| تعداد نشریات | 285 |
| تعداد نشریات سازمان | 83 |
| تعداد شماره‌ها | 3,086 |
| تعداد مقالات | 34,444 |
| تعداد مشاهده مقاله | 48,945,322 |
| تعداد دریافت فایل اصل مقاله | 79,372,218 |
برآورد میزان آبشستگی مصالح غیرچسبنده و حساسیت سنجی عوامل موثر بر آن در پایاب جام پرتابه ای با استفاده از شبکه عصبی و مدل آزمایشگاهی | ||
| تحقیقات مهندسی سازه های آبیاری و زهکشی | ||
| دوره 23، شماره 86، مرداد 1401، صفحه 111-132 اصل مقاله (1.24 M) | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22092/idser.2022.358589.1508 | ||
| نویسندگان | ||
| اقبال خرمی1؛ محمد مهدی حیدری* 2؛ رسول قبادیان3 | ||
| 1دانشجوی دکتری سازه‌های آبی گروه مهندسی آب، پردیس کشاورزی و منابع طبیعی، دانشگاه رازی، کرمانشاه، ایران | ||
| 2استادیار گروه مهندسی آب، پردیس کشاورزی و منابع طبیعی، دانشگاه رازی، کرمانشاه، ایران | ||
| 3دانشیار گروه مهندسی آب، پردیس کشاورزی و منابع طبیعی، دانشگاه رازی، کرمانشاه، ایران | ||
| چکیده | ||
| پرتاب کننده های جامی بعنوان یکی از انواع سیستم‌های استهلاک انرژی مقرون به صرفه همواره مورد توجه مهندسین هیدرولیک بوده است. علیرغم استهلاک بخش قابل توجهی از انرژی جریان در اثر پراکنده شدن در هوا، برخورد جت جریان با بستر پایاب، موجب آبشستگی شده و ممکن است خطراتی را برای سازه در پی داشته باشد. تنوع روابط برآورد آبشستگی و در برخی شرایط وجود نتایج متفاوت و یا حتی متناقض، انتخاب یکی از روابط را دشوار نموده است. در این تحقیق عوامل موثر بر آبشستگی شناسائی و با بهره‌گیری از شبکه عصبی و شاخص‌های حساسیت، اثر آن‌ها بر عمق آبشستگی بررسی شد. بیشترین و کمترین شاخص حساسیت عمق آبشستگی مربوط به دبی در واحد عرض جریان و عمق آب پایاب است، به طوری که افزایش 10 درصدی این متغیرها به ترتیب باعث افزایش 5/8 درصد و کاهش 9/3 درصد عمق آبشستگی می‌شود. در این تحقیق یک رابطه برای تخمین عمق فرسایش نیز ارائه شد و با انجام 13 آزمایش آبشستگی بر مدل آزمایشگاهی ساخته شده، دقت رابطه بررسی شد. متوسط خطا در برآورد عمق آبشستگی توسط شبکه عصبی و رابطة پیشنهادی به ترتیب 3/9 و 8/9 است، که با توجه به پارامترهای آماری، دقت مناسبی برای تخمین عمق آبشستگی دارند. | ||
| کلیدواژه‌ها | ||
| تخمین عمق فرسایش؛ پرش اسکی؛ مدل شبکه عصبی؛ شاخص‌های حساسیت | ||
| عنوان مقاله [English] | ||
| Estimation of downstream scour in non-cohesive materials and sensitization of factors affecting it in ski-jump using neural network and laboratory model | ||
| نویسندگان [English] | ||
| Eghbal Khorami1؛ Mohammad Mehdi Heidari2؛ RASOOL Ghobadian3 | ||
| 1Razi university | ||
| 2Assistant Professor, Campus of Agriculture and Natural Resources, Razi University, Kermanshah, Iran. | ||
| 3Razi UNIVERSITY | ||
| چکیده [English] | ||
| Introduction Ski jump as one of the energy dissipater systems at the end of weir has been considered by hydraulic engineers due to its cost-effectiveness. Despite the dissipation of a significant portion of the flow energy, the bed is subject to scouring and can pose hazards to the structure. In recent decades, equations have been proposed to estimate the scour rate of this structure, each of which has a specific problem. For example, in some of these relationships, there are some factors that affect scouring, such as the bucket radius or the lip angle of bucket. The equation of Azmatullah et al. Is one of the most complete relationships mentioned in most sources. In this equation, the depth of scour increases with increasing mean sediment size of bed materials, which contradicts previous research. In the present study, the neural network was trained to estimate the depth of scour and the transmission functions, number of layers and type of network training were optimally selected. Using neural network and sensitivity analysis, the effect of different variables on the scour depth was determined. A high-precision relationship was presented to predict the scour depth using explicit genetic programming Methodology In order to investigate the scour depth in flip-bucket, a laboratory model with a width of 0.5 m, height of 45 cm, length of 59 cm, bucket radius of 15 cm and lip angle of bucket of 45 degrees was constructed. The flow rate through the laboratory model is 7 to 21.2 lit/s, head between upper reservoir water level and tail water level is 0.3 to 0.38 m, the tail water depth is 0.028 to 0.1 m and the D50 bed materials is 4.3 mm. In the present study, in addition to the measured laboratory data, used data related to the research of Azmatullah et al. In this study, neural network has been used to predict the depth of scour hole. The optimal structure of the neural network is affected by variables such as the number of neurons in the latent layer, the stimulus functions between the neurons, and the number of latent layers. In the present study, the optimal structure of the neural network was determined by trial and error. It should be noted that in this study, NeuroSolutions software was used for the architecture of the multilayer neural network of perceptron. Results Stimulus functions, number of hidden layers and type of training are factors that affect the accuracy of the model. In this study, the optimal structure of the neural network was determined by considering the variables affecting network accuracy and using various laboratory data. The results show that if the number of hidden layers of a number is a function of hyperbolic tangent transfer and network training type, Levenberg-Marquardt, the accuracy of the neural network in estimating the maximum scour depth is better. Also in this