Loading...

Meteorological Drought Forecasting Using Conjunctive Model Of Adaptive Neuro Fuzzy Inference System And Wavelet Transforms (Case Study: Urmia Lake Watershed

Soleimani, Arash | 2014

845 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45430 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Abrishamchi, Ahmad
  7. Abstract:
  8. Drought is a common phenomena which has a lot of unwanted conse-quences on human being life and environment. Drought forecasting plays a significant role in water resources and environmental systems. Considering IRAN inappropriate location which is on the arid and semi-arid area of the earth and Widespread damages which are related to drought during recent years in iran; importance of developing an accurate model by using new technologies becomes quite inevitable. In the last decay Neural Networks have appeared very useful in non-Stationary and non-linear time Series forecasting and modeling.
    This study is about to use conjunctive model of adaptive neuro fuzzy inference system and wavelet transforms in drought forecasting. Hence at the first non-stationary hydrological time series such as precipitation were decomposed into their certain and uncertain components using wavelet transforms. Then the achieved subseries were used as the ANFIS model input in order to train the network and finally forecast the drought. Then these components were integrated to rebuild main series using wavelet transforms.
    Because of critical environmental situation of the URMIA lake, Urmia’s watershed was chosen as the case study.
    The final results represent more R^2 and less RMSE values in 1,3,6 month for introduced conjunctive model shows more accuracy in comparison with Artificial neural network model, adaptive neuro fuzzy inference system model and conjunctive model of artificial neural network and wavelet transforms
  9. Keywords:
  10. Forecasting ; Drought ; Lake Urmia Watershed ; Wavelet Transform ; Adaptive Neuro-Fuzzy Inference System (ANFIS)

 Digital Object List

 Bookmark

No TOC