ارزیابی مقایسه‌ای مدل عددی و شبکه عصبی مصنوعی در شبیه‌سازی کمی و کیفی آبخوان دشت نجف‌آباد

نویسندگان

1 دانشجوی دکترای مدیریت منابع آب، دانشکده مهندسی عمران و معماری، دانشگاه شهید چمران هواز

2 استاد مدیریت منابع آب، دانشکده مهندسی عمران و معماری، دانشگاه شهید چمران اهواز

3 استاد مدیریت منابع آب، دانشکده مهندسی عمران، دانشگاه صنعتی اصفهان

چکیده

هدف از انجام این مطالعه، بررسی کارایی مدل شبکه عصبی (ANN) در مدل‌سازی کمی و کیفی منابع آب زیرزمینی می‌باشد. بدین منظور، با استفاده از گزینه‌‌های MODFLOW وMT3DMS نرم افزار GMS v.10، از نظر کمی و کیفی، آب موجود در آبخوان دشت نجف آباد، واقع در حوضه آبریز گاوخونی در فلات مرکزی ایران، مدلسازی شد. بعد از واسنجی و صحت‌سنجی مدل در یک دوره مطالعه 11 ساله، محدوده تغییر ضریب هدایت هیدرولیکی بین 5/0 تا 16 (متر در روز)، آبدهی ویژه بین 023/0 تا 113/0 و ضریب پخشیدگی طولی بین 5/7 تا 2/18 (متر) بدست آمد. سپس، منطقه مورد مطالعه به دو ناحیه مجزا تقسیم و برای هر کدام یک مدل شبکه عصبی، طراحی شد. در ادامه، با استفاده از آمار 20 سال آبی و به کمک بهینه‌ساز الگوریتم ژنتیک، پارامتر‌های بهینه شبکه عصبی تعیین گردید. نهایتا، مقادیر مشاهده‌ای تراز متوسط سطح ایستابی و میانگین غلظت TDS با مقادیر محاسباتی توسط مدل عددی و شبکه عصبی، برای دو سال آبی متوالی 94-1393 و 95-1394، با یکدیگر مقایسه شدند. نتایج نشان داد که شبکه عصبی به خوبی قادر است رفتار کمی و کیفی سیستم آب زیرزمینی را شبیه‌سازی کند و می‌تواند به عنوان جایگزین مناسبی برای مدل عددی در اتصال به مدل‌های مدیریتی، استفاده شود.

کلیدواژه‌ها


عنوان مقاله [English]

Comparative Evaluation of Numerical Model and Artificial Neural Network for Quantity and Quality Simulation of Najafabad Aquifer

نویسندگان [English]

  • Masume zare 1
  • Hamid Reza Ghafouri 2
  • Hamid Reza Safavi 3
1 PhD Student of Water Resource Management, Faculty of Civil Engineering and Architecture, Shahid Chamran University of Ahvaz, Ahvaz, Iran
2 Professor of Water Resource Management, Faculty of Civil Engineering and Architecture, Shahid Chamran University of Ahvaz, Ahvaz, Iran
3 Prof. of Water Resource Management, Dept. of Civil Engineering, Isfahan University of Technology, Isfahan, Iran
چکیده [English]

The aim of this study was to investigate the efficiency of the neural network model in quantitative and qualitative modeling of groundwater resources. For this purpose, the groundwater of the Najafabad aquifer located in Gavkhoni basin at the central plateau of Iran, was modeled using MODFLOW and MT3DMS modules of GMS v. 10 software. After calibrating and validating the model for a 11 years time period, the ranges of hydraulic conductivity, specific yield and longitudinal dispersivity coefficient were found to be 0.5-16 (m day-1), 0.023-0.113 and 7.5-18.2 (m), respectively. Then the study area divided into two sub-regions and the ANN model was designed for each of the sub-regions. Afterwards, the optimal parameters of the ANN models were determined using the 20-year dataset of water year and the genetic algorithm optimization model. Finally, calculated values relevant to the average level of groundwater and the mean concentration of TDS, which were acquired by the ANN model and the numerical model, were compared with the observed values from 2014 to 2016. Results showed that the neural network model is capable in simulating the quantitative and qualitative treatment of the groundwater system and can be used as a suitable alternative for the numerical model linking the management models.

