During typhoon periods, accurate hourly rainfall forecasts are extremely important. A new typhoon rainfall forecasting model that integrates multi-objective genetic algorithm (MOGA) with support vector machines (SVM) is presented in this paper. Apart from the rainfall data, the meteorological variables are also considered. An application to high- and low-altitude meteorological stations has shown that the proposed model yields the best performance as compared to other models. Results indicate that meteorological variables are helpful. The proposed model significantly improves hourly typhoon rainfall forecasting, especially for the long lead time forecasting. Moreover, the optimal combination of inputs is determined by the proposed model for each lead time forecasting. The use of the optimal combination of inputs yields more accurate forecasts than the use of all inputs. In conclusion, the proposed model is expected to be useful for effective hourly typhoon rainfall forecasting.
Jhong, Bing-Chen and Lin, Gwo-Fong, "Integration Of Multi-Objective Genetic Algorithm And Support Vector Machine For Hourly Typhoon Rainfall Forecasting" (2014). CUNY Academic Works.