Comparison of Hybrid Model (ANFIS-PSO) and Tork Experimental Model in Reference Estimation of Evapotranspiration (Case study: Poldokhtar-Lorestan)

Document Type : Research Article

Authors

1 Faculty of Agriculture and Natural Resources, University of Lorestan

2 Faculty of New Sciences and Technologies Renewable Energies and Environment.

3 Watershed Management Engineering, Faculty of Agriculture and Natural Resources, University of Lorestan

Abstract

Evapotranspiration the catchment area is one of the main components of the water cycle and estimates the need for irrigation. The purpose of this study is to estimate the daily evapotranspiration and forwarding (ETo) using a neuro-fuzzy comparative inference system (ANFIS) and compare it with the experimental model of Tork. The input data of the turquoise model, including sunshine, air temperature, relative humidity and wind speed, were obtained from the Meteorological Station of Poldokhtar from the Lorestan Meteorological Station, and evapotranspiration was obtained using this method. Input data (ANFIS) including average temperature, mean humidity and latent evaporation rate were calculated using the corresponding and were given to the data model. The results of each method were compared separately with evaporation calculated at the station location using a vapor pan. In this study, the performance of the ANFIS based on the optimization algorithm (PSO) was satisfactory and the results were satisfactory. The efficiency of the compared methods was obtained using root mean square error (RMSE), mean square error MSE And the R2 determination coefficient was evaluated. The ANFIS-PSO method, with only three parameters, is the average of average humidity and the latent evapotranspiration rate, which can predict daily reference evapotranspiration and is more accurate than the empirical model of tork. So that the value of R2, RMSE, MSE for ANFIS model is 0.90, 2.65, 13.7 and for the experimental model of Tork is 0.63, 6.26, 39.24, respectively.

Keywords

Main Subjects

References

Karamuz M, Araghi Nezhad S. Advanced Hydrology. Tehran: Amirkabir university of technology; 2014. 468 p.
2.        Hossein Mirzaei HM. Comparison of Different Estimates of Evapotranspiration Potential. In: First National Conference on Irrigation and Drainage Networks Management. 2006.
3.        Sarmad MM-B-ME-M. Estimation of Reference crop Evapotranspiration Using the Least Meteorological Data. J Water Soil. 2009;23(1):91–9.
4.        Mohammadi KAGG-SMM-K. Comparison of Artificial Intelligence Systems (ANN & ANFIS) for Reference Evapotranspiration Estimation in the Extreme Arid Regions of Iran. J Water Soil. 2010;24(4):679–89.
5.        Shiri J, Ki\csi Ö, Landeras G, López JJ, Nazemi AH, Stuyt LCPM. Daily reference evapotranspiration modeling by using genetic programming approach in the Basque Country (Northern Spain). J Hydrol. 2012;414:302–16.
6.        Terzi Ö. Daily pan evaporation estimation using gene expression programming and adaptive neural-based fuzzy inference system. Neural Comput Appl. 2013;23(3–4):1035–44.
7.        Kumar P, Kumar D, Jaipaul J, Tiwari AK. Evaporation estimation using artificial neural networks and adaptive Neuro-Fuzzy inference system techniques. Pakistan J Meteorol. 2012;8(16).
8.        Mohamadrezapour O. Monthly Forecast of Potential Evapotranspiration Models Using Support Vector Machine (SVM), Genetic programming and Neural- Fuzzy Inference System. Irrig Water Eng. 2017;7(3):135–50.
9.        Haghighatjou P, Shahroudi ZMZA. Comparison of gene expression programming (GEP) and neuro-fuzzy methods for estimation of pan evaporation (case study: south Khorasan province). Water Soil Resour Conserv. 2017;6(4):107–17.
10.      Citakoglu H, Cobaner M, Haktanir T, Kisi O. Estimation of monthly mean reference evapotranspiration in Turkey. Water Resour Manag. 2014;28(1):99–113.
11.      Shayannejad masumeh najafi; VAM. An Evaluation of Accuracy of Intelligent Methods and Sensitivity Analysis of Reference Crop Evapotranspiration to Meteorological Parameters in Two Different Climates. Ecohydrology. 2014;1(1):17–24.
12.      M Nekooamal Kermani RMN. Determination of the Best Adaptive Neuro-Fuzzy Inference System (ANFIS) Model for Estimating Grass Reference Crop Evapotranspiration in Coastal Semi-arid Climate of Hormozgan. Water Soil Sci. 2016;26(2):239–58.
13.      Azad MRAP, Sattari and MT. Forecasting daily river flow of Ahar Chay River using Artificial Neural Networks (ANN) and comparison with Adaptive Neuro Fuzzy Inference System (ANFIS). J Water Soil Conserv. 2015;22(1):287–98.
14.      Kobold M, Suselj K, Polajnar J, Pogacnik N. Calibration techniques used for HBV hydrological model in Savinja catchment. In: XXIVth Conference of the Danubian Countries on the Hydrological Forecasting and Hydrological Bases of Water Managemet. 2008. p. 2–4.
15.      Kamali Bahareh SJM. HEC-HMS Conceptual Automated Calibration - Simulation and Optimization Approach. In: 5th National Congress on Civil Engineering [Internet]. 2010. Available from: https://www.tpbin.com/article/27170
16.      Jang J.S.R., Sun C.T. and ME. Neuro-fuzzy and Software Computing: a Computational Approach to Learning and Machine Intelligence. In: Prentice-Hall, New Jersey. 1997.
17.      Drake JT. Communications phase synchronization using the adaptive network fuzzy inference system (anfis). New Mexico State University; 2000.
18.      Sattari MT, Farnaz Nahrein VA. M5 Model Trees and Neural Networks Based Prediction of Daily ET0(Case Study: Bonab Station). Iran J lrrigation Drain. 2013;7(1):104–13.
19.      Hozhabr H, Moazed H, ShokriKhoochak S. Estimation of Reference Evapotranspiration (ETo) Using Empirical Models, Artificial Neural Network Modeling and Their Comparison with Lysimeter Data in Urmia Kahrizi Station. Irrig Water Eng. 2014;4(3):13–25.
Iranian journal of Ecohydrology
Volume 6, Issue 3
September 2019
Pages 685-694

Files

History

  • Receive Date: 22 December 2018
  • Revise Date: 21 June 2019
  • Accept Date: 21 June 2019

Share

How to cite

Statistics