Quantitative Analysis On Grain Security In Henan Province Based On Improved Grey Model

Grain the necessaries of human's life is the basic conditions for all activities.It can ensure the nation's lasting political stability only by regarding the grain security as the primary goal,and pay more attention to grain production.As the first high grain yield province,Henan plays a crucial role in national grain security.However,the unstable factors such as too many people and less land,the low level of agricultural technology,and the weak agricultural infrastructure will lead to the grain insecurity,restrict the sustainable and stable development of social economy,and directly influence the living standard of people.Therefore,studying the grain security,finding the current problems of grain security and adjusting the problem timely in Henan province are significant and far reaching impact.This paper analizes the architecture of grain security,studies the interaction relationships between certain and uncertain factors,and establishes the comprehensive model system to make the quantitative relationship based on the thoughts of system engineering and the grey model technology.Around this idea,this paper mainly does the following work:Firstly,this paper establishes the three-hierarchic influencing factors set of grain security,puts forward the quantitized index of grain security-comprehensive coefficient of grain security,and conducts the double-hierarchic diagnosis of influencing factors of grain security in Henan Province based on grey correlation analysis.The results show that: grain supply has the most significant impact on grain security,next came grain demand,and the grain market circulation has the smaller impact on grain security.Secondly,this paper combines the grey GM(1,1)model with the DEA cross model,introduces the theory of ideal point,and construct the grey DEA cross model based on the ideal point to measure the grain product efficient in Henan province.Comparing the results of proposed model and traditional DEA model can obtain the conclusion that the frequency of repeated solution by using proposed model is lower than the traditional DEA model,and the proposed model can effectively solve the problem that cross efficient is not unique.Thirdly,this paper puts forward the two-stage grey clustering model based on mixed center-point trianglewhitenization weight function and determines the weight of indexes by using entropy method.This model can solve the problem that clustering coefficient is too closed to judge the ownership of index,avoids the problem of extension of the index value,and objectively determine the weight of index.Evaluate the situation of grain security in Henan province by using proposed model,the results show that: the overall situation of grain security in Henan province is optimistic from 2000 t0 2015,the grain security in most of years are in the grey class of basic security and complete security,and only several years are in the grey class of light warning.Finally,this paper using the stochastic particle swarm algorithm to optimize the parameters of GM(1,1)model,and compare the prediction result of improved model and traditional GM(1,1)model.From the results we can obtain that the fitting accuracy and forecasting accuracy of improved GM(1,1)are obviously higher than traditional GM(1,1)model.According to the prediction result,this paper make the comprehensive warning of grain security in Henan province from 2016 to 2020.The results show that: the overall situation of grain security in Henan province is optimistic from 2016 t0 2020,but still exists some unsafe factors.The grain security involves many aspects,and the relationship among the aspects is very complicated.This paper only studies the influencing factors of grain security from grain supply,grain demand and grain market circulation,but ignores the other factors such as climate,natural disasters,policies and so on.Apart from that,the reasonableness of the determination of the center point and turning point of grey class and the applicability of the improved GM(1,1)model need to be further verified.