با همکاری انجمن آبخیزداری ایران

نوع مقاله : مقاله پژوهشی

نویسندگان

استادیار، پژوهشکده حفاظت خاک و آبخیزداری، سازمان تحقیقات، آموزش و ترویج کشاورزی، تهران، ایران

چکیده

برآورد صحیح مقدار رسوب معلق رودخانه‌ها، نقش مهمی در مطالعات فرسایش و رسوب، هیدرولوژی و مدیریت حوزه‌های آبخیز دارد. شبیه‌سازی رسوب معلق در سیستم‌های هیدرولوژیکی که دارای پیچیدگی‌های زیاد بوده و درعین‌حال درک و دانش ما از اجزاء و فرآیندهای درون آن‌ها همواره با عدم قطعیت روبرو است سبب کاربرد فراوان مدل‌های هوشمند و از جمله شبکه‌های عصبی مصنوعی شده است. با این حال، استفاده از این مدل‌های هوشمند نیز با چالش روبرو است. تعیین ساختار مناسب شبکه مستلزم بهینه نمودن پارامترهای مورداستفاده در آن (نظیر تعداد بهینه نرون‌ها و لایه‌ها، وزن و بایاس و نوع توابع فعال‌سازی) بوده که واسنجی مناسب آن‌ها به روش آزمون و خطا، ضمن کارایی کم، منجر به صرف زمان زیاد می‌شود. در پژوهش حاضر، به منظور شبیه‌سازی بار رسوب معلق روزانه رودخانه‌ نیرچای )در محل ایستگاه‌ آب سنجی نیر در استان اردبیل) از شبکه عصبی مصنوعی پرسپترون چند لایه استفاده شد. به منظور آموزش مدل شبکه عصبی، علاوه بر روش مرسوم پس انتشار خطا، از الگوریتم بهینه‌سازی ازدحام ذرات (Particle Swarm Optimization (PSO))، به منظور بهینه‌سازی مقادیر وزن و بایاس نرون‌های مدل‌های شبکه عصبی استفاده گردید. به منظور افزایش قدرت تعمیم دهی مدل‌ها، از خوشه‌‌‌‌بندی فازی استفاده شد. نتایج گرفته‌شده از پژوهش حاضر نشان داد که آموزش مدل‌های شبکه عصبی با الگوریتم PSO با کاهش خطای برآورد رسوب (کاهش خطای برآورد کل و ریشه میانگین مربعات خطا به ترتیب تا 3/0 درصد و 4/10 تن در روز) کارایی بیشتری نسبت به مدل‌های شبکه عصبی که صرفاً از روش‌های ‌پس انتشار خطا استفاده می‌نمایند داشته است. با توجه به اینکه در بهینه‌سازی پارامترهای شبکه عصبی، الگوریتمهای تکاملی (نظیر الگوریتم PSO) قادر به ارائه راهحلهای مناسبی هستند ، لذا در شبیه‌سازی پدیدهها و متغیرهای پیچیده حوزههای آبخیز (نظیر رسوب معلق) می‌توان از این توانمندی استفاده نمود.

کلیدواژه‌ها

عنوان مقاله [English]

Improvement of the efficiency of artificial neural network model in suspended sediment simulation using particle swarm optimization algorithm

نویسندگان [English]

  • Mahmoudreza Tabatabaei
  • Amin Salehpour Jam
  • Jamal Mosaffaie

Assistant Professor, Soil Conservation and Watershed Management Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Tehran, Iran

چکیده [English]

The proper estimation of the amount of suspended sediment in rivers has an important role in erosion and sediment studies, hydrology and management of watersheds. The simulation of suspended sediment in hydrological systems that has a lot of complexity and at the same time our understanding of the components and processes within them is always uncertain led to the use of many intelligent models, including artificial neural networks (ANNs). However, the use of these smart models also faces challenges. Determining the proper structure of the network requires optimization of the parameters used (such as the optimal number of neurons and layers, weight and bias, and the type of activation functions), which their proper calibration, using test and error, leads to a lot of time spent in low efficiency. In this study, a multilayer perceptron (MLP) was used to simulate the daily sediment load of the Nirchai River at the site of the Nair hydrometric station in Ardebil province. In order to train the models, in addition to the error back propagation (BP) algorithm, Particle Swarm Optimization (PSO) algorithm was used to optimize the weight and bias of ANNs. The fuzzy clustering method was also used to increase the power of generalization of the models. The results showed that training of ANN models with PSO algorithm with decreasing estimation error (decreasing the PBIAS of estimation and root mean square error up to 0.3% and 10.4 tons per day respectively) is more effective than ANN models that use only error BP techniques. Due to insufficient recorded sediment data in most hydrometric stations of the country on the one hand and the need to train ANNs with sufficient data on the other hand, the use of evolutionary algorithms (e.g. PSO algorithm) can be a good solution for improving the efficiency of intelligent models.

