Study On Application Of Partial Least Squares Method Wheat Scab Forecasting

Wheat scab is one of the main diseases and insect pests,whichhave serious effects on t he production of wheat in our province.Wheat diseases forecasting usually adopts linear regression,but the factor which influence the scab occurred between multiple correlation and using the method of general linear prediction model having poor stability.Accurate forecast of plant diseases and insect pests is the premise of preparing in advance,focusing on prevention and controlling and reducing the loss.Therefore chooses the appropriate algorithm to wheat scab disease prediction research has important practical significance.In this paper,11 years from 2004 to 2014 of Tongcheng,Anhui province wheat scab disease data as the foundation.Analysis of the cause and the research status is analyzed.Wheat induced heavy meteorological factor has the characteristics of multiple correlation,so chose partial least squares model which can be extracted by principal component regression so as to effectively solving the multiple correlation between factors.The prediction model of wheat scab was built,and the prediction system of wheat scab based on partial least square method was designed and realized.The main contents of the thesis are as follows:1.select the origins of meteorological factors as independent variables,using cross validation method to determine the number of principal components and establishing Anhui wheat 3 single dependent variable partial least-square regression forecast models.Then Using Matlab software programming algorithm and Through the comparison of the predicted values and the actual value comparison test,finally the model prediction accuracy reached satisfaction.2.Using original concept 11.5 development platform and combining with severs SQL 2005 database system,we designed and implemented scab big data analysis system version 1.0 of Anhui.The application of prediction model to ‘scab prediction’s in the system module,can not only make the pest data visually more intuitive display,also improve the operability of the heavy forecast work,which for scab predicting provides an intelligent,network,information platform.In this paper,the partial least square method was applied to scab prediction research,Through the establishment of forecasting model,system realization,finally get the results.The results provide reliable basis for scab forecasting work,at the same time for scab large data analysis system in Anhui follow-up study to lay the foundation,also play a reference for other diseases prediction.