Improving Hybrid Models for Precipitation Forecasting by Combining Nonlinear Machine Learning Methods

Abstract Precipitation forecast is key for water resources management in semi-arid climates. The traditional hybrid models simulate linear and nonlinear components of precipitation series separately. But they do not still provide accurate forecasts. This research aims to improve by using an ensemble models. Preprocessing configurations each the Gene Expression Programming (GEP), Support Vector Regression (SVR), Group Method Data Handling (GMDH) were used as They compared against proposed with a combination all these three performance was improved different methods. Two weather stations Tabriz Rasht Iran respectively annual monthly time steps selected test results showed that Theil’s coefficient, which measures inequality degree forecasts differ from observations, 9% 15% SVR GMDH relative GEP station. applied error criteria indicated have better representation observations than Mean square decreased 67% Nash Sutcliffe increased 5% station when we combined machine learning genetic algorithm instead SVR. Generally, within implications modeling highly systems full advantages