ارزیابی اثر متغیرهای تاثیرگذار بر پیش‌بینی سیلاب واریزه‌ای با استفاده از مدل شبکه بیزین

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشکده پردیس ابوریحان-دانشگاه تهران-تهران-ایران

2 دانشیار دانشگاه تهران

3 گروه مهندسی آبیاری و زهکشی، دانشکده پردیس ابوریحان، دانشگاه تهران، تهران، ایران

4 دانشگاه تهران-پردیس ابوریحان، تهران، ایران

چکیده

 
پیش­بینی سیلاب واریزه­ای جهت کاهش خسارات ناشی از آن از اهمیت ویژه­ای برخوردار است. هدف این تحقیق پیش­بینی غلظت رسوبات سیلاب (واریزه­ای و معمولی) توسط مدل­های شبکه بیزین و شبکه عصبی در حوضه‌های امامه، ناورود و کسیلیان است که به ترتیب در استان­های تهران، گیلان و مازندران واقع شده­اند. بدین­منظور، متوسط ارتفاع، شیب حوضه، مساحت حوضه، بارش فعلی، بارش پیشین (به مدت 3 روز قبل) و دبی 1 روز قبل به عنوان متغیرهای ورودی انتخاب شدند. سپس برای تعیین مؤثرترین عوامل بر غلظت رسوبات سیلاب، 32 سناریو ارزیابی شد. برای سناریو حاصل از کلیه عوامل منتخب، شاخص­های R2 و MAPE در مرحله آزمون، به ترتیب 97/0 و %55/8 برآورد گردید. ارزیابی اثر متغیرهای مختلف نشان داد مؤثرترین عوامل بر دقت پیش­بینی شبکه بیزین به ترتیب ارتفاع حوضه، بارش فعلی، دبی روز قبل، مساحت حوضه و بارش پیشین یک روز قبل می­باشند. شاخص­های R2 و MAPE برای این سناریو 91/0 و %01/11 است که به دلیل داشتن کمترین تعداد عوامل ورودی و بالاترین دقت به عنوان بهترین سناریو انتخاب گردید. مقایسه عملکرد مدل بیزین با مدل شبکه عصبی نشان داد مدل شبکه بیزین دقت پیش­بینی بالاتری دارد. مؤثرترین عوامل شناسایی شده می­تواند برای پیش­بینی سیلاب واریزه­ای در حوضه­های مشابه استفاده گردد.

کلیدواژه‌ها


عنوان مقاله [English]

Assessment of Effective Factors on the Forecasting of Debris Floods Using Bayesian Network Model

نویسندگان [English]

  • Mahsa Sheikh Kazemi 1
  • Mohammad Ebrahim Banihabib 2
  • Jaber Soltani 3
  • Abbas Roozbahani 3
  • Mitra Tanhapour 4
1 College of Abouraihan, University of Tehran, Tehran, Iran
2 Associate professor, University of Tehran
3 Department of Irrigation and Drainage Engineering, College of Abouraihan, University of Tehran, Tehran, Iran
4 University College of Abouraihan, University of Tehran, Tehran, Iran.
چکیده [English]

It is important to predict debris flood for reducing its damages. The aim of this study is the prediction of sediment concentration of debris floods and ordinary floods using bayesian network (BN) and artificial neural network (ANN) models in Ammameh, Navrood and Casilian basins which were located in Tehran, Gilan and Mazandaran provinces, respectively. Accordingly, average basin elevation (EL), average basin slope (S), watershed area (A), current day rainfall (R), antecedent rainfall (AR) of three-days ago and discharge of one-day ago were selected as input variables. Then, 32 scenarios were tested to determine the most effective factors on the sediment concentration of flood. For the scenario derived from all selected factors, indices R2 and MAPE in the test stage were obtained 0.97 and 8.55%, respectively. Assessment of the effect of different factors shows that the most effective factors on the BN model’s prediction accuracy are EL, R, PQ, A and AR one-day ago. Indices R2 and MAPE for this scenario were obtained 0.916 and 11.01%, respectively. It was selected as the best scenario because the least number of predictors and the highest accuracy. The most effective factors identified in this study can be used to predict debris flood in similar basins.

کلیدواژه‌ها [English]

