Liriomyza huidobrensis （Blanchard） is an important vegetable pest of pathology. In order to improve the accuracy of prediction of Liriomyza huidobrensis and to control the Liriomyza huidobrensis effectively, this pa...Liriomyza huidobrensis （Blanchard） is an important vegetable pest of pathology. In order to improve the accuracy of prediction of Liriomyza huidobrensis and to control the Liriomyza huidobrensis effectively, this paper presents a new prediction model by principal components analysis （PCA） and back propagation artificial neural network （BP-ANN） methods. The historical data from 1999 to 2007 on population occurrence are analyzed in order to find out a non-linear relationship between the pest occurrence and the meteorological factors. And then by using analysis results, the prediction model of Liriomyza huidobrensis occurrence in Jianshui in Yunnan is built. The new model has successfully applied to verify the paddy stem borer population occurrence in 2006. Test results show that the new prediction model with BP-ANN and PCA can improve the prediction accuracy.展开更多
基金supported by the Mega-Projection of National Key Technology R & D Program for the 11th Five-Year Plan under Grant No.2006BAD10A14
文摘Liriomyza huidobrensis （Blanchard） is an important vegetable pest of pathology. In order to improve the accuracy of prediction of Liriomyza huidobrensis and to control the Liriomyza huidobrensis effectively, this paper presents a new prediction model by principal components analysis （PCA） and back propagation artificial neural network （BP-ANN） methods. The historical data from 1999 to 2007 on population occurrence are analyzed in order to find out a non-linear relationship between the pest occurrence and the meteorological factors. And then by using analysis results, the prediction model of Liriomyza huidobrensis occurrence in Jianshui in Yunnan is built. The new model has successfully applied to verify the paddy stem borer population occurrence in 2006. Test results show that the new prediction model with BP-ANN and PCA can improve the prediction accuracy.
