Skip to Main content Skip to Navigation
Conference papers

Research on GA-RBF Optimization Algorithm in the Prediction of Yield Loss of Maize Diseases and Pests

Abstract : In view of the high complexity and nonlinearity of crop pests and diseases, using the traditional BP network and RBF network model method to predict is pretty difficult. And the prediction accuracy is low. Also the effect is not ideal when the sample size is small and the noise is more, therefore, this article presents a fusion optimization algorithm based on genetic-algorithm (GA) and radial basis function neural network (RBF). By unified coding the data center of RBF neural network and its corresponding expansion constant and weight, strengthened the cooperation between the hidden and output layer, furthermore using the functional characteristics of global search using genetic algorithm to obtain the optimal model of yield loss, finally predict on yield loss of maize diseases and pests. By making simulation test data of the National 863 project demonstration area - 13 village, Gong’ peng town in Jilin province Yu’shu County, the experimental results show that: After using the GA to optimize the RBF in the network’s structure and approximation has obvious improvement and enhancement, can effectively reflect the fluctuation characteristics of maize diseases and pests, has been widely application prospect in the agricultural field.
Document type :
Conference papers
Complete list of metadata
Contributor : Hal Ifip <>
Submitted on : Thursday, July 11, 2019 - 3:09:31 PM
Last modification on : Thursday, July 11, 2019 - 3:12:07 PM


Files produced by the author(s)


Distributed under a Creative Commons Attribution 4.0 International License



Guifen Chen, Dongxue Wang, Shan Zhao, Siwei Fu. Research on GA-RBF Optimization Algorithm in the Prediction of Yield Loss of Maize Diseases and Pests. 10th International Conference on Computer and Computing Technologies in Agriculture (CCTA), Oct 2016, Dongying, China. pp.257-267, ⟨10.1007/978-3-030-06155-5_25⟩. ⟨hal-02179963⟩



Record views


Files downloads