  • Debris flood
  • Sediment Concentration
  • Bayesian network model
  • Artificial Neural Network
Aguilera PA, Fernández A, Fernández R, Rumí R, Salmerón A (2011) Bayesian networks in environmental modelling. Environmental Modelling & Software 26(12):1376-1388
Anbari MJ, Tabesh M, Roozbahani A (2017) Risk assessment model to prioritize sewer pipes inspection in wastewater collection networks. Journal of Environmental Management 190:91-101
Asadi H, Shahedi R, Sidle C, and Kalami Heris S M (2019) Prediction of suspended sediment by intelligence models using hydrologic and hydrogeomorphic data. Iran-Water Resources Research 15(3):105-119 (In Persian (
Banihabib M (1999) Hydraulic roughness of flow with high concentrations of sediment. In: 2nd Conference Hydraulic, 16-18 November, Tehran, Iran, 174-181 (In Persian)
Banihabib ME, Bahram E (2009) Experimental analyses of sedimentation in the slit dam Reservoir. In: World Environmental and Water Resources Congress, May 17-21, Kansas, Missouri, United States, 1-12
Banihabib ME, Forghani A (2017) An assessment framework for the mitigation effects of check dams on debris flow. Catena152:277-84
Banihabib M E, Masumi A (2008) Effect of high-concentrated sediment transport on inundation of rivers: Case study Masuleh Flood. In: Iranian Hydraulic Conference, Tehran, Iran, 166-173 (In Persian)
Banihabib ME (2002) Mud flow and debris. In: Proceeding of Conference of Prevent and Reduce of Flood Risks, Gorgan, Iran, 1-8 (In Persian)
Bromley J, Jackson NA, Clymer OJ, Giacomello AM, Jensen FV (2005) The use of Hugin® to develop Bayesian networks as an aid to integrated water resource planning. Environmental Modelling & Software 20(2):231-42
Chang T C, Wang Z Y, Chien Y H (2010) Hazard assessment model for debris flow prediction. Environmental Earth Sciences 60(8):1619-1630
Ebrahimy E, Rozbahany A, Kardan Moghadam H (2015) Analysis of uncertainty effective parameters on forecasting the groundwater level with Bayesian network approach. In: Shahid Beheshti Conference, 17-18 Oct, Tehran, Iran, 1-10 (In Persian)
Emamgholizadeh S, Kashi H, Marofpoor I, Zalaghi E (2014) Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models. International Journal of Environmental Science and Technology 11(3):645-656
Hassan-Esfahani L, Banihabib M E (2016) The impact of slit and detention dams on debris flow control using GSTARS 3.0. Environmental Earth Sciences 75(4):1-11
Hesar A S, Tabatabaee H, Jalali M (2012) Monthly rainfall forecasting using Bayesian belief networks. International Research Journal of Applied and Basic Sciences 3(11):2226-2231
Hirano M, Moriyama T, Kawahara K (1995) Prediction of the occurrence of debris flow and a runoff analysis by the use of neural networks. Journal of Natural Disaster Science 17(2):53-63
Hirano M, Harada T, Banihabib ME, Kawahara K (1997) Estimation of hazard area due to debris flow. In: Debris-Flow Hazards Mitigation: Mechanics, Prediction, and Assessment proceedings of First International Conference, 7 Aug, San Francisco, 697-706
Jakob M, Weatherly H (2003) A hydroclimatic threshold for landslide initiation on the North Shore Mountains of Vancouver, British Columbia. Geomorphology 54(3-4):137-56
Kern AN, Addison P, Oommen T, Salazar SE, Coffman RA (2017) Machine learning based predictive modeling of debris flow probability following wildfire in the intermountain Western United States. Mathematical Geosciences 49(6):717-35
Khakzad N, Khan F, Amyotte P (2011) Safety analysis in process facilities: Comparison of fault tree and Bayesian network approaches. Reliability Engineering & System Safety 96(8):925-32
Liang W J, Zhuang D F, Jiang D, Pan J J, Ren H Y (2012) Assessment of debris flow hazards using a Bayesian Network. Geomorphology 171:94-100
Lin J W, Chen C W, Peng C Y, (2012) Potential hazard analysis and risk assessment of debris flow by fuzzy modeling. Natural Hazards 64(1):273-282
Madsen AL, Lang M, Kjærulff UB, Jensen F (2003) The Hugin tool for learning Bayesian networks. In: European Conference on Symbolic and Quantitative Approaches to Reasoning and Uncertainty, 2 Jul, 594-605
Mohajerani H, Mosaedi A, Kholghi M, Meftah Helaghi M, Saadoldin A (2010) Introducing bayesian decision making networks and their application in water resources management. In: First National Conference on Coastal Water Resources Management, December 17-17, Sari, Iran (In Persian)
Nikolopoulos EI, Destro E, Bhuiyan MA, Borga M, Anagnostou EN (2018) Evaluation of predictive models for post-fire debris flow occurrence in the western United States. Natural Hazards and Earth System Sciences 18(9):2331-2343
Pearl J (1988) Probabilistic reasoning in intelligent systems: networks of plausible inference. California, Morgan Kaufmann, 57
Peng M, Zhang LM (2012) Analysis of human risks due to dam-break floods-part 1: A new model based on Bayesian networks. Natural Hazards 64(1):903-33
Peng S H (2016) Hazard ratings of debris flow evacuation sites in hillside communities of Ershui town ship, Changhua County Taiwan. Water 8(2):54
Sharghi E, Nourani V, and Behfar N (2019) Evaluation and application of ensemble AIbased models for estimating piezometric heads in earth fill dams. Iran-Water Resources Research 14(4):164-173 (In Persian)
Tanhapour M, Banihabib ME, Roozbahani A (2017) Bayesian networks model to study the effect of previous precipitation in the forecasting of debris floods occurrence in Alborz Region of Iran. Iran-Water Resources Research 13(4):118-131 (In Persian)
Xu W, Yu W, Jing S, Zhang G, Huang J (2013) Debris flow susceptibility assessment by GIS and information value model in a large-scale region, Sichuan Province (China). Natural Hazards 65(3):1379-92
Zhang H, Liu X, Cai E, Huang G, Ding C (2013) Integration of dynamic rainfall data with environmental factors to forecast debris flow using an improved GMDH model. Computers and Geosciences 56:23-31
Zhuang J, Cui P, Wang G, Chen X, Iqbal J, Guo X (2015) Rainfall thresholds for the occurrence of debris flows in the Jiangjia Gully, Yunnan Province, China. Engineering Geology 195:335-346