prediction of monthly rainfall using artificial neural network mixture approach, case study: torbat-e heydariyeh

Rainfall is one of the most important elements of water cycle used in evaluating climate conditions of each region. longterm forecast of rainfall for arid and semiarid regions is very important for managing and planning of water resources. to forecast appropriately, accurate data regarding humidity, temperature, pressure, wind speed etc. is required.this article is analytical and its database includes 7336 records situated in 11 features from daily brainstorm data within a twenty year period. the samples were selected based on a case study in torbate heydariyeh. 70% were chosen for learning and 30% were chosen for taking tests. from 7181 available data, 75% and 25% were used for training and evaluating, respectively. this research studied the performance of different neural networks in order to predict precipitation and then presented an algorithm for combining neural networks with linear and nonlinear methods. after modeling and comparing their results using neural networks, the root mean square error was recorded for each method. in the first modeling, the artificial neural network error was 0.05, in the second modeling, linear combination of neural networks error was 0.07, and in the third model, nonlinear combination neural networks error was 0.001. reducing the error of forecasting precipitation has always been one of the goals of the researchers. this study, with the forecast of precipitation by neural networks, suggested that the use of a more robust method called a nonlinear combination neural network can lead to improve men is in for cast diagnostic accuracy.

Prediction of monthly rainfall using artificial neural network mixture approach, Case Study: Torbate Heydariyeh

Rainfall is one of the most important elements of water cycle used in evaluating climate conditions of each region. Longterm forecast of rainfall for arid and semiarid regions is very important for managing and planning of water resources. To forecast appropriately, accurate data regarding humidity, temperature, pressure, wind speed etc. is required.This article is analytical and its database includes 7336 records situated in 11 features from daily brainstorm data within a twenty year period. The samples were selected based on a case study in Torbate Heydariyeh. 70% were chosen for learning and 30% were chosen for taking tests. From 7181 available data, 75% and 25% were used for training and evaluating, respectively. This research studied the performance of different neural networks in order to predict precipitation and then presented an algorithm for combining neural networks with linear and nonlinear methods. After modeling and comparing their results using neural networks, the root mean square error was recorded for each method. In the first modeling, the artificial neural network error was 0.05, in the second modeling, linear combination of neural networks error was 0.07, and in the third model, nonlinear combination neural networks error was 0.001. Reducing the error of forecasting precipitation has always been one of the goals of the researchers. This study, with the forecast of precipitation by neural networks, suggested that the use of a more robust method called a nonlinear combination neural network can lead to improve men is in for cast diagnostic accuracy.