LCH), Ecole Polytechnique Federale de Lausanne EPFL, Switzerland. effect of shear-induced lift force. Scientific Reports, 10(1), 1-17.

8.       Bishop, C. M. (2006). Pattern recognition and machine learning. Cambridge: Springer. and consequent coastal land loss. Marine Geology, 129(3–4), 189–195.

9.       Davoodi, L., and Shafaei Bajestan, M. (2011). Application of submerged plates in bed sediment load control of branched catchments from trapezoidal irrigation canals. Water and Irrigation Management, 1 (2): 59-71. (In Persian).

10.   Ebrahimzadeh, A., Zarghami, M., Nooranif, V. (2019). Overtopping risk management by system dynamics and Monte-Carlo simulations, Hajilarchay Dam of Iran. Water and Irrigation Management, 5 (1): 96-81. (In Persian).

11.   Eghbalzadeh, A., and Javan, M. (2011). Numerical simulation of a turbidity current flowing over a solid obstacle. In 2nd International Conference on Environmental Science and Development.

12.   Georgoulas, A. N., Angelidis, P. B., Panagiotidis, T. G., and Kotsovinos, N. E. (2010). 3D numerical modelling of turbidity currents. Environmental fluid mechanics, 10(6), 603-635.

13.   Goodarzi, D., Lari, K. S., Khavasi, E., and Abolfathi, S. (2020). Large eddy simulation of turbidity currents in a narrow channel with different obstacle configurations. Scientific Reports, 10(1), 1-16.

14.   He, Z., Zhao, L., Hu, P., Yu, C. and Lin, Y.T., (2018). Investigations of dynamic behaviors of lock-exchange turbidity currents down a slope based on direct numerical simulation. Advances in Water Resources, 119, pp.164-177.

15.   Houichi L, Dechemi N, Heddam S, Achour B (2013). An evaluation of ANN methods for estimating the length of hydraulic jumps in U-shaped channel. J Hydroinform 15:147–154.

16. Huang, S., Huang, W., and Shen, Q. (2020). Effects of Bottom Obstacle Structure on Density-Induced Flow. E&ES, 455(1), 012024.

17.   Jang J-SR (1993) ANFIS: Adaptive-network-based fuzzy inference system. IEEE Transactions on Systems,Man, and Cybernetics 23(3):665–685.

18.   Kale, S. (2020). Development of an adaptive neuro-fuzzy inference system (ANFIS) model to predict sea surface temperature (SST). Oceanological and Hydrobiological Studies, 49(4), 354-373.

19.   Khavasi, E., Afshin, H., and Firoozabadi, B. (2012). Effect of selected parameters on the depositional behaviour of turbidity currents. Journal of Hydraulic Research, 50(1), 60–69.

20.   20 Koohandaz, A., Khavasi, E., Eyvazian, A., and Yousefi, H. (2020). Prediction of particles deposition in a dilute quasi-steady gravity current by Lagrangian markers:

21.   Ghorban Moghadam, A., and Ghomshi, M. (2015). Laboratory study of controlling concentrated salt flow by means of cylindrical barriers. Iranian Journal of Water Research, 9 (3 (18 in a row)), 111-120.

22.   Lai, Y. G., Huang, J., and Wu, K. (2015). Reservoir turbidity current modeling with a two-dimensional layer-averaged model. Journal of Hydraulic Engineering, 141(12), 04015029.

23.   Marosi, M., Ghomeshi, M. and Sarkardeh, H., (2015). Sedimentation control in the reservoirs by using an obstacle. Sadhana, 40(4), pp.1373-1383.

24.   Mohammadi, B., Linh, N. T. T., Pham, Q. B., Ahmed, A. N., Vojteková, J., Guan, Y., ... and El-Shafie, A. (2020). Adaptive neuro-fuzzy inference system coupled with shuffled frog leaping algorithm for predicting river streamflow time series. Hydrological Sciences Journal, 65(10), 1738-1751.

25.   Naftchali, A.K., Khozeymehnezhad, H., Akbarpour, A. and Varjavand, P., (2016). Experimental study on the effects of artificial vegetation density on forehead of saline current flow. Ain Shams Engineering Journal, 7(2), pp.799-809.

26.   Nakajima, T. (2002). Laboratory experiments and numerical simulation of sediment-wave formation by turbidity currents. Marine Geology, 192(1-3), 105-121.

27.   Nasr-Azadani, M. M., and Meiburg, E. (2014). Influence of seafloor topography on the depositional behavior of bidisperse turbidity currents: A three-dimensional, depthresolved numerical investigation. Environmental Fluid Mechanics, 14(2), 319–342.

28.   Oehy, C. D., and Schleiss, A. J. (2007). Control of turbidity currents in reservoirs by solid and permeable obstacles. Journal of Hydraulic Engineering, 133(6), 637-648.

29.   Oehy, C. D., De Cesare, G., and Schleiss, A. J. (2010). Effect of inclined jet screen on turbidity current. Journal of Hydraulic Research, 48(1), 81-90.

30.   Pirnia, S. P., Wali Samani, J. and  monaem, M. J. (2012). Spatial and temporal study of salinity infiltration in tidal river using COHERENS model: a case study of Bahmanshir river. Journal of Water and Irrigation Management, 3 (1): 27-13. (In Persian).

31.   Rojas, R., (2013). Neural networks: a systematic introduction. Springer Science & Business Media.

32.   Schleiss, A. J., Franca, M. J., Juez, C., and De Cesare, G. (2016). Reservoir sedimentation. Journal of Hydraulic Research, 54(6), 595–614.

33.   Stanley, D. J. (1996). Nile delta: extreme case of sediment entrapment on a delta plain

34.   Soler, M., Colomer, J., Folkard, A., and Serra, T. (2020). Particle size segregation of turbidity current deposits in vegetated canopies. Science of The Total Environment, 703, 134784.

35.   Takagi T, Sugeno M (1985) Fuzzy identification of systems and its applications to modeling and control. IEEE Trans Syst Man Cybern 1:116–132.

36.   Wang, Z.-y., and Hu, C. (2009). Strategies for managing reservoir sedimentation. International Journal of Sediment Research, 24(4), 369–384.

37.   Wilson, R. I., and Friedrich, H. (2014). Dynamic analysis of the interaction between unconfined turbidity currents and obstacles. In 9th International Symposium on Ultrasonic Doppler Methods for Fluid Mechanics and Fluid Engineering, Strasbourg, France.

38.   Xu, J., Li, Y., Xuan, G., Melville, B.W. and Macky, G.H., (2020). Numerical simulation of turbidity current in approach channels with a closed end. Journal of Waterway, Port, Coastal, and Ocean Engineering, 146(5), p.04020036.