Study On Prediction Accuracy Of Grey Model And Its Application In The Length Of The Branches Of Thymus

Grey prediction refers to explore and master the law of the future development of the system by creating the grey model when dealing with the original data. Its core idea is:through a small amount or incomplete information to create a differential prediction model, further complicated information, expressed no clear system for long-term development of the law described. The model study emphasis "poor information, less data uncertainty" issue.Thymus is a small and dense plant flowers, fragrant smell when distributing open, plant is short, easy to form a thick and rich elastic layer.Therefore, this paper through the continuous space ant colony algorithm to optimize parameters of gray model to achieve improved grey model to predict the accuracy of target, and the model is applied in the prediction of thymus shoot length, with a view to Inner Mongolia and northern arid urban landscape construction to provide new materials and new ideas.Paper carefully studied on the basis of gray system theory for gray GM (1,1) model,and carrise out the following three aspects of the research to solve the existing problems.1.based on the geometric meaning of the background value of the gray model, the new background value formula is constructed to improve the prediction accuracy. After analyzing the deficiency of the traditional grey model, this paper with the background value as the breakthrough point constructs a new background value formula IGM (1,1) model,and improves the prediction accuracy.2.Using the continuous space ant colony algorithm to solver the parameter method for GM(1,1) model.This method is the use of continuous space ant colony algorithm global search and local search operation to limit optimal solution, based on theACOGM (1,1) model is proposed based on the ant colony algorithm.3.The three models were applied to Hu Shuo highway thymus shoot length prediction problem. Through the comparison of the results,further verify the effectiveness and applicability of the new model.The experimental results show that the improved model can improve the prediction accuracy.During the completed the paper,I always exchange the idea with school outstanding entrepreneurship student who is agricultural grass responsible.the length of the branches of the data were derived from field measurements. The model is applied in the Hu Shuo highway thymus shoot length predictive,in order to master the resistance of thymus,hope that through a cooperation with alumnus in the late, the agriculture and the combination of information, better promote agricultural informatization venture project forward.