کلیدواژه‌ها [English]

  • Evolutionary Algorithm
  • Balekhlochi River
  • Fuzzy Clustering
  • Simulation
  • Intelligent Model
  1.  

    1. Alizamir, M. and S. Sobhanardakani. 2018. An Artificial Neural Network-Particle Swarm Optimization (ANN-PSO) approach to predict heavy metals contamination in groundwater resources. Jundishapur Journal of Health Sciences, 10(2): doi: 10.5812/jjhs.67544.
    2. Altunkaynak, A. 2009. Sediment load prediction by genetic algorithms. Advances in Engineering Software, 40: 928-934.
    3. Cheng, C.T., C.P. Ou and K.W. Chau. 2002. Combining a fuzzy optimal model with a genetic algorithm to solve multi-objective rainfall–runoff model calibration. Journal of Hydrology, 268: 72-86.
    4. Criss, R.E. and W.E. Winston. 2008. Do Nash values have value? discussion and alternate proposals. Hydrological Processes, 22(14): 2723-2738.
    5. Ebtehaj, I. and H. Bonakdari. 2016. Assessment of evolutionary algorithms in predicting non-deposition sediment transport. Urban Water Journal, 13: 499-510.
    6. Guo, W. and H. Wang. 2010. August PSO optimizing neural network for the Yangtze River sediment entering estuary prediction. In 2010 6th International Conference on Natural Computation, 4: 1769-1772.
    7. Gupta, H.V., S. Sorooshian and P.O. Yapo. 1999. Status of automatic calibration for hydrologic models: comparison with multilevel expert calibration. Journal of Hydrologic Engineering, 4(2): 135–143.
    8. Gupta, H.V., H. Kling, K.K. Yilmaz and G.F. Martineza. 2009. Decomposition of the mean squared error and NSE performance criteria: implications for improving hydrological modeling. Journal of Hydrology (Amsterdam), 377(1–2): 80–91.
    9. Hejazi, M.I., X. Cai and D.K. Borah. 2008. Calibrating a watershed simulation model involving human interference: an application of multi-objective genetic algorithms. Journal of Hydroinformatics, 10: 97-111.
    10. Hornik, K., M. Stinchcombe and H. White. 1989. Multilayer feedforward networks are universal approximators. Neural Networks, 2(5): 359-366.
    11. Jansson, M.B. 1996. Estimating a sediment rating curve of the Reventazon River at Palomo using logged mean loads within discharge classes. Journal of Hydrology, 183(3-4): 227-241.
    12. Kashefei Kavyani, A., S.A. Pourmosavi Kani and A. Jahanbani Adrakani. 2007. The training of multi-layer neural networks using PSO algorithm. First Joint Congress of Fuzzy Systems and Intelligent System, Mashhad, Ferdowsi University of Mashhad. Available online at:https://www.civilica.com/ Paper-FJCFIS01-FJCFIS01_229.html.
    13. Kaufman, L. and P.J. Rousseeuw. 2009. Finding groups in data: an introduction to cluster analysis (Vol. 344). John Wiley and Sons, New Jersey, USA, 342 pages.
    14. Kisi, O. and J. Shiri. 2012. River suspended sediment estimation by climatic variables implication: Comparative study among soft computing techniques. Computers and Geosciences, 43: 73-82.
    15. Kisi, O., A. Keshavarzi, J. Shiri, M. Zounemat-Kermani and E.S.E. Omran. 2017. Groundwater quality modeling using neuro-particle swarm optimization and neuro-differential evolution techniques. Hydrology Research, 48(2017): 1508-1519.
    16. Kisi, O. and C. Ozkan. 2017. A new approach for modeling sediment-discharge relationship: local weighted linear regression. Water Resources Management, 30(2): 1-23.
    17. Krause, P., D.P. Boyle and F. Bäse. 2005. Comparison of different efficiency criteria for hydrological model assessment. Advances in Geosciences, 5: 89-97.
    18. Kuok, K.K., S. Harun and S.M. Shamsuddin. 2010. Particle swarm optimization feedforward neural network for modeling runoff. International Journal of Environmental Science and Technology, 7: 67-78.