study, the scour depth sensitivity index to discharge per unit width flow, upstream head, lip angle of bucket, bucket radius, tail water depth and mean sediment size were calculated. The sensitivity index of scour depth to discharge per unit of flow width, fall height and lip angle of bucket is more than zero and this shows that with the increase of these variables, the depth of the scour hole increases. In order to reduce the amount of scour holes, the depth of the tail water can be increased or a riprap can be used. According to the calculated sensitivity indices, if the depth of tail water or particle diameter increases by 10%, the maximum depth of the scour hole decreases by about 3.9 and 1.3, respectively. Depending on the economic considerations and the feasibility of each case, the appropriate option can be selected. In this study, using the data of Azmatullah et al. And using GEP software, a relationship was presented to estimate the maximum scour rate in the flip bucket, with an average error rate of 9.8% which shows the model's ability to estimate scour. Conclusions In this study a neural network model was given to estimate the depth of the scour hole in the flip bucket. The results show that if the number of hidden layers of a number is a function of hyperbolic tangent transfer and the type of network training is Levenberg-Marquardt, the neural network accuracy in estimating the maximum scour depth is better. Also in this study, the sensitivity index of scour depth to input variables was calculated. The greatest effect on the scour depth is related to the water discharge per unit width rate, if the flow rate increases by 10%, the maximum depth of the scour hole will increase by 8.5%. In this study, a relation was presented to estimate the depth of the erosion hole, the accuracy of which is evaluated very well according to the criteria. | ||
| کلیدواژه‌ها [English] | ||
| Estimation of scour depth, ski jump, neural network model, sensitivity indicators | ||
| مراجع | ||
|
Amanian, N. (1995). Scour below a flip bucket spillway (Ph. D. Thesis), Utah State University, Logan, Utah, USA.
Annandale, G. (1995). Erodibility. Journal of hydraulic research. 33(4), 471-494.
Annandale, G.W. (2006). Scour Technology. New York: McGraw-Hill.
Azamathulla, H. M., Deo. M., & Deolalikar, P. (2006). Estimation of scour below spillways using neural networks. Journal of Hydraulic Research, 44(1), 61-69.
Azamathulla, H. M., Deo. M., & Deolalikar, P. (2005). Neural networks for estimation of scour downstream of a ski-jump bucket. Journal of Hydraulic Engineering, 131(10), 898-908.
Azamathulla, H. M. & Guven, A. (2012). Gene-expression programming for flip-bucket spillway scour. Water Science and technology, 65(11), 1982-1987.
Azamathulla, H. M., Deo. M., & Deolalikar, P. (2008). Alternative neural networks to estimate the scour below spillways. Advances in Engineering Software, 39(8), 689-698.
Bollaert, E. (2002). Transient Water Pressures in Joints and Formation of Rock Scour due to High-Velocity Jet Impact (PhD Thesis), Swiss Federal Institute of Technology, Lausanne, Switzerland.
Castillo, L. G. & Carrillo, J. M. (2017). Comparison of methods to estimate the scour downstream of a ski jump. International Journal of Multiphase Flow, 92, 171-180.
Haghiabi, A. (2017). Estimation of scour downstream of a ski-jump bucket using the multivariate adaptive regression splines. Scientia Iranica, 24(4), 1789-1801.
Haykin, S. (1994). Neural networks. New York: Mac millan.
Martins, R. B. F. (1973). Contribution to the Knowledge on the Scour Action of Free Jets on Rocky River Beds. in Transactions of the 11th International Congress on Large Dams, Vol. II, Question 41, Reply 44, Madrid. Spain.
Martins, R. B. F. (1975). Scouring of Rocky River Beds by Free Jet Spillways. International Water Power and Dam Construction, 27(5), 152-153.
Mason, P. J. & Arumugam, K. (1985). Free jet scour below dams and flip buckets. Journal of Hydraulic Engineering, 111(2), 220-235.
Moriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel, R.D. & Veith, T.L. (2007). Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3), 885-900.
Movahedi, A., Kavianpour, M., & Yamini, A. (2019). Experimental and numerical analysis of the scour profile downstream of flip bucket with change in bed material size. ISH Journal of Hydraulic Engineering, 25(2), 188-202.
Nayak, Satyaji Rao, Y. R., Sudheer, K. P. (2006). Groundwater level forecasting in a shallow aguifer using artifical neural. Water Resources Management, 2(11), 77-99.
Parsaie, A., Azamathulla, H.M., & Haghiabi, A.H. (2017). Physical and numerical modeling of performance of detention dams. Journal of Hydrology, https://doi.org/10.1016/j.jhydrol.2017.01.018.
Parsaie, A., Dehdar-Behbahani, S. & Haghiabi, A.H. (2016). Numerical modeling of cavitation on spillway’s flip bucket. Frontiers of Structural and Civil Engineering, 10 (4):438-444.
Parsaie, A. & Haghiabi, A. H. (2021). Hydraulic investigation of finite crested stepped spillways, Water Supply, 21 (5): 2437–2443.
Rumelhart, D. E., Hinton, G. H., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533-536.
Veronese, A. (1937). Erosioni de Fondo a Valle di uno Scarico. Annali dei Lavori Publicci, 75(9), 717-726.
| ||
|
آمار تعداد مشاهده مقاله: 1,583 تعداد دریافت فایل اصل مقاله: 259 |
||