Design The Prediction Of System Insect Pests Of Crop Based On GRNN Neural Network

With the rapid development of the agricultural informatization construction, putforward higher requirements on the construction of crop pests forecast system, andadvanced computer technology, network technology and information processingtechnology provides a strong technical support for the construction of agriculturalinformatization, which make it gradually toward automation, intelligent direction.The crop diseases and insect pests forecasting system of Hohhot of Inner Mongoliaas an example, design based on using generalized regression neural network algorithm ofradial basis function neural network. The algorithm model have realized the crop pestsforecast function and prove the accuracy and objectivity under the condition of themain factors known to influence crop pest occurrence quantity. The research contentsinclude:1. Design the generalized regression neural network to establish prediction modelusing of Matlab in view of the effect of main factors of crop diseases and insect pests,which the sample data through learning and training, optimized the structure of thenetwork. Through eight sets of test sample data verify the network prediction model ofreliability, accuracy and objectivity.2. Study the method of mixed programming between C#and Matlab, using themethod make the generalized regression neural network algorithm embed into the croppest forecasting system which completed the integration in Hohhot area of crop diseasesand insect pests forecasting system.3. Under the.NET platform to complete the Hohhot region structure and function ofcrop diseases and insect pests forecasting system design, and the intelligent predictionmodule and administrator module for data operation.Simulation and test results show that the crop pests forecasting model based ongeneralized regression neural network algorithm that realize basic function of short-termprediction for Inner Mongolia, Hohhot area of crop diseases and insect pests occurrence.