    19. Lafdani, E.K., A.M. Nia and A. Ahmadi. 2013. Daily suspended sediment load prediction using artificial neural networks and support vector machines. Journal of Hydrology, 478: 50-62.
    20. Legates, D.R. and G.J. McCabe. 1999. Evaluating the use of “goodness‐of‐fit” measures in hydrologic and hydroclimatic model validation. Water Resources Research, 35(1): 233-241.
    21. May, R.J., H.R. Maier and G.C. Dandy. 2010. Data splitting for artificial neural networks using SOM-based stratified sampling. Neural Networks, 23: 283-294.
    22. Mansourfar, K. 2009. Advanced statistical methods using applied software. University of Tehran Press, Tehran, Iran, 480 pages (in Persian).
    23. Moriasi, D.N. and J.G. Arnold, M.W. Liew Van, R.L. Bingner, R.D. Harmel and T.L. Veith. 2007. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3): 885-900.
    24. Muleta, M.K. 2011. Model performance sensitivity to objective function during automated calibrations. Journal of Hydrologic Engineering, 17(6): 756-767.
    25. Nash, J.E. and J.V. Sutcliffe. 1970. River flow forecasting through conceptual models, part I-A discussion of principles. Journal of Hydrology, 10(3): 282-290.
    26. Nasserabadi, F., A. Esmaali Ouri and S. Tolabi. 2013. Estimation of soil erosion potential and sediment yield using EPM model and GIS technique, case study: Balkhlochai River Basin. The First National Conference on Planning, Conservation and Sustainable Development, Hamedan, Hegmatane Environment Evaluators Association, available online at: https://www.civilica.com/ Paper-PCEPSD01-PCEPSD01_157.html.
    27. Partal, T. and H.K. Cigizoglu. 2008. Estimation and forecasting of daily suspended sediment data using wavelet–neural networks. Journal of Hydrology, 358(3-4): 317-331.
    28. Shahriar Shahhoseini, H., M. Moosavi and M. Mollajafari. 2011. Evolutionary algorithms- fundamentals, applications, implementation. Tehran Press Center, Iran University of Science and Technology, Tehran, 590 pages (in Persian).
    29. Swain, R. and B. Sahoo. 2017. Mapping of heavy metal pollution in river water at daily time-scale using spatio-temporal fusion of MODIS-aqua and Landsat satellite imageries. Journal of Environmental Management, 192: 1-14.
    30. Tabatabaei, M. and A. Salehpour Jam. 2017. Optimization of sediment rating curve coefficients using evolutionary algorithms and unsupervised artificial neural network. Caspian Journal of Environmental Sciences, 15(4): 387-401.
    31. Tabatabaei, M., A. Salehpour Jam and S.A. Hosseini. 2019. Suspended sediment load prediction using non-dominated sorting genetic algorithm II. International Soil and Water Conservation Research, 7(2): 119-129.
    32. Tayfur, G. 2012. Soft computing in water resources engineering: artificial neural networks, fuzzy logic and genetic algorithms. WIT Press, Dorset, UK, 288 pages.
    33. Tfwala, S.S. and Y.M. Wang. 2016. Estimating sediment discharge using sediment rating curves and artificial neural networks in the Shiwen River, Taiwan. Water, 8(53): 1-15.
    34. Ulke, A., G. Tayfur and S. Ozkul. 2009. Predicting suspended sediment loads and missing data for Gediz River, Turkey. Journal of Hydrologic Engineering, 14: 954-965.
    35. Yar Kiani, A. 2009. Intelligent systems. Press Center of Poyesh Andisheh, Tehran, Iran, 260 pages (in Persian).
    36. Zhou, H. and G. Schaefer. 2009. An overview of fuzzy C-Means based image clustering algorithms. Foundations of Computational Intelligence, 2: 295-310.
    37. Zounemat-Kermani, M. 2017. Assessment of several nonlinear methods in forecasting suspended sediment concentration in streams. Hydrology Research, 48(5): 1240-1252.