Toward Evolutionary Nonlinear Prediction Model for Temperature Forecast Using Less Weather Elements
This paper presents the notion of evolutionary nonlinear prediction technique for temperature forecast based on GP (Genetic Programming). The linear regression method is widely used in most of numeric weather prediction model. Their performances are acceptable, but some limitation is existed for nonlinear natures of the weather prediction. We explain how to apply symbolic regression method using GP for the nonlinear prediction model using less weather elements. In order to verify the possibility of the proposed method, experiments of temperature forecast for the sampled locations in South Korea are executed.
Keywordstemperature forecast genetic programming nonlinear regression
Unable to display preview. Download preview PDF.
- 1.Kim, Y., Park, O., Hwang, S.: Realtime Operation of the Korea Local Analysis and Prediction System at METRI. Asia-Pacific Journal of Atmospheric Sciences 38(1), 1–10 (2002)Google Scholar
- 5.United Kingdom Met Office, http://www.metoffice.gov.uk
- 6.Korean Meteorological Society, Introduction to Atmospheric Science. Sigma Press (2009)Google Scholar
- 7.Goldberg, D.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley (1989)Google Scholar
- 8.Zongker, D., Punch, B.: Lil-GP User’s Manual. Michigan State University (1995)Google Scholar