Research on sand-dust storm warning based on SVM with combined kernel function
To improve the correct rate of sand-dust storm forecasts,a support vector machine classifier with combined kernel function which integrates the polynomial kernel function with the Gussian radial kernel function together is presented,and then it is applied to the application of sand-dust storm warning.Taken Yanchi district in Ningxia as an example,a large number of projections are made based on its historical data.The experimental results show the Support Vector Machine Model with combined kernel function can forecast whether sand-dust storm occurred in some region accurately and the successful limit index exceeds that of the traditional support vector machine model with single kernel function by nearly 2.79 ％.
sand-dust storm warningcombined kernel functionsupport vector machineclassificationforecasting model
NETL
NSTL
万方数据
维普