کلیدواژه‌ها [English]

  • Artificial neural network
  • Genetic algorithm
  • MODFLOW
  • MT3DMS
  • Simulation
Alaviani F, Sedghi H, Moghaddam AA and Babazadeh H, 2018. Adopting GMS–PSO model to reduce groundwater withdrawal by integrated water resources management. International Journal of Environmental Research 12(5): 619-629.
Anonymous, 2010. Zayandehrood River Basin Report. Isfahan, Iran. (In Persian with English abstract)
Bahreini GR and Safavi HR, 2009.Water resources interaction modeling using satellite images and GIS techniques. Iran Water Resources Research 4(3):14-26. (In Persian with English abstract)
Chitzasan M and Sedghi MM, 2005. Physical modelling study of free and submerged flow on the obstructions in rivers with ten modes of transverse waves. 23th Earth Science Gathering. Tehran, Iran. (In Persian with English abstract)
Coppola EA, Rana AJ, Poulton MM, Szidarovszky F and Uhl VW, 2005. A neural network model for predicting aquifer water level elevations. Groundwater 43(2): 231-241.
Eshelman LJ and Shaffer DJ, 1993. Real-coded genetic algorithms and interval schemata. Foundations of Genetic Algorithms 2: 187-202. 
Kalantari M, 2016. Conjunctive management of groundwater and surface water using Honey-Bee mating algorithm. MSc Thesis, Isfahan University of Technology, Iran. (In Persian with English abstract)
Kardan Moghadam H and Banihabib ME, 2017. Investigation of interference of salt water in desert aquifers (case study: South Khorasan, Saryan aquifer). Water and Soil 31(3): 673-688. (In Persian with English abstract)
Kolahdoozan A, Dinpajooh Y, Mirabbasi Najafabadi R, Asadi E and Darbandi S, 2015. Effect of Zayandehrood river dry becoming on Najafabad aquifer oscillations. Iranian Journal of Soil and Water Research 46(1): 81-93. (In Persian with English abstract)
Lalehzari R and Abbaslou H, 2016. Simulating the effect of optimal water allocation on groundwater in monthly stress periods (Baghmalek plain, Khuzestan province). Water and Soil Science-University of Tabriz, 26(4.1): 307-320. (In Persian with English abstract)
Lalehzari R and Kerachian R, 2017. Quantity and quality modeling of groundwater flow in Shahrekord aquifer. Aquifer and Ganat 1(1): 26-37. (In Persian with English abstract)
Nadiri AA, Fijani E, Tsai FTC and Asghari Moghaddam A, 2013. Supervised committee machine with artificial intelligence for prediction of fluoride concentration. Journal of Hydroinformatics 15(4): 1474-1490.
Nadiri AA, Sadeghi Aghdam F and Asghari Moghaddam A, 2015. Using intelligence committee fuzzy logic model to estimate arsenic in water resources of Sahand dam basin. Geosiences Scientific Quarterly Journal 25(97): 397-406. (In Persian with English abstract)
Nayak PC, Rao YS and Sudheer KP, 2006. Groundwater level forecasting in a shallow aquifer using artificial neural network approach. Water Resources Management 20(1): 77-90.
Peralta RC, Forghani A and Fayad H, 2014. Multiobjective genetic algorithm conjunctive use optimization for production, cost, and energy with dynamic return flow. Journal of Hydrology 511: 776-785.
Sadeghfam S, Hassanzadeh Y, Khatibi R, Nadiri AA and Moazamnia M, 2019. Groundwater remediation through pump-treat-inject technology using optimum control by artificial intelligence (OCAI). Water Resources Management 33(3):1123-1145. 
Safavi HR and Enteshari S, 2016. Conjunctive use of surface and ground water resources using the ant system optimization. Agricultural Water Management 173: 23-34.
Sreekanth PD, Geethanjali N, Sreedevi PD, Ahmed S, Kumar NR and Jayanthi PK, 2009. Forecasting groundwater level using artificial neural networks. Current Science 933-939.
Vaheddoost B, Rezaie H and Behmanesh J, 2016. Effect of underground dam construction on discharge of qanats, using Modflow model (case study: Soofi-Makoo). Water and Soil Science-University of Tabriz, 26(4.1): 295-305. (In Persian with English abstract)