https://en.front-sci.com/index.php/JAI/article/view/37

Published on: 2019-06-18 | Updated on: 2019-09-09

Journal of Autonomous Intelligence

The precipitation will greatly affect people’s life and production. If there is too much rainfall, it will lead to flash flooding and natural disasters, even though will cause severe economic losses and inconveniences to human life. In recently years, there are many methods used in hydrologic forecasting. Besides lots of mechanism models, data driving models are popular in recently years. They can be divided into two classification: probability statistics method and time series analysis method.

Recently, using combined methods to predict time series is developing fast in different fields. The idea to hybrid data preprocessing method, forecasting model and optimization method to predict rainfall is attractive. Weide Li and Juan Zhang made an research of the innovated model  about intelligent algorithm for rainfall forecasting,The survey result published on the journal of autonomous intelligence.

In this paper, a novel hybrid model to forecast rainfall is developed by incorporating singular spectrum analysis (SSA) and dragonfly algorithm (DA) into support vector regression (SVR) method. Firstly, SSA is used for extracting the trend components of the hydrological data. Then, SVR is utilized to deal with the volatility and irregularity of the precipitation series. Finally, the parameter of SVR is optimized by DA. The proposed SSA-DA-SVR method is used to forecast the monthly precipitation for Songbai, Panshui, Lanma and Jiulongchi stations. To validate the efficiency of the method, four compared models, DA-SVR, SSA-GWO-SVR, SSA-PSO-SVR, SSA-CS-SVR are established.

Compared with DA-SVR, SSA-GWO-SVR, SSA-PSO-SVR, SSA-CS-SVR models, the proposed hybrid model can effectively improve the prediction accuracy for month average precipitation. Thus, the model can be used on rainfall forecasting in the future. In addition, as a prediction model, it can also be applied in wind speed and power load forecasting.

For more information, please visit:

https://en.front-sci.com/index.php/JAI/article/view/37